Compare commits

..

2 Commits

Author SHA1 Message Date
18376aa64b Backup commit 2025-03-14 09:59:05 +01:00
a5fdccc764 Introduce nats client
Initially submit every job received via rest on NATS
2025-02-17 17:26:51 +01:00
292 changed files with 25790 additions and 32454 deletions

View File

@@ -1,15 +0,0 @@
# To get started with Dependabot version updates, you'll need to specify which
# package ecosystems to update and where the package manifests are located.
# Please see the documentation for all configuration options:
# https://docs.github.com/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file
version: 2
updates:
- package-ecosystem: "gomod"
directory: "/"
schedule:
interval: "weekly"
- package-ecosystem: "npm"
directory: "/web/frontend"
schedule:
interval: "weekly"

331
.github/workflows/Release.yml vendored Normal file
View File

@@ -0,0 +1,331 @@
# See: https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions
# Workflow name
name: Release
# Run on tag push
on:
push:
tags:
- '**'
jobs:
#
# Build on AlmaLinux 8.5 using golang-1.18.2
#
AlmaLinux-RPM-build:
runs-on: ubuntu-latest
# See: https://hub.docker.com/_/almalinux
container: almalinux:8.5
# The job outputs link to the outputs of the 'rpmrename' step
# Only job outputs can be used in child jobs
outputs:
rpm : ${{steps.rpmrename.outputs.RPM}}
srpm : ${{steps.rpmrename.outputs.SRPM}}
steps:
# Use dnf to install development packages
- name: Install development packages
run: |
dnf --assumeyes group install "Development Tools" "RPM Development Tools"
dnf --assumeyes install wget openssl-devel diffutils delve which npm
dnf --assumeyes install 'dnf-command(builddep)'
# Checkout git repository and submodules
# fetch-depth must be 0 to use git describe
# See: https://github.com/marketplace/actions/checkout
- name: Checkout
uses: actions/checkout@v2
with:
submodules: recursive
fetch-depth: 0
# Use dnf to install build dependencies
- name: Install build dependencies
run: |
wget -q http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-bin-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-src-1.18.2-1.module_el8.7.0+1173+5d37c0fd.noarch.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/go-toolset-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm
rpm -i go*.rpm
npm install --global yarn rollup svelte rollup-plugin-svelte
#dnf --assumeyes builddep build/package/cc-backend.spec
- name: RPM build ClusterCockpit
id: rpmbuild
run: make RPM
# AlmaLinux 8.5 is a derivate of RedHat Enterprise Linux 8 (UBI8),
# so the created RPM both contain the substring 'el8' in the RPM file names
# This step replaces the substring 'el8' to 'alma85'. It uses the move operation
# because it is unclear whether the default AlmaLinux 8.5 container contains the
# 'rename' command. This way we also get the new names for output.
- name: Rename RPMs (s/el8/alma85/)
id: rpmrename
run: |
OLD_RPM="${{steps.rpmbuild.outputs.RPM}}"
OLD_SRPM="${{steps.rpmbuild.outputs.SRPM}}"
NEW_RPM="${OLD_RPM/el8/alma85}"
NEW_SRPM=${OLD_SRPM/el8/alma85}
mv "${OLD_RPM}" "${NEW_RPM}"
mv "${OLD_SRPM}" "${NEW_SRPM}"
echo "::set-output name=SRPM::${NEW_SRPM}"
echo "::set-output name=RPM::${NEW_RPM}"
# See: https://github.com/actions/upload-artifact
- name: Save RPM as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend RPM for AlmaLinux 8.5
path: ${{ steps.rpmrename.outputs.RPM }}
- name: Save SRPM as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend SRPM for AlmaLinux 8.5
path: ${{ steps.rpmrename.outputs.SRPM }}
#
# Build on UBI 8 using golang-1.18.2
#
UBI-8-RPM-build:
runs-on: ubuntu-latest
# See: https://catalog.redhat.com/software/containers/ubi8/ubi/5c359854d70cc534b3a3784e?container-tabs=gti
container: registry.access.redhat.com/ubi8/ubi:8.5-226.1645809065
# The job outputs link to the outputs of the 'rpmbuild' step
outputs:
rpm : ${{steps.rpmbuild.outputs.RPM}}
srpm : ${{steps.rpmbuild.outputs.SRPM}}
steps:
# Use dnf to install development packages
- name: Install development packages
run: dnf --assumeyes --disableplugin=subscription-manager install rpm-build go-srpm-macros rpm-build-libs rpm-libs gcc make python38 git wget openssl-devel diffutils delve which
# Checkout git repository and submodules
# fetch-depth must be 0 to use git describe
# See: https://github.com/marketplace/actions/checkout
- name: Checkout
uses: actions/checkout@v2
with:
submodules: recursive
fetch-depth: 0
# Use dnf to install build dependencies
- name: Install build dependencies
run: |
wget -q http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-bin-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/golang-src-1.18.2-1.module_el8.7.0+1173+5d37c0fd.noarch.rpm \
http://mirror.centos.org/centos/8-stream/AppStream/x86_64/os/Packages/go-toolset-1.18.2-1.module_el8.7.0+1173+5d37c0fd.x86_64.rpm
rpm -i go*.rpm
dnf --assumeyes --disableplugin=subscription-manager install npm
npm install --global yarn rollup svelte rollup-plugin-svelte
#dnf --assumeyes builddep build/package/cc-backend.spec
- name: RPM build ClusterCockpit
id: rpmbuild
run: make RPM
# See: https://github.com/actions/upload-artifact
- name: Save RPM as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend RPM for UBI 8
path: ${{ steps.rpmbuild.outputs.RPM }}
- name: Save SRPM as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend SRPM for UBI 8
path: ${{ steps.rpmbuild.outputs.SRPM }}
#
# Build on Ubuntu 20.04 using official go 1.19.1 package
#
Ubuntu-focal-build:
runs-on: ubuntu-latest
container: ubuntu:20.04
# The job outputs link to the outputs of the 'debrename' step
# Only job outputs can be used in child jobs
outputs:
deb : ${{steps.debrename.outputs.DEB}}
steps:
# Use apt to install development packages
- name: Install development packages
run: |
apt update && apt --assume-yes upgrade
apt --assume-yes install build-essential sed git wget bash
apt --assume-yes install npm
npm install --global yarn rollup svelte rollup-plugin-svelte
# Checkout git repository and submodules
# fetch-depth must be 0 to use git describe
# See: https://github.com/marketplace/actions/checkout
- name: Checkout
uses: actions/checkout@v2
with:
submodules: recursive
fetch-depth: 0
# Use official golang package
- name: Install Golang
run: |
wget -q https://go.dev/dl/go1.19.1.linux-amd64.tar.gz
tar -C /usr/local -xzf go1.19.1.linux-amd64.tar.gz
export PATH=/usr/local/go/bin:/usr/local/go/pkg/tool/linux_amd64:$PATH
go version
- name: DEB build ClusterCockpit
id: dpkg-build
run: |
ls -la
pwd
env
export PATH=/usr/local/go/bin:/usr/local/go/pkg/tool/linux_amd64:$PATH
git config --global --add safe.directory $(pwd)
make DEB
- name: Rename DEB (add '_ubuntu20.04')
id: debrename
run: |
OLD_DEB_NAME=$(echo "${{steps.dpkg-build.outputs.DEB}}" | rev | cut -d '.' -f 2- | rev)
NEW_DEB_FILE="${OLD_DEB_NAME}_ubuntu20.04.deb"
mv "${{steps.dpkg-build.outputs.DEB}}" "${NEW_DEB_FILE}"
echo "::set-output name=DEB::${NEW_DEB_FILE}"
# See: https://github.com/actions/upload-artifact
- name: Save DEB as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend DEB for Ubuntu 20.04
path: ${{ steps.debrename.outputs.DEB }}
#
# Build on Ubuntu 20.04 using official go 1.19.1 package
#
Ubuntu-jammy-build:
runs-on: ubuntu-latest
container: ubuntu:22.04
# The job outputs link to the outputs of the 'debrename' step
# Only job outputs can be used in child jobs
outputs:
deb : ${{steps.debrename.outputs.DEB}}
steps:
# Use apt to install development packages
- name: Install development packages
run: |
apt update && apt --assume-yes upgrade
apt --assume-yes install build-essential sed git wget bash npm
npm install --global yarn rollup svelte rollup-plugin-svelte
# Checkout git repository and submodules
# fetch-depth must be 0 to use git describe
# See: https://github.com/marketplace/actions/checkout
- name: Checkout
uses: actions/checkout@v2
with:
submodules: recursive
fetch-depth: 0
# Use official golang package
- name: Install Golang
run: |
wget -q https://go.dev/dl/go1.19.1.linux-amd64.tar.gz
tar -C /usr/local -xzf go1.19.1.linux-amd64.tar.gz
export PATH=/usr/local/go/bin:/usr/local/go/pkg/tool/linux_amd64:$PATH
go version
- name: DEB build ClusterCockpit
id: dpkg-build
run: |
ls -la
pwd
env
export PATH=/usr/local/go/bin:/usr/local/go/pkg/tool/linux_amd64:$PATH
git config --global --add safe.directory $(pwd)
make DEB
- name: Rename DEB (add '_ubuntu22.04')
id: debrename
run: |
OLD_DEB_NAME=$(echo "${{steps.dpkg-build.outputs.DEB}}" | rev | cut -d '.' -f 2- | rev)
NEW_DEB_FILE="${OLD_DEB_NAME}_ubuntu22.04.deb"
mv "${{steps.dpkg-build.outputs.DEB}}" "${NEW_DEB_FILE}"
echo "::set-output name=DEB::${NEW_DEB_FILE}"
# See: https://github.com/actions/upload-artifact
- name: Save DEB as artifact
uses: actions/upload-artifact@v2
with:
name: cc-backend DEB for Ubuntu 22.04
path: ${{ steps.debrename.outputs.DEB }}
#
# Create release with fresh RPMs
#
Release:
runs-on: ubuntu-latest
# We need the RPMs, so add dependency
needs: [AlmaLinux-RPM-build, UBI-8-RPM-build, Ubuntu-focal-build, Ubuntu-jammy-build]
steps:
# See: https://github.com/actions/download-artifact
- name: Download AlmaLinux 8.5 RPM
uses: actions/download-artifact@v2
with:
name: cc-backend RPM for AlmaLinux 8.5
- name: Download AlmaLinux 8.5 SRPM
uses: actions/download-artifact@v2
with:
name: cc-backend SRPM for AlmaLinux 8.5
- name: Download UBI 8 RPM
uses: actions/download-artifact@v2
with:
name: cc-backend RPM for UBI 8
- name: Download UBI 8 SRPM
uses: actions/download-artifact@v2
with:
name: cc-backend SRPM for UBI 8
- name: Download Ubuntu 20.04 DEB
uses: actions/download-artifact@v2
with:
name: cc-backend DEB for Ubuntu 20.04
- name: Download Ubuntu 22.04 DEB
uses: actions/download-artifact@v2
with:
name: cc-backend DEB for Ubuntu 22.04
# The download actions do not publish the name of the downloaded file,
# so we re-use the job outputs of the parent jobs. The files are all
# downloaded to the current folder.
# The gh-release action afterwards does not accept file lists but all
# files have to be listed at 'files'. The step creates one output per
# RPM package (2 per distro)
- name: Set RPM variables
id: files
run: |
ALMA_85_RPM=$(basename "${{ needs.AlmaLinux-RPM-build.outputs.rpm}}")
ALMA_85_SRPM=$(basename "${{ needs.AlmaLinux-RPM-build.outputs.srpm}}")
UBI_8_RPM=$(basename "${{ needs.UBI-8-RPM-build.outputs.rpm}}")
UBI_8_SRPM=$(basename "${{ needs.UBI-8-RPM-build.outputs.srpm}}")
U_2004_DEB=$(basename "${{ needs.Ubuntu-focal-build.outputs.deb}}")
U_2204_DEB=$(basename "${{ needs.Ubuntu-jammy-build.outputs.deb}}")
echo "ALMA_85_RPM::${ALMA_85_RPM}"
echo "ALMA_85_SRPM::${ALMA_85_SRPM}"
echo "UBI_8_RPM::${UBI_8_RPM}"
echo "UBI_8_SRPM::${UBI_8_SRPM}"
echo "U_2004_DEB::${U_2004_DEB}"
echo "U_2204_DEB::${U_2204_DEB}"
echo "::set-output name=ALMA_85_RPM::${ALMA_85_RPM}"
echo "::set-output name=ALMA_85_SRPM::${ALMA_85_SRPM}"
echo "::set-output name=UBI_8_RPM::${UBI_8_RPM}"
echo "::set-output name=UBI_8_SRPM::${UBI_8_SRPM}"
echo "::set-output name=U_2004_DEB::${U_2004_DEB}"
echo "::set-output name=U_2204_DEB::${U_2204_DEB}"
# See: https://github.com/softprops/action-gh-release
- name: Release
uses: softprops/action-gh-release@v1
if: startsWith(github.ref, 'refs/tags/')
with:
name: cc-backend-${{github.ref_name}}
files: |
${{ steps.files.outputs.ALMA_85_RPM }}
${{ steps.files.outputs.ALMA_85_SRPM }}
${{ steps.files.outputs.UBI_8_RPM }}
${{ steps.files.outputs.UBI_8_SRPM }}
${{ steps.files.outputs.U_2004_DEB }}
${{ steps.files.outputs.U_2204_DEB }}

View File

@@ -7,7 +7,7 @@ jobs:
- name: Install Go
uses: actions/setup-go@v4
with:
go-version: 1.25.x
go-version: 1.22.x
- name: Checkout code
uses: actions/checkout@v3
- name: Build, Vet & Test

24
.gitignore vendored
View File

@@ -1,29 +1,21 @@
/cc-backend
/.env
/config.json
/uiConfig.json
/var/job-archive
/var/machine-state
/var/job.db-shm
/var/job.db-wal
/var/*.db
/var/*.txt
/var/machine-state
/var/checkpoints*
migrateTimestamps.pl
test_ccms_write_api.sh
/.env
/config.json
/web/frontend/public/build
/web/frontend/node_modules
/.vscode/*
/archive-migration
/archive-manager
var/job.db-shm
var/job.db-wal
/internal/repository/testdata/job.db-shm
/internal/repository/testdata/job.db-wal
/.vscode/*
dist/
*.db
internal/repository/testdata/job.db-shm
internal/repository/testdata/job.db-wal

View File

@@ -2,7 +2,7 @@ TARGET = ./cc-backend
VAR = ./var
CFG = config.json .env
FRONTEND = ./web/frontend
VERSION = 1.4.4
VERSION = 1.4.2
GIT_HASH := $(shell git rev-parse --short HEAD || echo 'development')
CURRENT_TIME = $(shell date +"%Y-%m-%d:T%H:%M:%S")
LD_FLAGS = '-s -X main.date=${CURRENT_TIME} -X main.version=${VERSION} -X main.commit=${GIT_HASH}'
@@ -52,7 +52,7 @@ frontend:
swagger:
$(info ===> GENERATE swagger)
@go run github.com/swaggo/swag/cmd/swag init --parseDependency -d ./internal/api -g rest.go -o ./api
@go run github.com/swaggo/swag/cmd/swag init -d ./internal/api,./pkg/schema -g rest.go -o ./api
@mv ./api/docs.go ./internal/api/docs.go
graphql:
@@ -82,7 +82,7 @@ tags:
@ctags -R
$(VAR):
@mkdir -p $(VAR)
@mkdir $(VAR)
config.json:
$(info ===> Initialize config.json file)

View File

@@ -1,8 +1,5 @@
# NOTE
While we do our best to keep the master branch in a usable state, there is no guarantee the master branch works.
Please do not use it for production!
Please have a look at the [Release
Notes](https://github.com/ClusterCockpit/cc-backend/blob/master/ReleaseNotes.md)
for breaking changes!

View File

@@ -1,27 +1,13 @@
# `cc-backend` version 1.4.4
# `cc-backend` version 1.4.2
Supports job archive version 2 and database version 8.
This is a bug fix release of `cc-backend`, the API backend and frontend
This is a small bug fix release of `cc-backend`, the API backend and frontend
implementation of ClusterCockpit.
For release specific notes visit the [ClusterCockpit Documentation](https://clusterockpit.org/docs/release/).
## Breaking changes
The option `apiAllowedIPs` is now a required configuration attribute in
`config.json`. This option restricts access to the admin API.
To retain the previous behavior that the API is per default accessible from
everywhere set:
```json
"apiAllowedIPs": [
"*"
]
```
## Breaking changes for minor release 1.4.x
- You need to perform a database migration. Depending on your database size the
migration might require several hours!
- You need to adapt the `cluster.json` configuration files in the job-archive,
@@ -36,7 +22,20 @@ everywhere set:
## New features
- Enable to delete tags from the web interface
- Tags have a scope now. Tags created by a basic user are only visible by that
user. Tags created by an admin/support role can be configured to be visible by
all users (global scope) or only be admin/support role.
- Re-sampling support for running (requires a recent `cc-metric-store`) and
archived jobs. This greatly speeds up loading of large or very long jobs. You
need to add the new configuration key `enable-resampling` to the `config.json`
file.
- For finished jobs a total job energy is shown in the job view.
- Continuous scrolling in job lists is default now.
- All database queries (especially for sqlite) were optimized resulting in
dramatically faster load times.
- A performance and energy footprint can be freely configured on a per
subcluster base. One can filter for footprint statistics for running and
finished jobs.
## Known issues

View File

@@ -4,89 +4,61 @@ scalar Any
scalar NullableFloat
scalar MetricScope
scalar JobState
scalar SchedulerState
scalar MonitoringState
type Node {
id: ID!
hostname: String!
cluster: String!
subCluster: String!
jobsRunning: Int!
cpusAllocated: Int
memoryAllocated: Int
gpusAllocated: Int
schedulerState: SchedulerState!
healthState: MonitoringState!
metaData: Any
}
type NodeStates {
state: String!
count: Int!
}
type NodeStatesTimed {
state: String!
type: String!
count: Int!
time: Int!
}
type Job {
id: ID!
jobId: Int!
user: String!
project: String!
cluster: String!
subCluster: String!
startTime: Time!
duration: Int!
walltime: Int!
numNodes: Int!
numHWThreads: Int!
numAcc: Int!
energy: Float!
SMT: Int!
shared: String!
partition: String!
arrayJobId: Int!
id: ID!
jobId: Int!
user: String!
project: String!
cluster: String!
subCluster: String!
startTime: Time!
duration: Int!
walltime: Int!
numNodes: Int!
numHWThreads: Int!
numAcc: Int!
energy: Float!
SMT: Int!
exclusive: Int!
partition: String!
arrayJobId: Int!
monitoringStatus: Int!
state: JobState!
tags: [Tag!]!
resources: [Resource!]!
concurrentJobs: JobLinkResultList
footprint: [FootprintValue]
energyFootprint: [EnergyFootprintValue]
metaData: Any
userData: User
state: JobState!
tags: [Tag!]!
resources: [Resource!]!
concurrentJobs: JobLinkResultList
footprint: [FootprintValue]
energyFootprint: [EnergyFootprintValue]
metaData: Any
userData: User
}
type JobLink {
id: ID!
jobId: Int!
id: ID!
jobId: Int!
}
type Cluster {
name: String!
partitions: [String!]! # Slurm partitions
subClusters: [SubCluster!]! # Hardware partitions/subclusters
name: String!
partitions: [String!]! # Slurm partitions
subClusters: [SubCluster!]! # Hardware partitions/subclusters
}
type SubCluster {
name: String!
nodes: String!
numberOfNodes: Int!
processorType: String!
socketsPerNode: Int!
coresPerSocket: Int!
threadsPerCore: Int!
flopRateScalar: MetricValue!
flopRateSimd: MetricValue!
name: String!
nodes: String!
numberOfNodes: Int!
processorType: String!
socketsPerNode: Int!
coresPerSocket: Int!
threadsPerCore: Int!
flopRateScalar: MetricValue!
flopRateSimd: MetricValue!
memoryBandwidth: MetricValue!
topology: Topology!
metricConfig: [MetricConfig!]!
footprint: [String!]!
topology: Topology!
metricConfig: [MetricConfig!]!
footprint: [String!]!
}
type FootprintValue {
@@ -108,112 +80,75 @@ type MetricValue {
}
type Topology {
node: [Int!]
socket: [[Int!]!]
node: [Int!]
socket: [[Int!]!]
memoryDomain: [[Int!]!]
die: [[Int!]!]
core: [[Int!]!]
die: [[Int!]!]
core: [[Int!]!]
accelerators: [Accelerator!]
}
type Accelerator {
id: String!
type: String!
id: String!
type: String!
model: String!
}
type SubClusterConfig {
name: String!
peak: Float
normal: Float
name: String!
peak: Float
normal: Float
caution: Float
alert: Float
remove: Boolean
alert: Float
remove: Boolean
}
type MetricConfig {
name: String!
unit: Unit!
scope: MetricScope!
name: String!
unit: Unit!
scope: MetricScope!
aggregation: String!
timestep: Int!
peak: Float!
normal: Float
timestep: Int!
peak: Float!
normal: Float
caution: Float!
alert: Float!
alert: Float!
lowerIsBetter: Boolean
subClusters: [SubClusterConfig!]!
}
type Tag {
id: ID!
id: ID!
type: String!
name: String!
scope: String!
}
type Resource {
hostname: String!
hwthreads: [Int!]
accelerators: [String!]
hostname: String!
hwthreads: [Int!]
accelerators: [String!]
configuration: String
}
type JobMetricWithName {
name: String!
scope: MetricScope!
name: String!
scope: MetricScope!
metric: JobMetric!
}
type JobMetric {
unit: Unit
timestep: Int!
series: [Series!]
unit: Unit
timestep: Int!
series: [Series!]
statisticsSeries: StatsSeries
}
type Series {
hostname: String!
id: String
hostname: String!
id: String
statistics: MetricStatistics
data: [NullableFloat!]!
}
type StatsSeries {
mean: [NullableFloat!]!
median: [NullableFloat!]!
min: [NullableFloat!]!
max: [NullableFloat!]!
}
type NamedStatsWithScope {
name: String!
scope: MetricScope!
stats: [ScopedStats!]!
}
type ScopedStats {
hostname: String!
id: String
data: MetricStatistics!
}
type JobStats {
id: Int!
jobId: String!
startTime: Int!
duration: Int!
cluster: String!
subCluster: String!
numNodes: Int!
numHWThreads: Int
numAccelerators: Int
stats: [NamedStats!]!
}
type NamedStats {
name: String!
data: MetricStatistics!
data: [NullableFloat!]!
}
type Unit {
@@ -227,14 +162,21 @@ type MetricStatistics {
max: Float!
}
type StatsSeries {
mean: [NullableFloat!]!
median: [NullableFloat!]!
min: [NullableFloat!]!
max: [NullableFloat!]!
}
type MetricFootprints {
metric: String!
data: [NullableFloat!]!
data: [NullableFloat!]!
}
type Footprints {
timeWeights: TimeWeights!
metrics: [MetricFootprints!]!
metrics: [MetricFootprints!]!
}
type TimeWeights {
@@ -243,35 +185,20 @@ type TimeWeights {
coreHours: [NullableFloat!]!
}
enum Aggregate {
USER
PROJECT
CLUSTER
SUBCLUSTER
}
enum SortByAggregate {
TOTALWALLTIME
TOTALJOBS
TOTALUSERS
TOTALNODES
TOTALNODEHOURS
TOTALCORES
TOTALCOREHOURS
TOTALACCS
TOTALACCHOURS
}
enum Aggregate { USER, PROJECT, CLUSTER }
enum SortByAggregate { TOTALWALLTIME, TOTALJOBS, TOTALNODES, TOTALNODEHOURS, TOTALCORES, TOTALCOREHOURS, TOTALACCS, TOTALACCHOURS }
type NodeMetrics {
host: String!
host: String!
subCluster: String!
metrics: [JobMetricWithName!]!
metrics: [JobMetricWithName!]!
}
type NodesResultList {
items: [NodeMetrics!]!
items: [NodeMetrics!]!
offset: Int
limit: Int
count: Int
limit: Int
count: Int
totalNodes: Int
hasNextPage: Boolean
}
@@ -290,14 +217,14 @@ type GlobalMetricListItem {
}
type Count {
name: String!
name: String!
count: Int!
}
type User {
username: String!
name: String!
email: String!
name: String!
email: String!
}
input MetricStatItem {
@@ -306,84 +233,24 @@ input MetricStatItem {
}
type Query {
clusters: [Cluster!]! # List of all clusters
tags: [Tag!]! # List of all tags
globalMetrics: [GlobalMetricListItem!]!
clusters: [Cluster!]! # List of all clusters
tags: [Tag!]! # List of all tags
globalMetrics: [GlobalMetricListItem!]!
user(username: String!): User
allocatedNodes(cluster: String!): [Count!]!
## Node Queries New
node(id: ID!): Node
nodes(filter: [NodeFilter!], order: OrderByInput): NodeStateResultList!
nodeStates(filter: [NodeFilter!]): [NodeStates!]!
nodeStatesTimed(filter: [NodeFilter!]): [NodeStatesTimed!]!
job(id: ID!): Job
jobMetrics(
id: ID!
metrics: [String!]
scopes: [MetricScope!]
resolution: Int
): [JobMetricWithName!]!
jobStats(id: ID!, metrics: [String!]): [NamedStats!]!
scopedJobStats(
id: ID!
metrics: [String!]
scopes: [MetricScope!]
): [NamedStatsWithScope!]!
jobs(
filter: [JobFilter!]
page: PageRequest
order: OrderByInput
): JobResultList!
jobsStatistics(
filter: [JobFilter!]
metrics: [String!]
page: PageRequest
sortBy: SortByAggregate
groupBy: Aggregate
numDurationBins: String
numMetricBins: Int
): [JobsStatistics!]!
jobsMetricStats(filter: [JobFilter!], metrics: [String!]): [JobStats!]!
jobMetrics(id: ID!, metrics: [String!], scopes: [MetricScope!], resolution: Int): [JobMetricWithName!]!
jobsFootprints(filter: [JobFilter!], metrics: [String!]!): Footprints
rooflineHeatmap(
filter: [JobFilter!]!
rows: Int!
cols: Int!
minX: Float!
minY: Float!
maxX: Float!
maxY: Float!
): [[Float!]!]!
jobs(filter: [JobFilter!], page: PageRequest, order: OrderByInput): JobResultList!
jobsStatistics(filter: [JobFilter!], metrics: [String!], page: PageRequest, sortBy: SortByAggregate, groupBy: Aggregate, numDurationBins: String, numMetricBins: Int): [JobsStatistics!]!
nodeMetrics(
cluster: String!
nodes: [String!]
scopes: [MetricScope!]
metrics: [String!]
from: Time!
to: Time!
): [NodeMetrics!]!
rooflineHeatmap(filter: [JobFilter!]!, rows: Int!, cols: Int!, minX: Float!, minY: Float!, maxX: Float!, maxY: Float!): [[Float!]!]!
nodeMetricsList(
cluster: String!
subCluster: String!
nodeFilter: String!
scopes: [MetricScope!]
metrics: [String!]
from: Time!
to: Time!
page: PageRequest
resolution: Int
): NodesResultList!
nodeMetrics(cluster: String!, nodes: [String!], scopes: [MetricScope!], metrics: [String!], from: Time!, to: Time!): [NodeMetrics!]!
nodeMetricsList(cluster: String!, subCluster: String!, nodeFilter: String!, scopes: [MetricScope!], metrics: [String!], from: Time!, to: Time!, page: PageRequest, resolution: Int): NodesResultList!
}
type Mutation {
@@ -391,59 +258,41 @@ type Mutation {
deleteTag(id: ID!): ID!
addTagsToJob(job: ID!, tagIds: [ID!]!): [Tag!]!
removeTagsFromJob(job: ID!, tagIds: [ID!]!): [Tag!]!
removeTagFromList(tagIds: [ID!]!): [Int!]!
updateConfiguration(name: String!, value: String!): String
}
type IntRangeOutput {
from: Int!
to: Int!
}
type TimeRangeOutput {
range: String
from: Time!
to: Time!
}
input NodeFilter {
hostname: StringInput
cluster: StringInput
subcluster: StringInput
schedulerState: SchedulerState
healthState: MonitoringState
timeStart: Int
}
type IntRangeOutput { from: Int!, to: Int! }
type TimeRangeOutput { range: String, from: Time!, to: Time! }
input JobFilter {
tags: [ID!]
dbId: [ID!]
jobId: StringInput
arrayJobId: Int
user: StringInput
project: StringInput
jobName: StringInput
cluster: StringInput
partition: StringInput
duration: IntRange
energy: FloatRange
tags: [ID!]
jobId: StringInput
arrayJobId: Int
user: StringInput
project: StringInput
jobName: StringInput
cluster: StringInput
partition: StringInput
duration: IntRange
energy: FloatRange
minRunningFor: Int
numNodes: IntRange
numNodes: IntRange
numAccelerators: IntRange
numHWThreads: IntRange
numHWThreads: IntRange
startTime: TimeRange
state: [JobState!]
startTime: TimeRange
state: [JobState!]
metricStats: [MetricStatItem!]
shared: String
node: StringInput
exclusive: Int
node: StringInput
}
input OrderByInput {
field: String!
type: String!
type: String!,
order: SortDirectionEnum! = ASC
}
@@ -453,46 +302,34 @@ enum SortDirectionEnum {
}
input StringInput {
eq: String
neq: String
contains: String
eq: String
neq: String
contains: String
startsWith: String
endsWith: String
in: [String!]
endsWith: String
in: [String!]
}
input IntRange {
from: Int!
to: Int!
}
input TimeRange {
range: String
from: Time
to: Time
}
input IntRange { from: Int!, to: Int! }
input TimeRange { range: String, from: Time, to: Time }
input FloatRange {
from: Float!
to: Float!
}
type NodeStateResultList {
items: [Node!]!
count: Int
}
type JobResultList {
items: [Job!]!
items: [Job!]!
offset: Int
limit: Int
count: Int
limit: Int
count: Int
hasNextPage: Boolean
}
type JobLinkResultList {
listQuery: String
items: [JobLink!]!
count: Int
items: [JobLink!]!
count: Int
}
type HistoPoint {
@@ -514,28 +351,27 @@ type MetricHistoPoint {
max: Int
}
type JobsStatistics {
id: ID! # If `groupBy` was used, ID of the user/project/cluster/subcluster
name: String! # if User-Statistics: Given Name of Account (ID) Owner
totalUsers: Int! # if *not* User-Statistics: Number of active users (based on running jobs)
totalJobs: Int! # Number of jobs
runningJobs: Int! # Number of running jobs
shortJobs: Int! # Number of jobs with a duration of less than config'd ShortRunningJobsDuration
totalWalltime: Int! # Sum of the duration of all matched jobs in hours
totalNodes: Int! # Sum of the nodes of all matched jobs
totalNodeHours: Int! # Sum of the node hours of all matched jobs
totalCores: Int! # Sum of the cores of all matched jobs
totalCoreHours: Int! # Sum of the core hours of all matched jobs
totalAccs: Int! # Sum of the accs of all matched jobs
totalAccHours: Int! # Sum of the gpu hours of all matched jobs
histDuration: [HistoPoint!]! # value: hour, count: number of jobs with a rounded duration of value
histNumNodes: [HistoPoint!]! # value: number of nodes, count: number of jobs with that number of nodes
histNumCores: [HistoPoint!]! # value: number of cores, count: number of jobs with that number of cores
histNumAccs: [HistoPoint!]! # value: number of accs, count: number of jobs with that number of accs
histMetrics: [MetricHistoPoints!]! # metric: metricname, data array of histopoints: value: metric average bin, count: number of jobs with that metric average
type JobsStatistics {
id: ID! # If `groupBy` was used, ID of the user/project/cluster
name: String! # if User-Statistics: Given Name of Account (ID) Owner
totalJobs: Int! # Number of jobs
runningJobs: Int! # Number of running jobs
shortJobs: Int! # Number of jobs with a duration of less than duration
totalWalltime: Int! # Sum of the duration of all matched jobs in hours
totalNodes: Int! # Sum of the nodes of all matched jobs
totalNodeHours: Int! # Sum of the node hours of all matched jobs
totalCores: Int! # Sum of the cores of all matched jobs
totalCoreHours: Int! # Sum of the core hours of all matched jobs
totalAccs: Int! # Sum of the accs of all matched jobs
totalAccHours: Int! # Sum of the gpu hours of all matched jobs
histDuration: [HistoPoint!]! # value: hour, count: number of jobs with a rounded duration of value
histNumNodes: [HistoPoint!]! # value: number of nodes, count: number of jobs with that number of nodes
histNumCores: [HistoPoint!]! # value: number of cores, count: number of jobs with that number of cores
histNumAccs: [HistoPoint!]! # value: number of accs, count: number of jobs with that number of accs
histMetrics: [MetricHistoPoints!]! # metric: metricname, data array of histopoints: value: metric average bin, count: number of jobs with that metric average
}
input PageRequest {
itemsPerPage: Int!
page: Int!
page: Int!
}

View File

@@ -15,8 +15,9 @@
"version": "1.0.0"
},
"host": "localhost:8080",
"basePath": "/api",
"paths": {
"/api/clusters/": {
"/clusters/": {
"get": {
"security": [
{
@@ -73,7 +74,7 @@
}
}
},
"/api/jobs/": {
"/jobs/": {
"get": {
"security": [
{
@@ -168,7 +169,7 @@
}
}
},
"/api/jobs/delete_job/": {
"/jobs/delete_job/": {
"delete": {
"security": [
{
@@ -201,7 +202,7 @@
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -243,7 +244,7 @@
}
}
},
"/api/jobs/delete_job/{id}": {
"/jobs/delete_job/{id}": {
"delete": {
"security": [
{
@@ -271,7 +272,7 @@
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -313,7 +314,7 @@
}
}
},
"/api/jobs/delete_job_before/{ts}": {
"/jobs/delete_job_before/{ts}": {
"delete": {
"security": [
{
@@ -341,7 +342,7 @@
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -383,7 +384,7 @@
}
}
},
"/api/jobs/edit_meta/{id}": {
"/jobs/edit_meta/{id}": {
"post": {
"security": [
{
@@ -453,14 +454,14 @@
}
}
},
"/api/jobs/start_job/": {
"/jobs/start_job/": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job specified in request body will be saved to database as \"running\" with new DB ID.\nJob specifications follow the 'Job' scheme, API will fail to execute if requirements are not met.",
"description": "Job specified in request body will be saved to database as \"running\" with new DB ID.\nJob specifications follow the 'JobMeta' scheme, API will fail to execute if requirements are not met.",
"consumes": [
"application/json"
],
@@ -478,7 +479,7 @@
"in": "body",
"required": true,
"schema": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
}
],
@@ -486,7 +487,7 @@
"201": {
"description": "Job added successfully",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.StartJobApiResponse"
}
},
"400": {
@@ -522,14 +523,14 @@
}
}
},
"/api/jobs/stop_job/": {
"/jobs/stop_job/": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job to stop is specified by request body. All fields are required in this case.\nReturns full job resource information according to 'Job' scheme.",
"description": "Job to stop is specified by request body. All fields are required in this case.\nReturns full job resource information according to 'JobMeta' scheme.",
"produces": [
"application/json"
],
@@ -552,7 +553,7 @@
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
},
"400": {
@@ -580,7 +581,7 @@
}
},
"422": {
"description": "Unprocessable Entity: job has already been stopped",
"description": "Unprocessable Entity: finding job failed: sql: no rows in result set",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
@@ -594,7 +595,7 @@
}
}
},
"/api/jobs/tag_job/{id}": {
"/jobs/tag_job/{id}": {
"post": {
"security": [
{
@@ -667,14 +668,14 @@
}
}
},
"/api/jobs/{id}": {
"/jobs/{id}": {
"get": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.",
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.",
"produces": [
"application/json"
],
@@ -748,7 +749,7 @@
"ApiKeyAuth": []
}
],
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.",
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.",
"consumes": [
"application/json"
],
@@ -826,74 +827,119 @@
}
}
},
"/api/nodestats/": {
"/user/{id}": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.",
"description": "Modifies user defined by username (id) in one of four possible ways.\nIf more than one formValue is set then only the highest priority field is used.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"multipart/form-data"
],
"produces": [
"application/json"
"text/plain"
],
"tags": [
"Nodestates"
"User"
],
"summary": "Deliver updated Slurm node states",
"summary": "Updates an existing user",
"parameters": [
{
"description": "Request body containing nodes and their states",
"name": "request",
"in": "body",
"required": true,
"schema": {
"$ref": "#/definitions/api.UpdateNodeStatesRequest"
}
"type": "string",
"description": "Database ID of User",
"name": "id",
"in": "path",
"required": true
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "Priority 1: Role to add",
"name": "add-role",
"in": "formData"
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "Priority 2: Role to remove",
"name": "remove-role",
"in": "formData"
},
{
"type": "string",
"description": "Priority 3: Project to add",
"name": "add-project",
"in": "formData"
},
{
"type": "string",
"description": "Priority 4: Project to remove",
"name": "remove-project",
"in": "formData"
}
],
"responses": {
"200": {
"description": "Success message",
"description": "Success Response Message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"type": "string"
}
},
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"403": {
"description": "Forbidden",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: The user could not be updated",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
}
}
}
},
"/api/users/": {
"/users/": {
"get": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.",
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"produces": [
"application/json"
],
@@ -945,111 +991,70 @@
}
}
}
}
},
"/jobs/tag_job/{id}": {
"delete": {
},
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Removes tag(s) from a job specified by DB ID. Name and Type of Tag(s) must match.\nTag Scope is required for matching, options: \"global\", \"admin\". Private tags can not be deleted via API.\nIf tagged job is already finished: Tag will be removed from respective archive files.",
"description": "User specified in form data will be saved to database.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"application/json"
],
"produces": [
"application/json"
],
"tags": [
"Job add and modify"
],
"summary": "Removes one or more tags from a job",
"parameters": [
{
"type": "integer",
"description": "Job Database ID",
"name": "id",
"in": "path",
"required": true
},
{
"description": "Array of tag-objects to remove",
"name": "request",
"in": "body",
"required": true,
"schema": {
"type": "array",
"items": {
"$ref": "#/definitions/api.ApiTag"
}
}
}
],
"responses": {
"200": {
"description": "Updated job resource",
"schema": {
"$ref": "#/definitions/schema.Job"
}
},
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"404": {
"description": "Job or tag does not exist",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
}
}
}
},
"/tags/": {
"delete": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Removes tags by type and name. Name and Type of Tag(s) must match.\nTag Scope is required for matching, options: \"global\", \"admin\". Private tags can not be deleted via API.\nTag wills be removed from respective archive files.",
"consumes": [
"application/json"
"multipart/form-data"
],
"produces": [
"text/plain"
],
"tags": [
"Tag remove"
"User"
],
"summary": "Removes all tags and job-relations for type:name tuple",
"summary": "Adds a new user",
"parameters": [
{
"description": "Array of tag-objects to remove",
"name": "request",
"in": "body",
"required": true,
"schema": {
"type": "array",
"items": {
"$ref": "#/definitions/api.ApiTag"
}
}
"type": "string",
"description": "Unique user ID",
"name": "username",
"in": "formData",
"required": true
},
{
"type": "string",
"description": "User password",
"name": "password",
"in": "formData",
"required": true
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "User role",
"name": "role",
"in": "formData",
"required": true
},
{
"type": "string",
"description": "Managed project, required for new manager role user",
"name": "project",
"in": "formData"
},
{
"type": "string",
"description": "Users name",
"name": "name",
"in": "formData"
},
{
"type": "string",
"description": "Users email",
"name": "email",
"in": "formData"
}
],
"responses": {
@@ -1062,25 +1067,93 @@
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"404": {
"description": "Job or tag does not exist",
"403": {
"description": "Forbidden",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: creating user failed",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
}
}
},
"delete": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "User defined by username in form data will be deleted from database.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"multipart/form-data"
],
"produces": [
"text/plain"
],
"tags": [
"User"
],
"summary": "Deletes a user",
"parameters": [
{
"type": "string",
"description": "User ID to delete",
"name": "username",
"in": "formData",
"required": true
}
],
"responses": {
"200": {
"description": "User deleted successfully"
},
"400": {
"description": "Bad Request",
"schema": {
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"type": "string"
}
},
"403": {
"description": "Forbidden",
"schema": {
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: deleting user failed",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"type": "string"
}
}
}
@@ -1134,14 +1207,6 @@
}
}
},
"api.DefaultApiResponse": {
"type": "object",
"properties": {
"msg": {
"type": "string"
}
}
},
"api.DeleteJobApiRequest": {
"type": "object",
"required": [
@@ -1165,6 +1230,14 @@
}
}
},
"api.DeleteJobApiResponse": {
"type": "object",
"properties": {
"msg": {
"type": "string"
}
}
},
"api.EditMetaRequest": {
"type": "object",
"properties": {
@@ -1228,7 +1301,7 @@
"description": "Array of jobs",
"type": "array",
"items": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
},
"page": {
@@ -1251,35 +1324,11 @@
}
}
},
"api.Node": {
"api.StartJobApiResponse": {
"type": "object",
"properties": {
"cpusAllocated": {
"type": "integer"
},
"cpusTotal": {
"type": "integer"
},
"gpusAllocated": {
"type": "integer"
},
"gpusTotal": {
"type": "integer"
},
"hostname": {
"msg": {
"type": "string"
},
"memoryAllocated": {
"type": "integer"
},
"memoryTotal": {
"type": "integer"
},
"states": {
"type": "array",
"items": {
"type": "string"
}
}
}
},
@@ -1316,21 +1365,6 @@
}
}
},
"api.UpdateNodeStatesRequest": {
"type": "object",
"properties": {
"cluster": {
"type": "string",
"example": "fritz"
},
"nodes": {
"type": "array",
"items": {
"$ref": "#/definitions/api.Node"
}
}
}
},
"schema.Accelerator": {
"type": "object",
"properties": {
@@ -1366,6 +1400,7 @@
}
},
"schema.Job": {
"description": "Information of a HPC job.",
"type": "object",
"properties": {
"arrayJobId": {
@@ -1390,15 +1425,19 @@
"energyFootprint": {
"type": "object",
"additionalProperties": {
"type": "number",
"format": "float64"
"type": "number"
}
},
"exclusive": {
"type": "integer",
"maximum": 2,
"minimum": 0,
"example": 1
},
"footprint": {
"type": "object",
"additionalProperties": {
"type": "number",
"format": "float64"
"type": "number"
}
},
"id": {
@@ -1410,18 +1449,12 @@
},
"jobState": {
"enum": [
"boot_fail",
"cancelled",
"completed",
"deadline",
"failed",
"node_fail",
"out-of-memory",
"pending",
"preempted",
"running",
"suspended",
"timeout"
"cancelled",
"stopped",
"timeout",
"out_of_memory"
],
"allOf": [
{
@@ -1465,48 +1498,23 @@
"type": "string",
"example": "abcd200"
},
"requestedMemory": {
"description": "in MB",
"type": "integer",
"minimum": 1,
"example": 128000
},
"resources": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Resource"
}
},
"shared": {
"type": "string",
"enum": [
"none",
"single_user",
"multi_user"
]
},
"smt": {
"type": "integer",
"example": 4
},
"startTime": {
"type": "integer",
"example": 1649723812
},
"statistics": {
"type": "object",
"additionalProperties": {
"$ref": "#/definitions/schema.JobStatistics"
}
"type": "string"
},
"subCluster": {
"type": "string",
"example": "main"
},
"submitTime": {
"type": "integer",
"example": 1649723812
},
"tags": {
"type": "array",
"items": {
@@ -1549,6 +1557,147 @@
}
}
},
"schema.JobMeta": {
"description": "Meta data information of a HPC job.",
"type": "object",
"properties": {
"arrayJobId": {
"type": "integer",
"example": 123000
},
"cluster": {
"type": "string",
"example": "fritz"
},
"concurrentJobs": {
"$ref": "#/definitions/schema.JobLinkResultList"
},
"duration": {
"type": "integer",
"minimum": 1,
"example": 43200
},
"energy": {
"type": "number"
},
"energyFootprint": {
"type": "object",
"additionalProperties": {
"type": "number"
}
},
"exclusive": {
"type": "integer",
"maximum": 2,
"minimum": 0,
"example": 1
},
"footprint": {
"type": "object",
"additionalProperties": {
"type": "number"
}
},
"id": {
"type": "integer"
},
"jobId": {
"type": "integer",
"example": 123000
},
"jobState": {
"enum": [
"completed",
"failed",
"cancelled",
"stopped",
"timeout",
"out_of_memory"
],
"allOf": [
{
"$ref": "#/definitions/schema.JobState"
}
],
"example": "completed"
},
"metaData": {
"type": "object",
"additionalProperties": {
"type": "string"
}
},
"monitoringStatus": {
"type": "integer",
"maximum": 3,
"minimum": 0,
"example": 1
},
"numAcc": {
"type": "integer",
"minimum": 1,
"example": 2
},
"numHwthreads": {
"type": "integer",
"minimum": 1,
"example": 20
},
"numNodes": {
"type": "integer",
"minimum": 1,
"example": 2
},
"partition": {
"type": "string",
"example": "main"
},
"project": {
"type": "string",
"example": "abcd200"
},
"resources": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Resource"
}
},
"smt": {
"type": "integer",
"example": 4
},
"startTime": {
"type": "integer",
"minimum": 1,
"example": 1649723812
},
"statistics": {
"type": "object",
"additionalProperties": {
"$ref": "#/definitions/schema.JobStatistics"
}
},
"subCluster": {
"type": "string",
"example": "main"
},
"tags": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Tag"
}
},
"user": {
"type": "string",
"example": "abcd100h"
},
"walltime": {
"type": "integer",
"minimum": 1,
"example": 86400
}
}
},
"schema.JobMetric": {
"type": "object",
"properties": {
@@ -1572,32 +1721,24 @@
"schema.JobState": {
"type": "string",
"enum": [
"boot_fail",
"cancelled",
"completed",
"deadline",
"failed",
"node_fail",
"out_of_memory",
"pending",
"preempted",
"running",
"suspended",
"timeout"
"completed",
"failed",
"cancelled",
"stopped",
"timeout",
"preempted",
"out_of_memory"
],
"x-enum-varnames": [
"JobStateBootFail",
"JobStateCancelled",
"JobStateCompleted",
"JobStateDeadline",
"JobStateFailed",
"JobStateNodeFail",
"JobStateOutOfMemory",
"JobStatePending",
"JobStatePreempted",
"JobStateRunning",
"JobStateSuspended",
"JobStateTimeout"
"JobStateCompleted",
"JobStateFailed",
"JobStateCancelled",
"JobStateStopped",
"JobStateTimeout",
"JobStatePreempted",
"JobStateOutOfMemory"
]
},
"schema.JobStatistics": {
@@ -1796,8 +1937,7 @@
"additionalProperties": {
"type": "array",
"items": {
"type": "number",
"format": "float64"
"type": "number"
}
}
}
@@ -1885,9 +2025,6 @@
},
"remove": {
"type": "boolean"
},
"unit": {
"$ref": "#/definitions/schema.Unit"
}
}
},

View File

@@ -1,3 +1,4 @@
basePath: /api
definitions:
api.ApiReturnedUser:
properties:
@@ -31,11 +32,6 @@ definitions:
example: Debug
type: string
type: object
api.DefaultApiResponse:
properties:
msg:
type: string
type: object
api.DeleteJobApiRequest:
properties:
cluster:
@@ -53,6 +49,11 @@ definitions:
required:
- jobId
type: object
api.DeleteJobApiResponse:
properties:
msg:
type: string
type: object
api.EditMetaRequest:
properties:
key:
@@ -96,7 +97,7 @@ definitions:
jobs:
description: Array of jobs
items:
$ref: '#/definitions/schema.Job'
$ref: '#/definitions/schema.JobMeta'
type: array
page:
description: Page id returned
@@ -111,26 +112,10 @@ definitions:
scope:
$ref: '#/definitions/schema.MetricScope'
type: object
api.Node:
api.StartJobApiResponse:
properties:
cpusAllocated:
type: integer
cpusTotal:
type: integer
gpusAllocated:
type: integer
gpusTotal:
type: integer
hostname:
msg:
type: string
memoryAllocated:
type: integer
memoryTotal:
type: integer
states:
items:
type: string
type: array
type: object
api.StopJobApiRequest:
properties:
@@ -154,16 +139,6 @@ definitions:
- jobState
- stopTime
type: object
api.UpdateNodeStatesRequest:
properties:
cluster:
example: fritz
type: string
nodes:
items:
$ref: '#/definitions/api.Node'
type: array
type: object
schema.Accelerator:
properties:
id:
@@ -187,6 +162,7 @@ definitions:
type: array
type: object
schema.Job:
description: Information of a HPC job.
properties:
arrayJobId:
example: 123000
@@ -204,12 +180,15 @@ definitions:
type: number
energyFootprint:
additionalProperties:
format: float64
type: number
type: object
exclusive:
example: 1
maximum: 2
minimum: 0
type: integer
footprint:
additionalProperties:
format: float64
type: number
type: object
id:
@@ -221,18 +200,12 @@ definitions:
allOf:
- $ref: '#/definitions/schema.JobState'
enum:
- boot_fail
- cancelled
- completed
- deadline
- failed
- node_fail
- out-of-memory
- pending
- preempted
- running
- suspended
- cancelled
- stopped
- timeout
- out_of_memory
example: completed
metaData:
additionalProperties:
@@ -261,37 +234,18 @@ definitions:
project:
example: abcd200
type: string
requestedMemory:
description: in MB
example: 128000
minimum: 1
type: integer
resources:
items:
$ref: '#/definitions/schema.Resource'
type: array
shared:
enum:
- none
- single_user
- multi_user
type: string
smt:
example: 4
type: integer
startTime:
example: 1649723812
type: integer
statistics:
additionalProperties:
$ref: '#/definitions/schema.JobStatistics'
type: object
type: string
subCluster:
example: main
type: string
submitTime:
example: 1649723812
type: integer
tags:
items:
$ref: '#/definitions/schema.Tag'
@@ -320,6 +274,109 @@ definitions:
$ref: '#/definitions/schema.JobLink'
type: array
type: object
schema.JobMeta:
description: Meta data information of a HPC job.
properties:
arrayJobId:
example: 123000
type: integer
cluster:
example: fritz
type: string
concurrentJobs:
$ref: '#/definitions/schema.JobLinkResultList'
duration:
example: 43200
minimum: 1
type: integer
energy:
type: number
energyFootprint:
additionalProperties:
type: number
type: object
exclusive:
example: 1
maximum: 2
minimum: 0
type: integer
footprint:
additionalProperties:
type: number
type: object
id:
type: integer
jobId:
example: 123000
type: integer
jobState:
allOf:
- $ref: '#/definitions/schema.JobState'
enum:
- completed
- failed
- cancelled
- stopped
- timeout
- out_of_memory
example: completed
metaData:
additionalProperties:
type: string
type: object
monitoringStatus:
example: 1
maximum: 3
minimum: 0
type: integer
numAcc:
example: 2
minimum: 1
type: integer
numHwthreads:
example: 20
minimum: 1
type: integer
numNodes:
example: 2
minimum: 1
type: integer
partition:
example: main
type: string
project:
example: abcd200
type: string
resources:
items:
$ref: '#/definitions/schema.Resource'
type: array
smt:
example: 4
type: integer
startTime:
example: 1649723812
minimum: 1
type: integer
statistics:
additionalProperties:
$ref: '#/definitions/schema.JobStatistics'
type: object
subCluster:
example: main
type: string
tags:
items:
$ref: '#/definitions/schema.Tag'
type: array
user:
example: abcd100h
type: string
walltime:
example: 86400
minimum: 1
type: integer
type: object
schema.JobMetric:
properties:
series:
@@ -335,32 +392,24 @@ definitions:
type: object
schema.JobState:
enum:
- boot_fail
- cancelled
- completed
- deadline
- failed
- node_fail
- out_of_memory
- pending
- preempted
- running
- suspended
- completed
- failed
- cancelled
- stopped
- timeout
- preempted
- out_of_memory
type: string
x-enum-varnames:
- JobStateBootFail
- JobStateCancelled
- JobStateCompleted
- JobStateDeadline
- JobStateFailed
- JobStateNodeFail
- JobStateOutOfMemory
- JobStatePending
- JobStatePreempted
- JobStateRunning
- JobStateSuspended
- JobStateCompleted
- JobStateFailed
- JobStateCancelled
- JobStateStopped
- JobStateTimeout
- JobStatePreempted
- JobStateOutOfMemory
schema.JobStatistics:
description: Specification for job metric statistics.
properties:
@@ -497,7 +546,6 @@ definitions:
percentiles:
additionalProperties:
items:
format: float64
type: number
type: array
type: object
@@ -557,8 +605,6 @@ definitions:
type: number
remove:
type: boolean
unit:
$ref: '#/definitions/schema.Unit'
type: object
schema.Tag:
description: Defines a tag using name and type.
@@ -630,7 +676,7 @@ info:
title: ClusterCockpit REST API
version: 1.0.0
paths:
/api/clusters/:
/clusters/:
get:
description: Get a list of all cluster configs. Specific cluster can be requested
using query parameter.
@@ -667,7 +713,7 @@ paths:
summary: Lists all cluster configs
tags:
- Cluster query
/api/jobs/:
/jobs/:
get:
description: |-
Get a list of all jobs. Filters can be applied using query parameters.
@@ -732,11 +778,11 @@ paths:
summary: Lists all jobs
tags:
- Job query
/api/jobs/{id}:
/jobs/{id}:
get:
description: |-
Job to get is specified by database ID
Returns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.
Returns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.
parameters:
- description: Database ID of Job
in: path
@@ -789,7 +835,7 @@ paths:
- application/json
description: |-
Job to get is specified by database ID
Returns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.
Returns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.
parameters:
- description: Database ID of Job
in: path
@@ -841,7 +887,7 @@ paths:
summary: Get job meta and configurable metric data
tags:
- Job query
/api/jobs/delete_job/:
/jobs/delete_job/:
delete:
consumes:
- application/json
@@ -860,7 +906,7 @@ paths:
"200":
description: Success message
schema:
$ref: '#/definitions/api.DefaultApiResponse'
$ref: '#/definitions/api.DeleteJobApiResponse'
"400":
description: Bad Request
schema:
@@ -891,7 +937,7 @@ paths:
summary: Remove a job from the sql database
tags:
- Job remove
/api/jobs/delete_job/{id}:
/jobs/delete_job/{id}:
delete:
description: Job to remove is specified by database ID. This will not remove
the job from the job archive.
@@ -907,7 +953,7 @@ paths:
"200":
description: Success message
schema:
$ref: '#/definitions/api.DefaultApiResponse'
$ref: '#/definitions/api.DeleteJobApiResponse'
"400":
description: Bad Request
schema:
@@ -938,7 +984,7 @@ paths:
summary: Remove a job from the sql database
tags:
- Job remove
/api/jobs/delete_job_before/{ts}:
/jobs/delete_job_before/{ts}:
delete:
description: Remove all jobs with start time before timestamp. The jobs will
not be removed from the job archive.
@@ -954,7 +1000,7 @@ paths:
"200":
description: Success message
schema:
$ref: '#/definitions/api.DefaultApiResponse'
$ref: '#/definitions/api.DeleteJobApiResponse'
"400":
description: Bad Request
schema:
@@ -985,7 +1031,7 @@ paths:
summary: Remove a job from the sql database
tags:
- Job remove
/api/jobs/edit_meta/{id}:
/jobs/edit_meta/{id}:
post:
consumes:
- application/json
@@ -1032,27 +1078,27 @@ paths:
summary: Edit meta-data json
tags:
- Job add and modify
/api/jobs/start_job/:
/jobs/start_job/:
post:
consumes:
- application/json
description: |-
Job specified in request body will be saved to database as "running" with new DB ID.
Job specifications follow the 'Job' scheme, API will fail to execute if requirements are not met.
Job specifications follow the 'JobMeta' scheme, API will fail to execute if requirements are not met.
parameters:
- description: Job to add
in: body
name: request
required: true
schema:
$ref: '#/definitions/schema.Job'
$ref: '#/definitions/schema.JobMeta'
produces:
- application/json
responses:
"201":
description: Job added successfully
schema:
$ref: '#/definitions/api.DefaultApiResponse'
$ref: '#/definitions/api.StartJobApiResponse'
"400":
description: Bad Request
schema:
@@ -1079,11 +1125,11 @@ paths:
summary: Adds a new job as "running"
tags:
- Job add and modify
/api/jobs/stop_job/:
/jobs/stop_job/:
post:
description: |-
Job to stop is specified by request body. All fields are required in this case.
Returns full job resource information according to 'Job' scheme.
Returns full job resource information according to 'JobMeta' scheme.
parameters:
- description: All fields required
in: body
@@ -1097,7 +1143,7 @@ paths:
"200":
description: Success message
schema:
$ref: '#/definitions/schema.Job'
$ref: '#/definitions/schema.JobMeta'
"400":
description: Bad Request
schema:
@@ -1115,7 +1161,8 @@ paths:
schema:
$ref: '#/definitions/api.ErrorResponse'
"422":
description: 'Unprocessable Entity: job has already been stopped'
description: 'Unprocessable Entity: finding job failed: sql: no rows in
result set'
schema:
$ref: '#/definitions/api.ErrorResponse'
"500":
@@ -1127,7 +1174,7 @@ paths:
summary: Marks job as completed and triggers archiving
tags:
- Job add and modify
/api/jobs/tag_job/{id}:
/jobs/tag_job/{id}:
post:
consumes:
- application/json
@@ -1177,51 +1224,128 @@ paths:
summary: Adds one or more tags to a job
tags:
- Job add and modify
/api/nodestats/:
/user/{id}:
post:
consumes:
- multipart/form-data
description: |-
Returns a JSON-encoded list of users.
Required query-parameter defines if all users or only users with additional special roles are returned.
Modifies user defined by username (id) in one of four possible ways.
If more than one formValue is set then only the highest priority field is used.
Only accessible from IPs registered with apiAllowedIPs configuration option.
parameters:
- description: Request body containing nodes and their states
in: body
name: request
- description: Database ID of User
in: path
name: id
required: true
schema:
$ref: '#/definitions/api.UpdateNodeStatesRequest'
type: string
- description: 'Priority 1: Role to add'
enum:
- admin
- support
- manager
- user
- api
in: formData
name: add-role
type: string
- description: 'Priority 2: Role to remove'
enum:
- admin
- support
- manager
- user
- api
in: formData
name: remove-role
type: string
- description: 'Priority 3: Project to add'
in: formData
name: add-project
type: string
- description: 'Priority 4: Project to remove'
in: formData
name: remove-project
type: string
produces:
- application/json
- text/plain
responses:
"200":
description: Success message
description: Success Response Message
schema:
$ref: '#/definitions/api.DefaultApiResponse'
type: string
"400":
description: Bad Request
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
"401":
description: Unauthorized
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
"403":
description: Forbidden
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
"422":
description: 'Unprocessable Entity: The user could not be updated'
schema:
type: string
"500":
description: Internal Server Error
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
security:
- ApiKeyAuth: []
summary: Deliver updated Slurm node states
summary: Updates an existing user
tags:
- Nodestates
/api/users/:
- User
/users/:
delete:
consumes:
- multipart/form-data
description: |-
User defined by username in form data will be deleted from database.
Only accessible from IPs registered with apiAllowedIPs configuration option.
parameters:
- description: User ID to delete
in: formData
name: username
required: true
type: string
produces:
- text/plain
responses:
"200":
description: User deleted successfully
"400":
description: Bad Request
schema:
type: string
"401":
description: Unauthorized
schema:
type: string
"403":
description: Forbidden
schema:
type: string
"422":
description: 'Unprocessable Entity: deleting user failed'
schema:
type: string
"500":
description: Internal Server Error
schema:
type: string
security:
- ApiKeyAuth: []
summary: Deletes a user
tags:
- User
get:
description: |-
Returns a JSON-encoded list of users.
Required query-parameter defines if all users or only users with additional special roles are returned.
Only accessible from IPs registered with apiAllowedIPs configuration option.
parameters:
- description: If returned list should contain all users or only users with
additional special roles
@@ -1259,73 +1383,46 @@ paths:
summary: Returns a list of users
tags:
- User
/jobs/tag_job/{id}:
delete:
post:
consumes:
- application/json
- multipart/form-data
description: |-
Removes tag(s) from a job specified by DB ID. Name and Type of Tag(s) must match.
Tag Scope is required for matching, options: "global", "admin". Private tags can not be deleted via API.
If tagged job is already finished: Tag will be removed from respective archive files.
User specified in form data will be saved to database.
Only accessible from IPs registered with apiAllowedIPs configuration option.
parameters:
- description: Job Database ID
in: path
name: id
- description: Unique user ID
in: formData
name: username
required: true
type: integer
- description: Array of tag-objects to remove
in: body
name: request
type: string
- description: User password
in: formData
name: password
required: true
schema:
items:
$ref: '#/definitions/api.ApiTag'
type: array
produces:
- application/json
responses:
"200":
description: Updated job resource
schema:
$ref: '#/definitions/schema.Job'
"400":
description: Bad Request
schema:
$ref: '#/definitions/api.ErrorResponse'
"401":
description: Unauthorized
schema:
$ref: '#/definitions/api.ErrorResponse'
"404":
description: Job or tag does not exist
schema:
$ref: '#/definitions/api.ErrorResponse'
"500":
description: Internal Server Error
schema:
$ref: '#/definitions/api.ErrorResponse'
security:
- ApiKeyAuth: []
summary: Removes one or more tags from a job
tags:
- Job add and modify
/tags/:
delete:
consumes:
- application/json
description: |-
Removes tags by type and name. Name and Type of Tag(s) must match.
Tag Scope is required for matching, options: "global", "admin". Private tags can not be deleted via API.
Tag wills be removed from respective archive files.
parameters:
- description: Array of tag-objects to remove
in: body
name: request
type: string
- description: User role
enum:
- admin
- support
- manager
- user
- api
in: formData
name: role
required: true
schema:
items:
$ref: '#/definitions/api.ApiTag'
type: array
type: string
- description: Managed project, required for new manager role user
in: formData
name: project
type: string
- description: Users name
in: formData
name: name
type: string
- description: Users email
in: formData
name: email
type: string
produces:
- text/plain
responses:
@@ -1336,24 +1433,28 @@ paths:
"400":
description: Bad Request
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
"401":
description: Unauthorized
schema:
$ref: '#/definitions/api.ErrorResponse'
"404":
description: Job or tag does not exist
type: string
"403":
description: Forbidden
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
"422":
description: 'Unprocessable Entity: creating user failed'
schema:
type: string
"500":
description: Internal Server Error
schema:
$ref: '#/definitions/api.ErrorResponse'
type: string
security:
- ApiKeyAuth: []
summary: Removes all tags and job-relations for type:name tuple
summary: Adds a new user
tags:
- Tag remove
- User
securityDefinitions:
ApiKeyAuth:
in: header

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package main
@@ -7,13 +7,12 @@ package main
import "flag"
var (
flagReinitDB, flagInit, flagServer, flagSyncLDAP, flagGops, flagMigrateDB, flagRevertDB,
flagForceDB, flagDev, flagVersion, flagLogDateTime, flagApplyTags bool
flagNewUser, flagDelUser, flagGenJWT, flagConfigFile, flagImportJob, flagLogLevel string
flagReinitDB, flagInit, flagServer, flagSyncLDAP, flagGops, flagMigrateDB, flagRevertDB, flagForceDB, flagDev, flagVersion, flagLogDateTime bool
flagNewUser, flagDelUser, flagGenJWT, flagConfigFile, flagImportJob, flagLogLevel string
)
func cliInit() {
flag.BoolVar(&flagInit, "init", false, "Setup var directory, initialize sqlite database file, config.json and .env")
flag.BoolVar(&flagInit, "init", false, "Setup var directory, initialize swlite database file, config.json and .env")
flag.BoolVar(&flagReinitDB, "init-db", false, "Go through job-archive and re-initialize the 'job', 'tag', and 'jobtag' tables (all running jobs will be lost!)")
flag.BoolVar(&flagSyncLDAP, "sync-ldap", false, "Sync the 'hpc_user' table with ldap")
flag.BoolVar(&flagServer, "server", false, "Start a server, continues listening on port after initialization and argument handling")
@@ -22,14 +21,13 @@ func cliInit() {
flag.BoolVar(&flagVersion, "version", false, "Show version information and exit")
flag.BoolVar(&flagMigrateDB, "migrate-db", false, "Migrate database to supported version and exit")
flag.BoolVar(&flagRevertDB, "revert-db", false, "Migrate database to previous version and exit")
flag.BoolVar(&flagApplyTags, "apply-tags", false, "Run taggers on all completed jobs and exit")
flag.BoolVar(&flagForceDB, "force-db", false, "Force database version, clear dirty flag and exit")
flag.BoolVar(&flagLogDateTime, "logdate", false, "Set this flag to add date and time to log messages")
flag.StringVar(&flagConfigFile, "config", "./config.json", "Specify alternative path to `config.json`")
flag.StringVar(&flagNewUser, "add-user", "", "Add a new user. Argument format: <username>:[admin,support,manager,api,user]:<password>")
flag.StringVar(&flagDelUser, "del-user", "", "Remove a existing user. Argument format: <username>")
flag.StringVar(&flagNewUser, "add-user", "", "Add a new user. Argument format: `<username>:[admin,support,manager,api,user]:<password>`")
flag.StringVar(&flagDelUser, "del-user", "", "Remove user by `username`")
flag.StringVar(&flagGenJWT, "jwt", "", "Generate and print a JWT for the user specified by its `username`")
flag.StringVar(&flagImportJob, "import-job", "", "Import a job. Argument format: `<path-to-meta.json>:<path-to-data.json>,...`")
flag.StringVar(&flagLogLevel, "loglevel", "warn", "Sets the logging level: `[debug, info (default), warn, err, crit]`")
flag.StringVar(&flagLogLevel, "loglevel", "warn", "Sets the logging level: `[debug,info,warn (default),err,fatal,crit]`")
flag.Parse()
}

View File

@@ -1,15 +1,16 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package main
import (
"fmt"
"os"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/util"
"github.com/ClusterCockpit/cc-backend/internal/util"
"github.com/ClusterCockpit/cc-backend/pkg/log"
)
const envString = `
@@ -32,18 +33,6 @@ const configString = `
"jwts": {
"max-age": "2000h"
},
"apiAllowedIPs": [
"*"
],
"enable-resampling": {
"trigger": 30,
"resolutions": [
600,
300,
120,
60
]
},
"clusters": [
{
"name": "name",
@@ -73,23 +62,24 @@ const configString = `
func initEnv() {
if util.CheckFileExists("var") {
cclog.Exit("Directory ./var already exists. Cautiously exiting application initialization.")
fmt.Print("Directory ./var already exists. Exiting!\n")
os.Exit(0)
}
if err := os.WriteFile("config.json", []byte(configString), 0o666); err != nil {
cclog.Abortf("Could not write default ./config.json with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
log.Fatalf("Writing config.json failed: %s", err.Error())
}
if err := os.WriteFile(".env", []byte(envString), 0o666); err != nil {
cclog.Abortf("Could not write default ./.env file with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
log.Fatalf("Writing .env failed: %s", err.Error())
}
if err := os.Mkdir("var", 0o777); err != nil {
cclog.Abortf("Could not create default ./var folder with permissions '0o777'. Application initialization failed, exited.\nError: %s\n", err.Error())
log.Fatalf("Mkdir var failed: %s", err.Error())
}
err := repository.MigrateDB("sqlite3", "./var/job.db")
if err != nil {
cclog.Abortf("Could not initialize default sqlite3 database as './var/job.db'. Application initialization failed, exited.\nError: %s\n", err.Error())
log.Fatalf("Initialize job.db failed: %s", err.Error())
}
}

View File

@@ -1,11 +1,10 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package main
import (
"encoding/json"
"fmt"
"os"
"os/signal"
@@ -18,20 +17,14 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/auth"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/importer"
"github.com/ClusterCockpit/cc-backend/internal/memorystore"
"github.com/ClusterCockpit/cc-backend/internal/metricdata"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/internal/tagger"
"github.com/ClusterCockpit/cc-backend/internal/taskManager"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
"github.com/ClusterCockpit/cc-backend/web"
ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/runtimeEnv"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-lib/util"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/runtimeEnv"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/google/gops/agent"
"github.com/joho/godotenv"
_ "github.com/go-sql-driver/mysql"
_ "github.com/mattn/go-sqlite3"
@@ -65,102 +58,87 @@ func main() {
os.Exit(0)
}
cclog.Init(flagLogLevel, flagLogDateTime)
// If init flag set, run tasks here before any file dependencies cause errors
if flagInit {
initEnv()
cclog.Exit("Successfully setup environment!\n" +
"Please review config.json and .env and adjust it to your needs.\n" +
"Add your job-archive at ./var/job-archive.")
}
// Apply config flags for pkg/log
log.Init(flagLogLevel, flagLogDateTime)
// See https://github.com/google/gops (Runtime overhead is almost zero)
if flagGops {
if err := agent.Listen(agent.Options{}); err != nil {
cclog.Abortf("Could not start gops agent with 'gops/agent.Listen(agent.Options{})'. Application startup failed, exited.\nError: %s\n", err.Error())
log.Fatalf("gops/agent.Listen failed: %s", err.Error())
}
}
err := godotenv.Load()
if err != nil {
cclog.Abortf("Could not parse existing .env file at location './.env'. Application startup failed, exited.\nError: %s\n", err.Error())
if err := runtimeEnv.LoadEnv("./.env"); err != nil && !os.IsNotExist(err) {
log.Fatalf("parsing './.env' file failed: %s", err.Error())
}
// Initialize sub-modules and handle command line flags.
// The order here is important!
ccconf.Init(flagConfigFile)
config.Init(flagConfigFile)
// Load and check main configuration
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
config.Init(cfg, clustercfg)
} else {
cclog.Abort("Cluster configuration must be present")
}
} else {
cclog.Abort("Main configuration must be present")
// As a special case for `db`, allow using an environment variable instead of the value
// stored in the config. This can be done for people having security concerns about storing
// the password for their mysql database in config.json.
if strings.HasPrefix(config.Keys.DB, "env:") {
envvar := strings.TrimPrefix(config.Keys.DB, "env:")
config.Keys.DB = os.Getenv(envvar)
}
if flagMigrateDB {
err := repository.MigrateDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
cclog.Abortf("MigrateDB Failed: Could not migrate '%s' database at location '%s' to version %d.\nError: %s\n", config.Keys.DBDriver, config.Keys.DB, repository.Version, err.Error())
log.Fatal(err)
}
cclog.Exitf("MigrateDB Success: Migrated '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, repository.Version)
os.Exit(0)
}
if flagRevertDB {
err := repository.RevertDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
cclog.Abortf("RevertDB Failed: Could not revert '%s' database at location '%s' to version %d.\nError: %s\n", config.Keys.DBDriver, config.Keys.DB, (repository.Version - 1), err.Error())
log.Fatal(err)
}
cclog.Exitf("RevertDB Success: Reverted '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, (repository.Version - 1))
os.Exit(0)
}
if flagForceDB {
err := repository.ForceDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
cclog.Abortf("ForceDB Failed: Could not force '%s' database at location '%s' to version %d.\nError: %s\n", config.Keys.DBDriver, config.Keys.DB, repository.Version, err.Error())
log.Fatal(err)
}
cclog.Exitf("ForceDB Success: Forced '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, repository.Version)
os.Exit(0)
}
repository.Connect(config.Keys.DBDriver, config.Keys.DB)
if flagInit {
initEnv()
fmt.Print("Successfully setup environment!\n")
fmt.Print("Please review config.json and .env and adjust it to your needs.\n")
fmt.Print("Add your job-archive at ./var/job-archive.\n")
os.Exit(0)
}
if !config.Keys.DisableAuthentication {
if cfg := ccconf.GetPackageConfig("auth"); cfg != nil {
auth.Init(&cfg)
} else {
cclog.Warn("Authentication disabled due to missing configuration")
auth.Init(nil)
}
auth.Init()
if flagNewUser != "" {
parts := strings.SplitN(flagNewUser, ":", 3)
if len(parts) != 3 || len(parts[0]) == 0 {
cclog.Abortf("Add User: Could not parse supplied argument format: No changes.\n"+
"Want: <username>:[admin,support,manager,api,user]:<password>\n"+
"Have: %s\n", flagNewUser)
log.Fatal("invalid argument format for user creation")
}
ur := repository.GetUserRepository()
if err := ur.AddUser(&schema.User{
Username: parts[0], Projects: make([]string, 0), Password: parts[2], Roles: strings.Split(parts[1], ","),
}); err != nil {
cclog.Abortf("Add User: Could not add new user authentication for '%s' and roles '%s'.\nError: %s\n", parts[0], parts[1], err.Error())
} else {
cclog.Printf("Add User: Added new user '%s' with roles '%s'.\n", parts[0], parts[1])
log.Fatalf("adding '%s' user authentication failed: %v", parts[0], err)
}
}
if flagDelUser != "" {
ur := repository.GetUserRepository()
if err := ur.DelUser(flagDelUser); err != nil {
cclog.Abortf("Delete User: Could not delete user '%s' from DB.\nError: %s\n", flagDelUser, err.Error())
} else {
cclog.Printf("Delete User: Deleted user '%s' from DB.\n", flagDelUser)
log.Fatalf("deleting user failed: %v", err)
}
}
@@ -168,101 +146,68 @@ func main() {
if flagSyncLDAP {
if authHandle.LdapAuth == nil {
cclog.Abort("Sync LDAP: LDAP authentication is not configured, could not synchronize. No changes, exited.")
log.Fatal("cannot sync: LDAP authentication is not configured")
}
if err := authHandle.LdapAuth.Sync(); err != nil {
cclog.Abortf("Sync LDAP: Could not synchronize, failed with error.\nError: %s\n", err.Error())
log.Fatalf("LDAP sync failed: %v", err)
}
cclog.Print("Sync LDAP: LDAP synchronization successfull.")
log.Info("LDAP sync successfull")
}
if flagGenJWT != "" {
ur := repository.GetUserRepository()
user, err := ur.GetUser(flagGenJWT)
if err != nil {
cclog.Abortf("JWT: Could not get supplied user '%s' from DB. No changes, exited.\nError: %s\n", flagGenJWT, err.Error())
log.Fatalf("could not get user from JWT: %v", err)
}
if !user.HasRole(schema.RoleApi) {
cclog.Warnf("JWT: User '%s' does not have the role 'api'. REST API endpoints will return error!\n", user.Username)
log.Warnf("user '%s' does not have the API role", user.Username)
}
jwt, err := authHandle.JwtAuth.ProvideJWT(user)
if err != nil {
cclog.Abortf("JWT: User '%s' found in DB, but failed to provide JWT.\nError: %s\n", user.Username, err.Error())
log.Fatalf("failed to provide JWT to user '%s': %v", user.Username, err)
}
cclog.Printf("JWT: Successfully generated JWT for user '%s': %s\n", user.Username, jwt)
fmt.Printf("MAIN > JWT for '%s': %s\n", user.Username, jwt)
}
} else if flagNewUser != "" || flagDelUser != "" {
cclog.Abort("Error: Arguments '--add-user' and '--del-user' can only be used if authentication is enabled. No changes, exited.")
log.Fatal("arguments --add-user and --del-user can only be used if authentication is enabled")
}
if archiveCfg := ccconf.GetPackageConfig("archive"); archiveCfg != nil {
err = archive.Init(archiveCfg, config.Keys.DisableArchive)
} else {
err = archive.Init(json.RawMessage("{\"kind\":\"file\",\"path\":\"./var/job-archive\"}"), config.Keys.DisableArchive)
}
if err != nil {
cclog.Abortf("Init: Failed to initialize archive.\nError: %s\n", err.Error())
if err := archive.Init(config.Keys.Archive, config.Keys.DisableArchive); err != nil {
log.Fatalf("failed to initialize archive: %s", err.Error())
}
if err := metricdata.Init(); err != nil {
cclog.Abortf("Init: Failed to initialize metricdata repository.\nError %s\n", err.Error())
log.Fatalf("failed to initialize metricdata repository: %s", err.Error())
}
if flagReinitDB {
if err := importer.InitDB(); err != nil {
cclog.Abortf("Init DB: Failed to re-initialize repository DB.\nError: %s\n", err.Error())
} else {
cclog.Print("Init DB: Sucessfully re-initialized repository DB.")
log.Fatalf("failed to re-initialize repository DB: %s", err.Error())
}
}
if flagImportJob != "" {
if err := importer.HandleImportFlag(flagImportJob); err != nil {
cclog.Abortf("Import Job: Job import failed.\nError: %s\n", err.Error())
} else {
cclog.Printf("Import Job: Imported Job '%s' into DB.\n", flagImportJob)
}
}
if config.Keys.EnableJobTaggers {
tagger.Init()
}
if flagApplyTags {
if err := tagger.RunTaggers(); err != nil {
cclog.Abortf("Running job taggers.\nError: %s\n", err.Error())
log.Fatalf("job import failed: %s", err.Error())
}
}
if !flagServer {
cclog.Exit("No errors, server flag not set. Exiting cc-backend.")
return
}
archiver.Start(repository.GetJobRepository())
taskManager.Start()
serverInit()
var wg sync.WaitGroup
// Metric Store starts after all flags have been processes
if memorystore.InternalCCMSFlag {
if mscfg := ccconf.GetPackageConfig("metric-store"); mscfg != nil {
memorystore.Init(mscfg, &wg)
} else {
cclog.Abort("Metric Store configuration must be present")
}
}
archiver.Start(repository.GetJobRepository())
taskManager.Start(ccconf.GetPackageConfig("cron"),
ccconf.GetPackageConfig("archive"))
cfg := ccconf.GetPackageConfig("ui")
web.Init(cfg)
serverInit()
wg.Add(1)
go func() {
defer wg.Done()
@@ -279,8 +224,6 @@ func main() {
serverShutdown()
util.FsWatcherShutdown()
taskManager.Shutdown()
}()
@@ -289,5 +232,5 @@ func main() {
}
runtimeEnv.SystemdNotifiy(true, "running")
wg.Wait()
cclog.Print("Graceful shutdown completed!")
log.Print("Graceful shutdown completed!")
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package main
@@ -18,7 +18,6 @@ import (
"time"
"github.com/99designs/gqlgen/graphql/handler"
"github.com/99designs/gqlgen/graphql/handler/transport"
"github.com/99designs/gqlgen/graphql/playground"
"github.com/ClusterCockpit/cc-backend/internal/api"
"github.com/ClusterCockpit/cc-backend/internal/archiver"
@@ -26,11 +25,10 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/graph"
"github.com/ClusterCockpit/cc-backend/internal/graph/generated"
"github.com/ClusterCockpit/cc-backend/internal/memorystore"
"github.com/ClusterCockpit/cc-backend/internal/routerConfig"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/runtimeEnv"
"github.com/ClusterCockpit/cc-backend/web"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/runtimeEnv"
"github.com/gorilla/handlers"
"github.com/gorilla/mux"
httpSwagger "github.com/swaggo/http-swagger"
@@ -55,29 +53,18 @@ func serverInit() {
// Setup the http.Handler/Router used by the server
graph.Init()
resolver := graph.GetResolverInstance()
graphQLServer := handler.New(
graphQLEndpoint := handler.NewDefaultServer(
generated.NewExecutableSchema(generated.Config{Resolvers: resolver}))
// graphQLServer.AddTransport(transport.SSE{})
graphQLServer.AddTransport(transport.POST{})
// graphQLServer.AddTransport(transport.Websocket{
// KeepAlivePingInterval: 10 * time.Second,
// Upgrader: websocket.Upgrader{
// CheckOrigin: func(r *http.Request) bool {
// return true
// },
// },
// })
if os.Getenv("DEBUG") != "1" {
// Having this handler means that a error message is returned via GraphQL instead of the connection simply beeing closed.
// The problem with this is that then, no more stacktrace is printed to stderr.
graphQLServer.SetRecoverFunc(func(ctx context.Context, err any) error {
graphQLEndpoint.SetRecoverFunc(func(ctx context.Context, err interface{}) error {
switch e := err.(type) {
case string:
return fmt.Errorf("MAIN > Panic: %s", e)
case error:
return fmt.Errorf("MAIN > Panic caused by: %s", e.Error())
return fmt.Errorf("MAIN > Panic caused by: %w", e)
}
return errors.New("MAIN > Internal server error (panic)")
@@ -91,10 +78,10 @@ func serverInit() {
router = mux.NewRouter()
buildInfo := web.Build{Version: version, Hash: commit, Buildtime: date}
info := map[string]any{}
info := map[string]interface{}{}
info["hasOpenIDConnect"] = false
if auth.Keys.OpenIDConfig != nil {
if config.Keys.OpenIDConfig != nil {
openIDConnect := auth.NewOIDC(authHandle)
openIDConnect.RegisterEndpoints(router)
info["hasOpenIDConnect"] = true
@@ -102,7 +89,7 @@ func serverInit() {
router.HandleFunc("/login", func(rw http.ResponseWriter, r *http.Request) {
rw.Header().Add("Content-Type", "text/html; charset=utf-8")
cclog.Debugf("##%v##", info)
log.Debugf("##%v##", info)
web.RenderTemplate(rw, "login.tmpl", &web.Page{Title: "Login", Build: buildInfo, Infos: info})
}).Methods(http.MethodGet)
router.HandleFunc("/imprint", func(rw http.ResponseWriter, r *http.Request) {
@@ -119,7 +106,6 @@ func serverInit() {
userapi := router.PathPrefix("/userapi").Subrouter()
configapi := router.PathPrefix("/config").Subrouter()
frontendapi := router.PathPrefix("/frontend").Subrouter()
metricstoreapi := router.PathPrefix("/metricstore").Subrouter()
if !config.Keys.DisableAuthentication {
router.Handle("/login", authHandle.Login(
@@ -185,7 +171,7 @@ func serverInit() {
})
securedapi.Use(func(next http.Handler) http.Handler {
return authHandle.AuthAPI(
return authHandle.AuthApi(
// On success;
next,
// On failure: JSON Response
@@ -193,15 +179,7 @@ func serverInit() {
})
userapi.Use(func(next http.Handler) http.Handler {
return authHandle.AuthUserAPI(
// On success;
next,
// On failure: JSON Response
onFailureResponse)
})
metricstoreapi.Use(func(next http.Handler) http.Handler {
return authHandle.AuthMetricStoreAPI(
return authHandle.AuthUserApi(
// On success;
next,
// On failure: JSON Response
@@ -209,7 +187,7 @@ func serverInit() {
})
configapi.Use(func(next http.Handler) http.Handler {
return authHandle.AuthConfigAPI(
return authHandle.AuthConfigApi(
// On success;
next,
// On failure: JSON Response
@@ -217,7 +195,7 @@ func serverInit() {
})
frontendapi.Use(func(next http.Handler) http.Handler {
return authHandle.AuthFrontendAPI(
return authHandle.AuthFrontendApi(
// On success;
next,
// On failure: JSON Response
@@ -230,7 +208,7 @@ func serverInit() {
router.PathPrefix("/swagger/").Handler(httpSwagger.Handler(
httpSwagger.URL("http://" + config.Keys.Addr + "/swagger/doc.json"))).Methods(http.MethodGet)
}
secured.Handle("/query", graphQLServer)
secured.Handle("/query", graphQLEndpoint)
// Send a searchId and then reply with a redirect to a user, or directly send query to job table for jobid and project.
secured.HandleFunc("/search", func(rw http.ResponseWriter, r *http.Request) {
@@ -244,18 +222,14 @@ func serverInit() {
apiHandle.MountConfigApiRoutes(configapi)
apiHandle.MountFrontendApiRoutes(frontendapi)
if memorystore.InternalCCMSFlag {
apiHandle.MountMetricStoreApiRoutes(metricstoreapi)
}
if config.Keys.EmbedStaticFiles {
if i, err := os.Stat("./var/img"); err == nil {
if i.IsDir() {
cclog.Info("Use local directory for static images")
log.Info("Use local directory for static images")
router.PathPrefix("/img/").Handler(http.StripPrefix("/img/", http.FileServer(http.Dir("./var/img"))))
}
}
router.PathPrefix("/").Handler(http.StripPrefix("/", web.ServeFiles()))
router.PathPrefix("/").Handler(web.ServeFiles())
} else {
router.PathPrefix("/").Handler(http.FileServer(http.Dir(config.Keys.StaticFiles)))
}
@@ -267,46 +241,17 @@ func serverInit() {
handlers.AllowedHeaders([]string{"X-Requested-With", "Content-Type", "Authorization", "Origin"}),
handlers.AllowedMethods([]string{"GET", "POST", "HEAD", "OPTIONS"}),
handlers.AllowedOrigins([]string{"*"})))
// secured.NotFoundHandler = http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
// page := web.Page{
// Title: "ClusterCockpit - Not Found",
// Build: buildInfo,
// }
// rw.Header().Add("Content-Type", "text/html; charset=utf-8")
// web.RenderTemplate(rw, "404.tmpl", &page)
// })
// secured.NotFoundHandler = http.HandlerFunc(http.NotFound)
// router.NotFoundHandler = router.NewRoute().HandlerFunc(http.NotFound).GetHandler()
// printEndpoints(router)
}
// func printEndpoints(r *mux.Router) {
// r.Walk(func(route *mux.Route, router *mux.Router, ancestors []*mux.Route) error {
// path, err := route.GetPathTemplate()
// if err != nil {
// path = "nopath"
// }
// methods, err := route.GetMethods()
// if err != nil {
// methods = append(methods, "nomethod")
// }
// fmt.Printf("%v %s\n", methods, path)
// return nil
// })
// }
func serverStart() {
handler := handlers.CustomLoggingHandler(io.Discard, router, func(_ io.Writer, params handlers.LogFormatterParams) {
if strings.HasPrefix(params.Request.RequestURI, "/api/") {
cclog.Debugf("%s %s (%d, %.02fkb, %dms)",
log.Debugf("%s %s (%d, %.02fkb, %dms)",
params.Request.Method, params.URL.RequestURI(),
params.StatusCode, float32(params.Size)/1024,
time.Since(params.TimeStamp).Milliseconds())
} else {
cclog.Debugf("%s %s (%d, %.02fkb, %dms)",
log.Debugf("%s %s (%d, %.02fkb, %dms)",
params.Request.Method, params.URL.RequestURI(),
params.StatusCode, float32(params.Size)/1024,
time.Since(params.TimeStamp).Milliseconds())
@@ -323,20 +268,20 @@ func serverStart() {
// Start http or https server
listener, err := net.Listen("tcp", config.Keys.Addr)
if err != nil {
cclog.Abortf("Server Start: Starting http listener on '%s' failed.\nError: %s\n", config.Keys.Addr, err.Error())
log.Fatalf("starting http listener failed: %v", err)
}
if !strings.HasSuffix(config.Keys.Addr, ":80") && config.Keys.RedirectHTTPTo != "" {
if !strings.HasSuffix(config.Keys.Addr, ":80") && config.Keys.RedirectHttpTo != "" {
go func() {
http.ListenAndServe(":80", http.RedirectHandler(config.Keys.RedirectHTTPTo, http.StatusMovedPermanently))
http.ListenAndServe(":80", http.RedirectHandler(config.Keys.RedirectHttpTo, http.StatusMovedPermanently))
}()
}
if config.Keys.HTTPSCertFile != "" && config.Keys.HTTPSKeyFile != "" {
if config.Keys.HttpsCertFile != "" && config.Keys.HttpsKeyFile != "" {
cert, err := tls.LoadX509KeyPair(
config.Keys.HTTPSCertFile, config.Keys.HTTPSKeyFile)
config.Keys.HttpsCertFile, config.Keys.HttpsKeyFile)
if err != nil {
cclog.Abortf("Server Start: Loading X509 keypair failed. Check options 'https-cert-file' and 'https-key-file' in 'config.json'.\nError: %s\n", err.Error())
log.Fatalf("loading X509 keypair failed: %v", err)
}
listener = tls.NewListener(listener, &tls.Config{
Certificates: []tls.Certificate{cert},
@@ -347,20 +292,20 @@ func serverStart() {
MinVersion: tls.VersionTLS12,
PreferServerCipherSuites: true,
})
cclog.Printf("HTTPS server listening at %s...\n", config.Keys.Addr)
fmt.Printf("HTTPS server listening at %s...", config.Keys.Addr)
} else {
cclog.Printf("HTTP server listening at %s...\n", config.Keys.Addr)
fmt.Printf("HTTP server listening at %s...", config.Keys.Addr)
}
//
// Because this program will want to bind to a privileged port (like 80), the listener must
// be established first, then the user can be changed, and after that,
// the actual http server can be started.
if err := runtimeEnv.DropPrivileges(config.Keys.Group, config.Keys.User); err != nil {
cclog.Abortf("Server Start: Error while preparing server start.\nError: %s\n", err.Error())
log.Fatalf("error while preparing server start: %s", err.Error())
}
if err = server.Serve(listener); err != nil && err != http.ErrServerClosed {
cclog.Abortf("Server Start: Starting server failed.\nError: %s\n", err.Error())
log.Fatalf("starting server failed: %v", err)
}
}
@@ -368,11 +313,6 @@ func serverShutdown() {
// First shut down the server gracefully (waiting for all ongoing requests)
server.Shutdown(context.Background())
// Archive all the metric store data
if memorystore.InternalCCMSFlag {
memorystore.Shutdown()
}
// Then, wait for any async archivings still pending...
archiver.WaitForArchiving()
}

View File

@@ -1,42 +1,30 @@
{
"main": {
"addr": "127.0.0.1:8080",
"short-running-jobs-duration": 300,
"resampling": {
"trigger": 30,
"resolutions": [
600,
300,
120,
60
]
},
"apiAllowedIPs": [
"*"
],
"emission-constant": 317
},
"cron": {
"commit-job-worker": "2m",
"duration-worker": "5m",
"footprint-worker": "10m"
},
"addr": "127.0.0.1:8080",
"short-running-jobs-duration": 300,
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"auth": {
"jwts": {
"max-age": "2000h"
}
"jwts": {
"max-age": "2000h"
},
"enable-resampling": {
"trigger": 30,
"resolutions": [
600,
300,
120,
60
]
},
"emission-constant": 317,
"clusters": [
{
"name": "fritz",
"metricDataRepository": {
"kind": "cc-metric-store-internal",
"kind": "cc-metric-store",
"url": "http://localhost:8082",
"token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJFZERTQSJ9.eyJ1c2VyIjoiYWRtaW4iLCJyb2xlcyI6WyJST0xFX0FETUlOIiwiUk9MRV9BTkFMWVNUIiwiUk9MRV9VU0VSIl19.d-3_3FZTsadPjDEdsWrrQ7nS0edMAR4zjl-eK7rJU3HziNBfI9PDHDIpJVHTNN5E5SlLGLFXctWyKAkwhXL-Dw"
"token": ""
},
"filterRanges": {
"numNodes": {
@@ -56,9 +44,9 @@
{
"name": "alex",
"metricDataRepository": {
"kind": "cc-metric-store-internal",
"kind": "cc-metric-store",
"url": "http://localhost:8082",
"token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJFZERTQSJ9.eyJ1c2VyIjoiYWRtaW4iLCJyb2xlcyI6WyJST0xFX0FETUlOIiwiUk9MRV9BTkFMWVNUIiwiUk9MRV9VU0VSIl19.d-3_3FZTsadPjDEdsWrrQ7nS0edMAR4zjl-eK7rJU3HziNBfI9PDHDIpJVHTNN5E5SlLGLFXctWyKAkwhXL-Dw"
"token": ""
},
"filterRanges": {
"numNodes": {
@@ -75,19 +63,5 @@
}
}
}
],
"metric-store": {
"checkpoints": {
"file-format": "avro",
"interval": "1h",
"directory": "./var/checkpoints",
"restore": "48h"
},
"archive": {
"interval": "1h",
"directory": "./var/archive"
},
"retention-in-memory": "48h"
},
"ui-file": "./configs/uiConfig.json"
}
]
}

View File

@@ -12,7 +12,12 @@
"db": "clustercockpit:demo@tcp(127.0.0.1:3306)/clustercockpit",
"enable-resampling": {
"trigger": 30,
"resolutions": [600, 300, 120, 60]
"resolutions": [
600,
300,
120,
60
]
},
"emission-constant": 317,
"clusters": [

View File

@@ -1,50 +1,50 @@
{
"main": {
"addr": "0.0.0.0:443",
"ldap": {
"url": "ldaps://test",
"user_base": "ou=people,ou=hpc,dc=test,dc=de",
"search_dn": "cn=hpcmonitoring,ou=roadm,ou=profile,ou=hpc,dc=test,dc=de",
"user_bind": "uid={username},ou=people,ou=hpc,dc=test,dc=de",
"user_filter": "(&(objectclass=posixAccount))"
},
"https-cert-file": "/etc/letsencrypt/live/url/fullchain.pem",
"https-key-file": "/etc/letsencrypt/live/url/privkey.pem",
"user": "clustercockpit",
"group": "clustercockpit",
"validate": false,
"apiAllowedIPs": ["*"],
"short-running-jobs-duration": 300,
"resampling": {
"trigger": 30,
"resolutions": [600, 300, 120, 60]
}
},
"cron": {
"commit-job-worker": "2m",
"duration-worker": "5m",
"footprint-worker": "10m"
},
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"clusters": [
{
"name": "test",
"metricDataRepository": {
"kind": "cc-metric-store",
"url": "http://localhost:8082",
"token": "eyJhbGciOiJF-E-pQBQ"
},
"filterRanges": {
"numNodes": {
"from": 1,
"to": 64
},
"duration": {
"from": 0,
"to": 86400
},
"startTime": {
"from": "2022-01-01T00:00:00Z",
"to": null
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"validate": true,
"clusters": [
{
"name": "test",
"metricDataRepository": {
"kind": "cc-metric-store",
"url": "http://localhost:8082",
"token": "eyJhbGciOiJF-E-pQBQ"
},
"filterRanges": {
"numNodes": {
"from": 1,
"to": 64
},
"duration": {
"from": 0,
"to": 86400
},
"startTime": {
"from": "2022-01-01T00:00:00Z",
"to": null
}
}
}
}
}
]
],
"jwts": {
"cookieName": "",
"validateUser": false,
"max-age": "2000h",
"trustedIssuer": ""
},
"short-running-jobs-duration": 300
}

View File

@@ -0,0 +1,12 @@
{
"clusters": [
{
"name": "fritz",
"default_metrics": "cpu_load, flops_any, core_power, lustre_open, mem_used, mem_bw, net_bytes_in"
},
{
"name": "alex",
"default_metrics": "flops_any, mem_bw, mem_used, vectorization_ratio"
}
]
}

View File

@@ -1,45 +0,0 @@
{
"jobList": {
"usePaging": false,
"showFootprint":false
},
"jobView": {
"showPolarPlot": true,
"showFootprint": true,
"showRoofline": true,
"showStatTable": true
},
"metricConfig": {
"jobListMetrics": ["mem_bw", "flops_dp"],
"jobViewPlotMetrics": ["mem_bw", "flops_dp"],
"jobViewTableMetrics": ["mem_bw", "flops_dp"],
"clusters": [
{
"name": "test",
"subClusters": [
{
"name": "one",
"jobListMetrics": ["mem_used", "flops_sp"]
}
]
}
]
},
"nodeList": {
"usePaging": true
},
"plotConfiguration": {
"plotsPerRow": 3,
"colorBackground": true,
"lineWidth": 3,
"colorScheme": [
"#00bfff",
"#0000ff",
"#ff00ff",
"#ff0000",
"#ff8000",
"#ffff00",
"#80ff00"
]
}
}

109
go.mod
View File

@@ -1,40 +1,36 @@
module github.com/ClusterCockpit/cc-backend
go 1.24.0
toolchain go1.24.1
go 1.23.5
require (
github.com/99designs/gqlgen v0.17.81
github.com/ClusterCockpit/cc-lib v0.10.1
github.com/99designs/gqlgen v0.17.63
github.com/ClusterCockpit/cc-lib v0.0.0-20250216162506-b737e48423c5
github.com/ClusterCockpit/cc-units v0.4.0
github.com/Masterminds/squirrel v1.5.4
github.com/coreos/go-oidc/v3 v3.16.0
github.com/expr-lang/expr v1.17.6
github.com/go-co-op/gocron/v2 v2.16.0
github.com/go-ldap/ldap/v3 v3.4.10
github.com/go-sql-driver/mysql v1.9.0
github.com/golang-jwt/jwt/v5 v5.2.2
github.com/golang-migrate/migrate/v4 v4.18.2
github.com/coreos/go-oidc/v3 v3.11.0
github.com/go-co-op/gocron/v2 v2.9.0
github.com/go-ldap/ldap/v3 v3.4.8
github.com/go-sql-driver/mysql v1.8.1
github.com/golang-jwt/jwt/v5 v5.2.1
github.com/golang-migrate/migrate/v4 v4.17.1
github.com/google/gops v0.3.28
github.com/gorilla/handlers v1.5.2
github.com/gorilla/mux v1.8.1
github.com/gorilla/sessions v1.4.0
github.com/influxdata/line-protocol/v2 v2.2.1
github.com/influxdata/influxdb-client-go/v2 v2.14.0
github.com/jmoiron/sqlx v1.4.0
github.com/joho/godotenv v1.5.1
github.com/linkedin/goavro/v2 v2.14.0
github.com/mattn/go-sqlite3 v1.14.24
github.com/nats-io/nats.go v1.47.0
github.com/prometheus/client_golang v1.23.2
github.com/prometheus/common v0.66.1
github.com/mattn/go-sqlite3 v1.14.22
github.com/prometheus/client_golang v1.20.5
github.com/prometheus/common v0.55.0
github.com/qustavo/sqlhooks/v2 v2.1.0
github.com/santhosh-tekuri/jsonschema/v5 v5.3.1
github.com/swaggo/http-swagger v1.3.4
github.com/swaggo/swag v1.16.6
github.com/vektah/gqlparser/v2 v2.5.30
golang.org/x/crypto v0.43.0
golang.org/x/oauth2 v0.32.0
golang.org/x/time v0.13.0
github.com/swaggo/swag v1.16.4
github.com/vektah/gqlparser/v2 v2.5.22
golang.org/x/crypto v0.33.0
golang.org/x/exp v0.0.0-20250215185904-eff6e970281f
golang.org/x/oauth2 v0.21.0
golang.org/x/time v0.10.0
)
require (
@@ -42,62 +38,61 @@ require (
github.com/Azure/go-ntlmssp v0.0.0-20221128193559-754e69321358 // indirect
github.com/KyleBanks/depth v1.2.1 // indirect
github.com/agnivade/levenshtein v1.2.1 // indirect
github.com/apapsch/go-jsonmerge/v2 v2.0.0 // indirect
github.com/beorn7/perks v1.0.1 // indirect
github.com/cespare/xxhash/v2 v2.3.0 // indirect
github.com/cpuguy83/go-md2man/v2 v2.0.7 // indirect
github.com/cpuguy83/go-md2man/v2 v2.0.6 // indirect
github.com/expr-lang/expr v1.16.9 // indirect
github.com/felixge/httpsnoop v1.0.4 // indirect
github.com/fsnotify/fsnotify v1.9.0 // indirect
github.com/go-asn1-ber/asn1-ber v1.5.7 // indirect
github.com/go-jose/go-jose/v4 v4.1.3 // indirect
github.com/go-openapi/jsonpointer v0.22.1 // indirect
github.com/go-openapi/jsonreference v0.21.2 // indirect
github.com/go-openapi/spec v0.22.0 // indirect
github.com/go-openapi/swag/conv v0.25.1 // indirect
github.com/go-openapi/swag/jsonname v0.25.1 // indirect
github.com/go-openapi/swag/jsonutils v0.25.1 // indirect
github.com/go-openapi/swag/loading v0.25.1 // indirect
github.com/go-openapi/swag/stringutils v0.25.1 // indirect
github.com/go-openapi/swag/typeutils v0.25.1 // indirect
github.com/go-openapi/swag/yamlutils v0.25.1 // indirect
github.com/go-viper/mapstructure/v2 v2.4.0 // indirect
github.com/golang/snappy v0.0.4 // indirect
github.com/go-jose/go-jose/v4 v4.0.3 // indirect
github.com/go-openapi/jsonpointer v0.21.0 // indirect
github.com/go-openapi/jsonreference v0.21.0 // indirect
github.com/go-openapi/spec v0.21.0 // indirect
github.com/go-openapi/swag v0.23.0 // indirect
github.com/go-viper/mapstructure/v2 v2.2.1 // indirect
github.com/google/uuid v1.6.0 // indirect
github.com/gorilla/securecookie v1.1.2 // indirect
github.com/gorilla/websocket v1.5.3 // indirect
github.com/hashicorp/errwrap v1.1.0 // indirect
github.com/hashicorp/go-multierror v1.1.1 // indirect
github.com/hashicorp/golang-lru/v2 v2.0.7 // indirect
github.com/jonboulle/clockwork v0.5.0 // indirect
github.com/influxdata/line-protocol v0.0.0-20210922203350-b1ad95c89adf // indirect
github.com/influxdata/line-protocol/v2 v2.2.1 // indirect
github.com/jonboulle/clockwork v0.4.0 // indirect
github.com/josharian/intern v1.0.0 // indirect
github.com/jpillora/backoff v1.0.0 // indirect
github.com/json-iterator/go v1.1.12 // indirect
github.com/klauspost/compress v1.18.0 // indirect
github.com/klauspost/compress v1.17.9 // indirect
github.com/lann/builder v0.0.0-20180802200727-47ae307949d0 // indirect
github.com/lann/ps v0.0.0-20150810152359-62de8c46ede0 // indirect
github.com/mailru/easyjson v0.9.0 // indirect
github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
github.com/modern-go/reflect2 v1.0.2 // indirect
github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f // indirect
github.com/nats-io/nkeys v0.4.11 // indirect
github.com/nats-io/nats.go v1.39.0 // indirect
github.com/nats-io/nkeys v0.4.9 // indirect
github.com/nats-io/nuid v1.0.1 // indirect
github.com/prometheus/client_model v0.6.2 // indirect
github.com/prometheus/procfs v0.16.1 // indirect
github.com/oapi-codegen/runtime v1.1.1 // indirect
github.com/opencontainers/image-spec v1.1.0-rc2.0.20221005185240-3a7f492d3f1b // indirect
github.com/prometheus/client_model v0.6.1 // indirect
github.com/prometheus/procfs v0.15.1 // indirect
github.com/robfig/cron/v3 v3.0.1 // indirect
github.com/russross/blackfriday/v2 v2.1.0 // indirect
github.com/sosodev/duration v1.3.1 // indirect
github.com/swaggo/files v1.0.1 // indirect
github.com/urfave/cli/v2 v2.27.7 // indirect
github.com/xrash/smetrics v0.0.0-20250705151800-55b8f293f342 // indirect
github.com/urfave/cli/v2 v2.27.5 // indirect
github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1 // indirect
go.uber.org/atomic v1.11.0 // indirect
go.yaml.in/yaml/v2 v2.4.3 // indirect
go.yaml.in/yaml/v3 v3.0.4 // indirect
golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b // indirect
golang.org/x/mod v0.29.0 // indirect
golang.org/x/net v0.46.0 // indirect
golang.org/x/sync v0.17.0 // indirect
golang.org/x/sys v0.37.0 // indirect
golang.org/x/text v0.30.0 // indirect
golang.org/x/tools v0.38.0 // indirect
google.golang.org/protobuf v1.36.9 // indirect
golang.org/x/mod v0.23.0 // indirect
golang.org/x/net v0.35.0 // indirect
golang.org/x/sync v0.11.0 // indirect
golang.org/x/sys v0.30.0 // indirect
golang.org/x/text v0.22.0 // indirect
golang.org/x/tools v0.30.0 // indirect
google.golang.org/protobuf v1.36.1 // indirect
gopkg.in/yaml.v2 v2.4.0 // indirect
gopkg.in/yaml.v3 v3.0.1 // indirect
sigs.k8s.io/yaml v1.6.0 // indirect
sigs.k8s.io/yaml v1.4.0 // indirect
)

295
go.sum
View File

@@ -1,124 +1,102 @@
filippo.io/edwards25519 v1.1.0 h1:FNf4tywRC1HmFuKW5xopWpigGjJKiJSV0Cqo0cJWDaA=
filippo.io/edwards25519 v1.1.0/go.mod h1:BxyFTGdWcka3PhytdK4V28tE5sGfRvvvRV7EaN4VDT4=
github.com/99designs/gqlgen v0.17.81 h1:kCkN/xVyRb5rEQpuwOHRTYq83i0IuTQg9vdIiwEerTs=
github.com/99designs/gqlgen v0.17.81/go.mod h1:vgNcZlLwemsUhYim4dC1pvFP5FX0pr2Y+uYUoHFb1ig=
github.com/99designs/gqlgen v0.17.63 h1:HCdaYDPd9HqUXRchEvmE3EFzELRwLlaJ8DBuyC8Cqto=
github.com/99designs/gqlgen v0.17.63/go.mod h1:sVCM2iwIZisJjTI/DEC3fpH+HFgxY1496ZJ+jbT9IjA=
github.com/Azure/go-ansiterm v0.0.0-20230124172434-306776ec8161 h1:L/gRVlceqvL25UVaW/CKtUDjefjrs0SPonmDGUVOYP0=
github.com/Azure/go-ansiterm v0.0.0-20230124172434-306776ec8161/go.mod h1:xomTg63KZ2rFqZQzSB4Vz2SUXa1BpHTVz9L5PTmPC4E=
github.com/Azure/go-ntlmssp v0.0.0-20221128193559-754e69321358 h1:mFRzDkZVAjdal+s7s0MwaRv9igoPqLRdzOLzw/8Xvq8=
github.com/Azure/go-ntlmssp v0.0.0-20221128193559-754e69321358/go.mod h1:chxPXzSsl7ZWRAuOIE23GDNzjWuZquvFlgA8xmpunjU=
github.com/ClusterCockpit/cc-lib v0.10.1 h1:tjGEH8mFGgznYxO8BKLiiar0eZR1Oytk8x5iIQHZR5s=
github.com/ClusterCockpit/cc-lib v0.10.1/go.mod h1:nvTZuxFCTwlos8I1rL5O1RPab7vRtkU8E/PGiaF6pQA=
github.com/ClusterCockpit/cc-lib v0.0.0-20250216162506-b737e48423c5 h1:Suu9ivW5rz1dXAqHvA4V9dHI0jm4aIGzfpQGFvGzGEc=
github.com/ClusterCockpit/cc-lib v0.0.0-20250216162506-b737e48423c5/go.mod h1:T2iDLmsbGjyMy1UtDkR+hofoQXfTMuwSdeFettTav9Q=
github.com/ClusterCockpit/cc-units v0.4.0 h1:zP5DOu99GmErW0tCDf0gcLrlWt42RQ9dpoONEOh4cI0=
github.com/ClusterCockpit/cc-units v0.4.0/go.mod h1:3S3PAhAayS3pbgcT4q9Vn9VJw22Op51X0YimtG77zBw=
github.com/KyleBanks/depth v1.2.1 h1:5h8fQADFrWtarTdtDudMmGsC7GPbOAu6RVB3ffsVFHc=
github.com/KyleBanks/depth v1.2.1/go.mod h1:jzSb9d0L43HxTQfT+oSA1EEp2q+ne2uh6XgeJcm8brE=
github.com/Masterminds/squirrel v1.5.4 h1:uUcX/aBc8O7Fg9kaISIUsHXdKuqehiXAMQTYX8afzqM=
github.com/Masterminds/squirrel v1.5.4/go.mod h1:NNaOrjSoIDfDA40n7sr2tPNZRfjzjA400rg+riTZj10=
github.com/Microsoft/go-winio v0.6.2 h1:F2VQgta7ecxGYO8k3ZZz3RS8fVIXVxONVUPlNERoyfY=
github.com/Microsoft/go-winio v0.6.2/go.mod h1:yd8OoFMLzJbo9gZq8j5qaps8bJ9aShtEA8Ipt1oGCvU=
github.com/NVIDIA/go-nvml v0.13.0-1 h1:OLX8Jq3dONuPOQPC7rndB6+iDmDakw0XTYgzMxObkEw=
github.com/NVIDIA/go-nvml v0.13.0-1/go.mod h1:+KNA7c7gIBH7SKSJ1ntlwkfN80zdx8ovl4hrK3LmPt4=
github.com/PuerkitoBio/goquery v1.10.3 h1:pFYcNSqHxBD06Fpj/KsbStFRsgRATgnf3LeXiUkhzPo=
github.com/PuerkitoBio/goquery v1.10.3/go.mod h1:tMUX0zDMHXYlAQk6p35XxQMqMweEKB7iK7iLNd4RH4Y=
github.com/Microsoft/go-winio v0.6.1 h1:9/kr64B9VUZrLm5YYwbGtUJnMgqWVOdUAXu6Migciow=
github.com/Microsoft/go-winio v0.6.1/go.mod h1:LRdKpFKfdobln8UmuiYcKPot9D2v6svN5+sAH+4kjUM=
github.com/PuerkitoBio/goquery v1.9.3 h1:mpJr/ikUA9/GNJB/DBZcGeFDXUtosHRyRrwh7KGdTG0=
github.com/PuerkitoBio/goquery v1.9.3/go.mod h1:1ndLHPdTz+DyQPICCWYlYQMPl0oXZj0G6D4LCYA6u4U=
github.com/RaveNoX/go-jsoncommentstrip v1.0.0/go.mod h1:78ihd09MekBnJnxpICcwzCMzGrKSKYe4AqU6PDYYpjk=
github.com/agnivade/levenshtein v1.2.1 h1:EHBY3UOn1gwdy/VbFwgo4cxecRznFk7fKWN1KOX7eoM=
github.com/agnivade/levenshtein v1.2.1/go.mod h1:QVVI16kDrtSuwcpd0p1+xMC6Z/VfhtCyDIjcwga4/DU=
github.com/alexbrainman/sspi v0.0.0-20231016080023-1a75b4708caa h1:LHTHcTQiSGT7VVbI0o4wBRNQIgn917usHWOd6VAffYI=
github.com/alexbrainman/sspi v0.0.0-20231016080023-1a75b4708caa/go.mod h1:cEWa1LVoE5KvSD9ONXsZrj0z6KqySlCCNKHlLzbqAt4=
github.com/andreyvit/diff v0.0.0-20170406064948-c7f18ee00883 h1:bvNMNQO63//z+xNgfBlViaCIJKLlCJ6/fmUseuG0wVQ=
github.com/andreyvit/diff v0.0.0-20170406064948-c7f18ee00883/go.mod h1:rCTlJbsFo29Kk6CurOXKm700vrz8f0KW0JNfpkRJY/8=
github.com/andybalholm/cascadia v1.3.3 h1:AG2YHrzJIm4BZ19iwJ/DAua6Btl3IwJX+VI4kktS1LM=
github.com/andybalholm/cascadia v1.3.3/go.mod h1:xNd9bqTn98Ln4DwST8/nG+H0yuB8Hmgu1YHNnWw0GeA=
github.com/andybalholm/cascadia v1.3.2 h1:3Xi6Dw5lHF15JtdcmAHD3i1+T8plmv7BQ/nsViSLyss=
github.com/andybalholm/cascadia v1.3.2/go.mod h1:7gtRlve5FxPPgIgX36uWBX58OdBsSS6lUvCFb+h7KvU=
github.com/apapsch/go-jsonmerge/v2 v2.0.0 h1:axGnT1gRIfimI7gJifB699GoE/oq+F2MU7Dml6nw9rQ=
github.com/apapsch/go-jsonmerge/v2 v2.0.0/go.mod h1:lvDnEdqiQrp0O42VQGgmlKpxL1AP2+08jFMw88y4klk=
github.com/arbovm/levenshtein v0.0.0-20160628152529-48b4e1c0c4d0 h1:jfIu9sQUG6Ig+0+Ap1h4unLjW6YQJpKZVmUzxsD4E/Q=
github.com/arbovm/levenshtein v0.0.0-20160628152529-48b4e1c0c4d0/go.mod h1:t2tdKJDJF9BV14lnkjHmOQgcvEKgtqs5a1N3LNdJhGE=
github.com/beorn7/perks v1.0.1 h1:VlbKKnNfV8bJzeqoa4cOKqO6bYr3WgKZxO8Z16+hsOM=
github.com/beorn7/perks v1.0.1/go.mod h1:G2ZrVWU2WbWT9wwq4/hrbKbnv/1ERSJQ0ibhJ6rlkpw=
github.com/bmatcuk/doublestar v1.1.1/go.mod h1:UD6OnuiIn0yFxxA2le/rnRU1G4RaI4UvFv1sNto9p6w=
github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs=
github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
github.com/coreos/go-oidc/v3 v3.16.0 h1:qRQUCFstKpXwmEjDQTIbyY/5jF00+asXzSkmkoa/mow=
github.com/coreos/go-oidc/v3 v3.16.0/go.mod h1:wqPbKFrVnE90vty060SB40FCJ8fTHTxSwyXJqZH+sI8=
github.com/cpuguy83/go-md2man/v2 v2.0.7 h1:zbFlGlXEAKlwXpmvle3d8Oe3YnkKIK4xSRTd3sHPnBo=
github.com/cpuguy83/go-md2man/v2 v2.0.7/go.mod h1:oOW0eioCTA6cOiMLiUPZOpcVxMig6NIQQ7OS05n1F4g=
github.com/coreos/go-oidc/v3 v3.11.0 h1:Ia3MxdwpSw702YW0xgfmP1GVCMA9aEFWu12XUZ3/OtI=
github.com/coreos/go-oidc/v3 v3.11.0/go.mod h1:gE3LgjOgFoHi9a4ce4/tJczr0Ai2/BoDhf0r5lltWI0=
github.com/cpuguy83/go-md2man/v2 v2.0.6 h1:XJtiaUW6dEEqVuZiMTn1ldk455QWwEIsMIJlo5vtkx0=
github.com/cpuguy83/go-md2man/v2 v2.0.6/go.mod h1:oOW0eioCTA6cOiMLiUPZOpcVxMig6NIQQ7OS05n1F4g=
github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/dgryski/trifles v0.0.0-20230903005119-f50d829f2e54 h1:SG7nF6SRlWhcT7cNTs5R6Hk4V2lcmLz2NsG2VnInyNo=
github.com/dgryski/trifles v0.0.0-20230903005119-f50d829f2e54/go.mod h1:if7Fbed8SFyPtHLHbg49SI7NAdJiC5WIA09pe59rfAA=
github.com/dhui/dktest v0.4.4 h1:+I4s6JRE1yGuqflzwqG+aIaMdgXIorCf5P98JnaAWa8=
github.com/dhui/dktest v0.4.4/go.mod h1:4+22R4lgsdAXrDyaH4Nqx2JEz2hLp49MqQmm9HLCQhM=
github.com/distribution/reference v0.6.0 h1:0IXCQ5g4/QMHHkarYzh5l+u8T3t73zM5QvfrDyIgxBk=
github.com/distribution/reference v0.6.0/go.mod h1:BbU0aIcezP1/5jX/8MP0YiH4SdvB5Y4f/wlDRiLyi3E=
github.com/docker/docker v27.2.0+incompatible h1:Rk9nIVdfH3+Vz4cyI/uhbINhEZ/oLmc+CBXmH6fbNk4=
github.com/docker/docker v27.2.0+incompatible/go.mod h1:eEKB0N0r5NX/I1kEveEz05bcu8tLC/8azJZsviup8Sk=
github.com/docker/go-connections v0.5.0 h1:USnMq7hx7gwdVZq1L49hLXaFtUdTADjXGp+uj1Br63c=
github.com/docker/go-connections v0.5.0/go.mod h1:ov60Kzw0kKElRwhNs9UlUHAE/F9Fe6GLaXnqyDdmEXc=
github.com/dhui/dktest v0.4.1 h1:/w+IWuDXVymg3IrRJCHHOkMK10m9aNVMOyD0X12YVTg=
github.com/dhui/dktest v0.4.1/go.mod h1:DdOqcUpL7vgyP4GlF3X3w7HbSlz8cEQzwewPveYEQbA=
github.com/docker/distribution v2.8.2+incompatible h1:T3de5rq0dB1j30rp0sA2rER+m322EBzniBPB6ZIzuh8=
github.com/docker/distribution v2.8.2+incompatible/go.mod h1:J2gT2udsDAN96Uj4KfcMRqY0/ypR+oyYUYmja8H+y+w=
github.com/docker/docker v24.0.9+incompatible h1:HPGzNmwfLZWdxHqK9/II92pyi1EpYKsAqcl4G0Of9v0=
github.com/docker/docker v24.0.9+incompatible/go.mod h1:eEKB0N0r5NX/I1kEveEz05bcu8tLC/8azJZsviup8Sk=
github.com/docker/go-connections v0.4.0 h1:El9xVISelRB7BuFusrZozjnkIM5YnzCViNKohAFqRJQ=
github.com/docker/go-connections v0.4.0/go.mod h1:Gbd7IOopHjR8Iph03tsViu4nIes5XhDvyHbTtUxmeec=
github.com/docker/go-units v0.5.0 h1:69rxXcBk27SvSaaxTtLh/8llcHD8vYHT7WSdRZ/jvr4=
github.com/docker/go-units v0.5.0/go.mod h1:fgPhTUdO+D/Jk86RDLlptpiXQzgHJF7gydDDbaIK4Dk=
github.com/expr-lang/expr v1.17.6 h1:1h6i8ONk9cexhDmowO/A64VPxHScu7qfSl2k8OlINec=
github.com/expr-lang/expr v1.17.6/go.mod h1:8/vRC7+7HBzESEqt5kKpYXxrxkr31SaO8r40VO/1IT4=
github.com/expr-lang/expr v1.16.9 h1:WUAzmR0JNI9JCiF0/ewwHB1gmcGw5wW7nWt8gc6PpCI=
github.com/expr-lang/expr v1.16.9/go.mod h1:8/vRC7+7HBzESEqt5kKpYXxrxkr31SaO8r40VO/1IT4=
github.com/felixge/httpsnoop v1.0.4 h1:NFTV2Zj1bL4mc9sqWACXbQFVBBg2W3GPvqp8/ESS2Wg=
github.com/felixge/httpsnoop v1.0.4/go.mod h1:m8KPJKqk1gH5J9DgRY2ASl2lWCfGKXixSwevea8zH2U=
github.com/frankban/quicktest v1.11.0/go.mod h1:K+q6oSqb0W0Ininfk863uOk1lMy69l/P6txr3mVT54s=
github.com/frankban/quicktest v1.11.2/go.mod h1:K+q6oSqb0W0Ininfk863uOk1lMy69l/P6txr3mVT54s=
github.com/frankban/quicktest v1.13.0 h1:yNZif1OkDfNoDfb9zZa9aXIpejNR4F23Wely0c+Qdqk=
github.com/frankban/quicktest v1.13.0/go.mod h1:qLE0fzW0VuyUAJgPU19zByoIr0HtCHN/r/VLSOOIySU=
github.com/fsnotify/fsnotify v1.9.0 h1:2Ml+OJNzbYCTzsxtv8vKSFD9PbJjmhYF14k/jKC7S9k=
github.com/fsnotify/fsnotify v1.9.0/go.mod h1:8jBTzvmWwFyi3Pb8djgCCO5IBqzKJ/Jwo8TRcHyHii0=
github.com/go-asn1-ber/asn1-ber v1.5.5/go.mod h1:hEBeB/ic+5LoWskz+yKT7vGhhPYkProFKoKdwZRWMe0=
github.com/go-asn1-ber/asn1-ber v1.5.7 h1:DTX+lbVTWaTw1hQ+PbZPlnDZPEIs0SS/GCZAl535dDk=
github.com/go-asn1-ber/asn1-ber v1.5.7/go.mod h1:hEBeB/ic+5LoWskz+yKT7vGhhPYkProFKoKdwZRWMe0=
github.com/go-co-op/gocron/v2 v2.16.0 h1:uqUF6WFZ4enRU45pWFNcn1xpDLc+jBOTKhPQI16Z1xs=
github.com/go-co-op/gocron/v2 v2.16.0/go.mod h1:opexeOFy5BplhsKdA7bzY9zeYih8I8/WNJ4arTIFPVc=
github.com/go-jose/go-jose/v4 v4.1.3 h1:CVLmWDhDVRa6Mi/IgCgaopNosCaHz7zrMeF9MlZRkrs=
github.com/go-jose/go-jose/v4 v4.1.3/go.mod h1:x4oUasVrzR7071A4TnHLGSPpNOm2a21K9Kf04k1rs08=
github.com/go-ldap/ldap/v3 v3.4.10 h1:ot/iwPOhfpNVgB1o+AVXljizWZ9JTp7YF5oeyONmcJU=
github.com/go-ldap/ldap/v3 v3.4.10/go.mod h1:JXh4Uxgi40P6E9rdsYqpUtbW46D9UTjJ9QSwGRznplY=
github.com/go-logr/logr v1.4.2 h1:6pFjapn8bFcIbiKo3XT4j/BhANplGihG6tvd+8rYgrY=
github.com/go-logr/logr v1.4.2/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
github.com/go-openapi/jsonpointer v0.22.1 h1:sHYI1He3b9NqJ4wXLoJDKmUmHkWy/L7rtEo92JUxBNk=
github.com/go-openapi/jsonpointer v0.22.1/go.mod h1:pQT9OsLkfz1yWoMgYFy4x3U5GY5nUlsOn1qSBH5MkCM=
github.com/go-openapi/jsonreference v0.21.2 h1:Wxjda4M/BBQllegefXrY/9aq1fxBA8sI5M/lFU6tSWU=
github.com/go-openapi/jsonreference v0.21.2/go.mod h1:pp3PEjIsJ9CZDGCNOyXIQxsNuroxm8FAJ/+quA0yKzQ=
github.com/go-openapi/spec v0.22.0 h1:xT/EsX4frL3U09QviRIZXvkh80yibxQmtoEvyqug0Tw=
github.com/go-openapi/spec v0.22.0/go.mod h1:K0FhKxkez8YNS94XzF8YKEMULbFrRw4m15i2YUht4L0=
github.com/go-openapi/swag v0.19.15 h1:D2NRCBzS9/pEY3gP9Nl8aDqGUcPFrwG2p+CNFrLyrCM=
github.com/go-openapi/swag/conv v0.25.1 h1:+9o8YUg6QuqqBM5X6rYL/p1dpWeZRhoIt9x7CCP+he0=
github.com/go-openapi/swag/conv v0.25.1/go.mod h1:Z1mFEGPfyIKPu0806khI3zF+/EUXde+fdeksUl2NiDs=
github.com/go-openapi/swag/jsonname v0.25.1 h1:Sgx+qbwa4ej6AomWC6pEfXrA6uP2RkaNjA9BR8a1RJU=
github.com/go-openapi/swag/jsonname v0.25.1/go.mod h1:71Tekow6UOLBD3wS7XhdT98g5J5GR13NOTQ9/6Q11Zo=
github.com/go-openapi/swag/jsonutils v0.25.1 h1:AihLHaD0brrkJoMqEZOBNzTLnk81Kg9cWr+SPtxtgl8=
github.com/go-openapi/swag/jsonutils v0.25.1/go.mod h1:JpEkAjxQXpiaHmRO04N1zE4qbUEg3b7Udll7AMGTNOo=
github.com/go-openapi/swag/jsonutils/fixtures_test v0.25.1 h1:DSQGcdB6G0N9c/KhtpYc71PzzGEIc/fZ1no35x4/XBY=
github.com/go-openapi/swag/jsonutils/fixtures_test v0.25.1/go.mod h1:kjmweouyPwRUEYMSrbAidoLMGeJ5p6zdHi9BgZiqmsg=
github.com/go-openapi/swag/loading v0.25.1 h1:6OruqzjWoJyanZOim58iG2vj934TysYVptyaoXS24kw=
github.com/go-openapi/swag/loading v0.25.1/go.mod h1:xoIe2EG32NOYYbqxvXgPzne989bWvSNoWoyQVWEZicc=
github.com/go-openapi/swag/stringutils v0.25.1 h1:Xasqgjvk30eUe8VKdmyzKtjkVjeiXx1Iz0zDfMNpPbw=
github.com/go-openapi/swag/stringutils v0.25.1/go.mod h1:JLdSAq5169HaiDUbTvArA2yQxmgn4D6h4A+4HqVvAYg=
github.com/go-openapi/swag/typeutils v0.25.1 h1:rD/9HsEQieewNt6/k+JBwkxuAHktFtH3I3ysiFZqukA=
github.com/go-openapi/swag/typeutils v0.25.1/go.mod h1:9McMC/oCdS4BKwk2shEB7x17P6HmMmA6dQRtAkSnNb8=
github.com/go-openapi/swag/yamlutils v0.25.1 h1:mry5ez8joJwzvMbaTGLhw8pXUnhDK91oSJLDPF1bmGk=
github.com/go-openapi/swag/yamlutils v0.25.1/go.mod h1:cm9ywbzncy3y6uPm/97ysW8+wZ09qsks+9RS8fLWKqg=
github.com/go-co-op/gocron/v2 v2.9.0 h1:+0nTyI3mjc2FGIClBdDWpaLPCNrJ+62o9xbS0ZklEKQ=
github.com/go-co-op/gocron/v2 v2.9.0/go.mod h1:xY7bJxGazKam1cz04EebrlP4S9q4iWdiAylMGP3jY9w=
github.com/go-jose/go-jose/v4 v4.0.3 h1:o8aphO8Hv6RPmH+GfzVuyf7YXSBibp+8YyHdOoDESGo=
github.com/go-jose/go-jose/v4 v4.0.3/go.mod h1:NKb5HO1EZccyMpiZNbdUw/14tiXNyUJh188dfnMCAfc=
github.com/go-ldap/ldap/v3 v3.4.8 h1:loKJyspcRezt2Q3ZRMq2p/0v8iOurlmeXDPw6fikSvQ=
github.com/go-ldap/ldap/v3 v3.4.8/go.mod h1:qS3Sjlu76eHfHGpUdWkAXQTw4beih+cHsco2jXlIXrk=
github.com/go-openapi/jsonpointer v0.21.0 h1:YgdVicSA9vH5RiHs9TZW5oyafXZFc6+2Vc1rr/O9oNQ=
github.com/go-openapi/jsonpointer v0.21.0/go.mod h1:IUyH9l/+uyhIYQ/PXVA41Rexl+kOkAPDdXEYns6fzUY=
github.com/go-openapi/jsonreference v0.21.0 h1:Rs+Y7hSXT83Jacb7kFyjn4ijOuVGSvOdF2+tg1TRrwQ=
github.com/go-openapi/jsonreference v0.21.0/go.mod h1:LmZmgsrTkVg9LG4EaHeY8cBDslNPMo06cago5JNLkm4=
github.com/go-openapi/spec v0.21.0 h1:LTVzPc3p/RzRnkQqLRndbAzjY0d0BCL72A6j3CdL9ZY=
github.com/go-openapi/spec v0.21.0/go.mod h1:78u6VdPw81XU44qEWGhtr982gJ5BWg2c0I5XwVMotYk=
github.com/go-openapi/swag v0.23.0 h1:vsEVJDUo2hPJ2tu0/Xc+4noaxyEffXNIs3cOULZ+GrE=
github.com/go-openapi/swag v0.23.0/go.mod h1:esZ8ITTYEsH1V2trKHjAN8Ai7xHb8RV+YSZ577vPjgQ=
github.com/go-sql-driver/mysql v1.4.1/go.mod h1:zAC/RDZ24gD3HViQzih4MyKcchzm+sOG5ZlKdlhCg5w=
github.com/go-sql-driver/mysql v1.8.1 h1:LedoTUt/eveggdHS9qUFC1EFSa8bU2+1pZjSRpvNJ1Y=
github.com/go-sql-driver/mysql v1.8.1/go.mod h1:wEBSXgmK//2ZFJyE+qWnIsVGmvmEKlqwuVSjsCm7DZg=
github.com/go-sql-driver/mysql v1.9.0 h1:Y0zIbQXhQKmQgTp44Y1dp3wTXcn804QoTptLZT1vtvo=
github.com/go-sql-driver/mysql v1.9.0/go.mod h1:pDetrLJeA3oMujJuvXc8RJoasr589B6A9fwzD3QMrqw=
github.com/go-viper/mapstructure/v2 v2.4.0 h1:EBsztssimR/CONLSZZ04E8qAkxNYq4Qp9LvH92wZUgs=
github.com/go-viper/mapstructure/v2 v2.4.0/go.mod h1:oJDH3BJKyqBA2TXFhDsKDGDTlndYOZ6rGS0BRZIxGhM=
github.com/go-viper/mapstructure/v2 v2.2.1 h1:ZAaOCxANMuZx5RCeg0mBdEZk7DZasvvZIxtHqx8aGss=
github.com/go-viper/mapstructure/v2 v2.2.1/go.mod h1:oJDH3BJKyqBA2TXFhDsKDGDTlndYOZ6rGS0BRZIxGhM=
github.com/gogo/protobuf v1.3.2 h1:Ov1cvc58UF3b5XjBnZv7+opcTcQFZebYjWzi34vdm4Q=
github.com/gogo/protobuf v1.3.2/go.mod h1:P1XiOD3dCwIKUDQYPy72D8LYyHL2YPYrpS2s69NZV8Q=
github.com/golang-jwt/jwt/v5 v5.2.2 h1:Rl4B7itRWVtYIHFrSNd7vhTiz9UpLdi6gZhZ3wEeDy8=
github.com/golang-jwt/jwt/v5 v5.2.2/go.mod h1:pqrtFR0X4osieyHYxtmOUWsAWrfe1Q5UVIyoH402zdk=
github.com/golang-migrate/migrate/v4 v4.18.2 h1:2VSCMz7x7mjyTXx3m2zPokOY82LTRgxK1yQYKo6wWQ8=
github.com/golang-migrate/migrate/v4 v4.18.2/go.mod h1:2CM6tJvn2kqPXwnXO/d3rAQYiyoIm180VsO8PRX6Rpk=
github.com/golang/snappy v0.0.1/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
github.com/golang/snappy v0.0.4 h1:yAGX7huGHXlcLOEtBnF4w7FQwA26wojNCwOYAEhLjQM=
github.com/golang/snappy v0.0.4/go.mod h1:/XxbfmMg8lxefKM7IXC3fBNl/7bRcc72aCRzEWrmP2Q=
github.com/golang-jwt/jwt/v5 v5.2.1 h1:OuVbFODueb089Lh128TAcimifWaLhJwVflnrgM17wHk=
github.com/golang-jwt/jwt/v5 v5.2.1/go.mod h1:pqrtFR0X4osieyHYxtmOUWsAWrfe1Q5UVIyoH402zdk=
github.com/golang-migrate/migrate/v4 v4.17.1 h1:4zQ6iqL6t6AiItphxJctQb3cFqWiSpMnX7wLTPnnYO4=
github.com/golang-migrate/migrate/v4 v4.17.1/go.mod h1:m8hinFyWBn0SA4QKHuKh175Pm9wjmxj3S2Mia7dbXzM=
github.com/google/go-cmp v0.5.2/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
github.com/google/go-cmp v0.5.5/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
github.com/google/go-cmp v0.5.9/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8=
github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU=
github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
github.com/google/gofuzz v1.2.0 h1:xRy4A+RhZaiKjJ1bPfwQ8sedCA+YS2YcCHW6ec7JMi0=
github.com/google/gofuzz v1.2.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
@@ -173,16 +151,17 @@ github.com/jcmturner/rpc/v2 v2.0.3 h1:7FXXj8Ti1IaVFpSAziCZWNzbNuZmnvw/i6CqLNdWfZ
github.com/jcmturner/rpc/v2 v2.0.3/go.mod h1:VUJYCIDm3PVOEHw8sgt091/20OJjskO/YJki3ELg/Hc=
github.com/jmoiron/sqlx v1.4.0 h1:1PLqN7S1UYp5t4SrVVnt4nUVNemrDAtxlulVe+Qgm3o=
github.com/jmoiron/sqlx v1.4.0/go.mod h1:ZrZ7UsYB/weZdl2Bxg6jCRO9c3YHl8r3ahlKmRT4JLY=
github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4=
github.com/jonboulle/clockwork v0.5.0 h1:Hyh9A8u51kptdkR+cqRpT1EebBwTn1oK9YfGYbdFz6I=
github.com/jonboulle/clockwork v0.5.0/go.mod h1:3mZlmanh0g2NDKO5TWZVJAfofYk64M7XN3SzBPjZF60=
github.com/jonboulle/clockwork v0.4.0 h1:p4Cf1aMWXnXAUh8lVfewRBx1zaTSYKrKMF2g3ST4RZ4=
github.com/jonboulle/clockwork v0.4.0/go.mod h1:xgRqUGwRcjKCO1vbZUEtSLrqKoPSsUpK7fnezOII0kc=
github.com/josharian/intern v1.0.0 h1:vlS4z54oSdjm0bgjRigI+G1HpF+tI+9rE5LLzOg8HmY=
github.com/josharian/intern v1.0.0/go.mod h1:5DoeVV0s6jJacbCEi61lwdGj/aVlrQvzHFFd8Hwg//Y=
github.com/jpillora/backoff v1.0.0 h1:uvFg412JmmHBHw7iwprIxkPMI+sGQ4kzOWsMeHnm2EA=
github.com/jpillora/backoff v1.0.0/go.mod h1:J/6gKK9jxlEcS3zixgDgUAsiuZ7yrSoa/FX5e0EB2j4=
github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM=
github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo=
github.com/klauspost/compress v1.18.0 h1:c/Cqfb0r+Yi+JtIEq73FWXVkRonBlf0CRNYc8Zttxdo=
github.com/klauspost/compress v1.18.0/go.mod h1:2Pp+KzxcywXVXMr50+X0Q/Lsb43OQHYWRCY2AiWywWQ=
github.com/juju/gnuflag v0.0.0-20171113085948-2ce1bb71843d/go.mod h1:2PavIy+JPciBPrBUjwbNvtwB6RQlve+hkpll6QSNmOE=
github.com/klauspost/compress v1.17.9 h1:6KIumPrER1LHsvBVuDa0r5xaG0Es51mhhB9BQB2qeMA=
github.com/klauspost/compress v1.17.9/go.mod h1:Di0epgTjJY877eYKx5yC51cX2A2Vl2ibi7bDH9ttBbw=
github.com/kr/pretty v0.2.1/go.mod h1:ipq/a2n7PKx3OHsz4KJII5eveXtPO4qwEXGdVfWzfnI=
github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
@@ -190,6 +169,8 @@ github.com/kr/pty v1.1.1/go.mod h1:pFQYn66WHrOpPYNljwOMqo10TkYh1fy3cYio2l3bCsQ=
github.com/kr/text v0.1.0/go.mod h1:4Jbv+DJW3UT/LiOwJeYQe1efqtUx/iVham/4vfdArNI=
github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
github.com/kylelemons/godebug v1.1.0 h1:RPNrshWIDI6G2gRW9EHilWtl7Z6Sb1BR0xunSBf0SNc=
github.com/kylelemons/godebug v1.1.0/go.mod h1:9/0rRGxNHcop5bhtWyNeEfOS8JIWk580+fNqagV/RAw=
github.com/lann/builder v0.0.0-20180802200727-47ae307949d0 h1:SOEGU9fKiNWd/HOJuq6+3iTQz8KNCLtVX6idSoTLdUw=
github.com/lann/builder v0.0.0-20180802200727-47ae307949d0/go.mod h1:dXGbAdH5GtBTC4WfIxhKZfyBF/HBFgRZSWwZ9g/He9o=
github.com/lann/ps v0.0.0-20150810152359-62de8c46ede0 h1:P6pPBnrTSX3DEVR4fDembhRWSsG5rVo6hYhAB/ADZrk=
@@ -197,14 +178,11 @@ github.com/lann/ps v0.0.0-20150810152359-62de8c46ede0/go.mod h1:vmVJ0l/dxyfGW6Fm
github.com/lib/pq v1.2.0/go.mod h1:5WUZQaWbwv1U+lTReE5YruASi9Al49XbQIvNi/34Woo=
github.com/lib/pq v1.10.9 h1:YXG7RB+JIjhP29X+OtkiDnYaXQwpS4JEWq7dtCCRUEw=
github.com/lib/pq v1.10.9/go.mod h1:AlVN5x4E4T544tWzH6hKfbfQvm3HdbOxrmggDNAPY9o=
github.com/linkedin/goavro/v2 v2.14.0 h1:aNO/js65U+Mwq4yB5f1h01c3wiM458qtRad1DN0CMUI=
github.com/linkedin/goavro/v2 v2.14.0/go.mod h1:KXx+erlq+RPlGSPmLF7xGo6SAbh8sCQ53x064+ioxhk=
github.com/mailru/easyjson v0.9.0 h1:PrnmzHw7262yW8sTBwxi1PdJA3Iw/EKBa8psRf7d9a4=
github.com/mailru/easyjson v0.9.0/go.mod h1:1+xMtQp2MRNVL/V1bOzuP3aP8VNwRW55fQUto+XFtTU=
github.com/mattn/go-sqlite3 v1.10.0/go.mod h1:FPy6KqzDD04eiIsT53CuJW3U88zkxoIYsOqkbpncsNc=
github.com/mattn/go-sqlite3 v1.14.22 h1:2gZY6PC6kBnID23Tichd1K+Z0oS6nE/XwU+Vz/5o4kU=
github.com/mattn/go-sqlite3 v1.14.22/go.mod h1:Uh1q+B4BYcTPb+yiD3kU8Ct7aC0hY9fxUwlHK0RXw+Y=
github.com/mattn/go-sqlite3 v1.14.24 h1:tpSp2G2KyMnnQu99ngJ47EIkWVmliIizyZBfPrBWDRM=
github.com/mattn/go-sqlite3 v1.14.24/go.mod h1:Uh1q+B4BYcTPb+yiD3kU8Ct7aC0hY9fxUwlHK0RXw+Y=
github.com/moby/docker-image-spec v1.3.1 h1:jMKff3w6PgbfSa69GfNg+zN/XLhfXJGnEx3Nl2EsFP0=
github.com/moby/docker-image-spec v1.3.1/go.mod h1:eKmb5VW8vQEh/BAr2yvVNvuiJuY6UIocYsFu/DxxRpo=
github.com/moby/term v0.5.0 h1:xt8Q1nalod/v7BqbG21f8mQPqH+xAaC9C3N3wfWbVP0=
github.com/moby/term v0.5.0/go.mod h1:8FzsFHVUBGZdbDsJw/ot+X+d5HLUbvklYLJ9uGfcI3Y=
github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
@@ -218,10 +196,10 @@ github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 h1:C3w9PqII01/Oq
github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822/go.mod h1:+n7T8mK8HuQTcFwEeznm/DIxMOiR9yIdICNftLE1DvQ=
github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f h1:KUppIJq7/+SVif2QVs3tOP0zanoHgBEVAwHxUSIzRqU=
github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f/go.mod h1:qRWi+5nqEBWmkhHvq77mSJWrCKwh8bxhgT7d/eI7P4U=
github.com/nats-io/nats.go v1.47.0 h1:YQdADw6J/UfGUd2Oy6tn4Hq6YHxCaJrVKayxxFqYrgM=
github.com/nats-io/nats.go v1.47.0/go.mod h1:iRWIPokVIFbVijxuMQq4y9ttaBTMe0SFdlZfMDd+33g=
github.com/nats-io/nkeys v0.4.11 h1:q44qGV008kYd9W1b1nEBkNzvnWxtRSQ7A8BoqRrcfa0=
github.com/nats-io/nkeys v0.4.11/go.mod h1:szDimtgmfOi9n25JpfIdGw12tZFYXqhGxjhVxsatHVE=
github.com/nats-io/nats.go v1.39.0 h1:2/yg2JQjiYYKLwDuBzV0FbB2sIV+eFNkEevlRi4n9lI=
github.com/nats-io/nats.go v1.39.0/go.mod h1:MgRb8oOdigA6cYpEPhXJuRVH6UE/V4jblJ2jQ27IXYM=
github.com/nats-io/nkeys v0.4.9 h1:qe9Faq2Gxwi6RZnZMXfmGMZkg3afLLOtrU+gDZJ35b0=
github.com/nats-io/nkeys v0.4.9/go.mod h1:jcMqs+FLG+W5YO36OX6wFIFcmpdAns+w1Wm6D3I/evE=
github.com/nats-io/nuid v1.0.1 h1:5iA8DT8V7q8WK2EScv2padNa/rTESc1KdnPw4TC2paw=
github.com/nats-io/nuid v1.0.1/go.mod h1:19wcPz3Ph3q0Jbyiqsd0kePYG7A95tJPxeL+1OSON2c=
github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e/go.mod h1:zD1mROLANZcx1PVRCS0qkT7pwLkGfwJo4zjcN/Tysno=
@@ -229,27 +207,27 @@ github.com/oapi-codegen/runtime v1.1.1 h1:EXLHh0DXIJnWhdRPN2w4MXAzFyE4CskzhNLUmt
github.com/oapi-codegen/runtime v1.1.1/go.mod h1:SK9X900oXmPWilYR5/WKPzt3Kqxn/uS/+lbpREv+eCg=
github.com/opencontainers/go-digest v1.0.0 h1:apOUWs51W5PlhuyGyz9FCeeBIOUDA/6nW8Oi/yOhh5U=
github.com/opencontainers/go-digest v1.0.0/go.mod h1:0JzlMkj0TRzQZfJkVvzbP0HBR3IKzErnv2BNG4W4MAM=
github.com/opencontainers/image-spec v1.1.0 h1:8SG7/vwALn54lVB/0yZ/MMwhFrPYtpEHQb2IpWsCzug=
github.com/opencontainers/image-spec v1.1.0/go.mod h1:W4s4sFTMaBeK1BQLXbG4AdM2szdn85PY75RI83NrTrM=
github.com/opencontainers/image-spec v1.1.0-rc2.0.20221005185240-3a7f492d3f1b h1:YWuSjZCQAPM8UUBLkYUk1e+rZcvWHJmFb6i6rM44Xs8=
github.com/opencontainers/image-spec v1.1.0-rc2.0.20221005185240-3a7f492d3f1b/go.mod h1:3OVijpioIKYWTqjiG0zfF6wvoJ4fAXGbjdZuI2NgsRQ=
github.com/opentracing/opentracing-go v1.1.0/go.mod h1:UkNAQd3GIcIGf0SeVgPpRdFStlNbqXla1AfSYxPUl2o=
github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
github.com/prometheus/client_golang v1.23.2 h1:Je96obch5RDVy3FDMndoUsjAhG5Edi49h0RJWRi/o0o=
github.com/prometheus/client_golang v1.23.2/go.mod h1:Tb1a6LWHB3/SPIzCoaDXI4I8UHKeFTEQ1YCr+0Gyqmg=
github.com/prometheus/client_model v0.6.2 h1:oBsgwpGs7iVziMvrGhE53c/GrLUsZdHnqNwqPLxwZyk=
github.com/prometheus/client_model v0.6.2/go.mod h1:y3m2F6Gdpfy6Ut/GBsUqTWZqCUvMVzSfMLjcu6wAwpE=
github.com/prometheus/common v0.66.1 h1:h5E0h5/Y8niHc5DlaLlWLArTQI7tMrsfQjHV+d9ZoGs=
github.com/prometheus/common v0.66.1/go.mod h1:gcaUsgf3KfRSwHY4dIMXLPV0K/Wg1oZ8+SbZk/HH/dA=
github.com/prometheus/procfs v0.16.1 h1:hZ15bTNuirocR6u0JZ6BAHHmwS1p8B4P6MRqxtzMyRg=
github.com/prometheus/procfs v0.16.1/go.mod h1:teAbpZRB1iIAJYREa1LsoWUXykVXA1KlTmWl8x/U+Is=
github.com/prometheus/client_golang v1.20.5 h1:cxppBPuYhUnsO6yo/aoRol4L7q7UFfdm+bR9r+8l63Y=
github.com/prometheus/client_golang v1.20.5/go.mod h1:PIEt8X02hGcP8JWbeHyeZ53Y/jReSnHgO035n//V5WE=
github.com/prometheus/client_model v0.6.1 h1:ZKSh/rekM+n3CeS952MLRAdFwIKqeY8b62p8ais2e9E=
github.com/prometheus/client_model v0.6.1/go.mod h1:OrxVMOVHjw3lKMa8+x6HeMGkHMQyHDk9E3jmP2AmGiY=
github.com/prometheus/common v0.55.0 h1:KEi6DK7lXW/m7Ig5i47x0vRzuBsHuvJdi5ee6Y3G1dc=
github.com/prometheus/common v0.55.0/go.mod h1:2SECS4xJG1kd8XF9IcM1gMX6510RAEL65zxzNImwdc8=
github.com/prometheus/procfs v0.15.1 h1:YagwOFzUgYfKKHX6Dr+sHT7km/hxC76UB0learggepc=
github.com/prometheus/procfs v0.15.1/go.mod h1:fB45yRUv8NstnjriLhBQLuOUt+WW4BsoGhij/e3PBqk=
github.com/qustavo/sqlhooks/v2 v2.1.0 h1:54yBemHnGHp/7xgT+pxwmIlMSDNYKx5JW5dfRAiCZi0=
github.com/qustavo/sqlhooks/v2 v2.1.0/go.mod h1:aMREyKo7fOKTwiLuWPsaHRXEmtqG4yREztO0idF83AU=
github.com/robfig/cron/v3 v3.0.1 h1:WdRxkvbJztn8LMz/QEvLN5sBU+xKpSqwwUO1Pjr4qDs=
github.com/robfig/cron/v3 v3.0.1/go.mod h1:eQICP3HwyT7UooqI/z+Ov+PtYAWygg1TEWWzGIFLtro=
github.com/rogpeppe/go-internal v1.12.0 h1:exVL4IDcn6na9z1rAb56Vxr+CgyK3nn3O+epU5NdKM8=
github.com/rogpeppe/go-internal v1.12.0/go.mod h1:E+RYuTGaKKdloAfM02xzb0FW3Paa99yedzYV+kq4uf4=
github.com/rogpeppe/go-internal v1.11.0 h1:cWPaGQEPrBb5/AsnsZesgZZ9yb1OQ+GOISoDNXVBh4M=
github.com/rogpeppe/go-internal v1.11.0/go.mod h1:ddIwULY96R17DhadqLgMfk9H9tvdUzkipdSkR5nkCZA=
github.com/russross/blackfriday/v2 v2.1.0 h1:JIOH55/0cWyOuilr9/qlrm0BSXldqnqwMsf35Ld67mk=
github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
github.com/santhosh-tekuri/jsonschema/v5 v5.3.1 h1:lZUw3E0/J3roVtGQ+SCrUrg3ON6NgVqpn3+iol9aGu4=
@@ -258,6 +236,7 @@ github.com/sergi/go-diff v1.3.1 h1:xkr+Oxo4BOQKmkn/B9eMK0g5Kg/983T9DqqPHwYqD+8=
github.com/sergi/go-diff v1.3.1/go.mod h1:aMJSSKb2lpPvRNec0+w3fl7LP9IOFzdc9Pa4NFbPK1I=
github.com/sosodev/duration v1.3.1 h1:qtHBDMQ6lvMQsL15g4aopM4HEfOaYuhWBw3NPTtlqq4=
github.com/sosodev/duration v1.3.1/go.mod h1:RQIBBX0+fMLc/D9+Jb/fwvVmo0eZvDDEERAikUR6SDg=
github.com/spkg/bom v0.0.0-20160624110644-59b7046e48ad/go.mod h1:qLr4V1qq6nMqFKkMo8ZTx3f+BZEkzsRUY10Xsm2mwU0=
github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
@@ -265,58 +244,40 @@ github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXf
github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
github.com/stretchr/testify v1.4.0/go.mod h1:j7eGeouHqKxXV5pUuKE4zz7dFj8WfuZ+81PSLYec5m4=
github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
github.com/stretchr/testify v1.7.5/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu7U=
github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
github.com/stretchr/testify v1.10.0 h1:Xv5erBjTwe/5IxqUQTdXv5kgmIvbHo3QQyRwhJsOfJA=
github.com/stretchr/testify v1.10.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
github.com/swaggo/files v1.0.1 h1:J1bVJ4XHZNq0I46UU90611i9/YzdrF7x92oX1ig5IdE=
github.com/swaggo/files v1.0.1/go.mod h1:0qXmMNH6sXNf+73t65aKeB+ApmgxdnkQzVTAj2uaMUg=
github.com/swaggo/http-swagger v1.3.4 h1:q7t/XLx0n15H1Q9/tk3Y9L4n210XzJF5WtnDX64a5ww=
github.com/swaggo/http-swagger v1.3.4/go.mod h1:9dAh0unqMBAlbp1uE2Uc2mQTxNMU/ha4UbucIg1MFkQ=
github.com/swaggo/swag v1.16.6 h1:qBNcx53ZaX+M5dxVyTrgQ0PJ/ACK+NzhwcbieTt+9yI=
github.com/swaggo/swag v1.16.6/go.mod h1:ngP2etMK5a0P3QBizic5MEwpRmluJZPHjXcMoj4Xesg=
github.com/urfave/cli/v2 v2.27.7 h1:bH59vdhbjLv3LAvIu6gd0usJHgoTTPhCFib8qqOwXYU=
github.com/urfave/cli/v2 v2.27.7/go.mod h1:CyNAG/xg+iAOg0N4MPGZqVmv2rCoP267496AOXUZjA4=
github.com/vektah/gqlparser/v2 v2.5.30 h1:EqLwGAFLIzt1wpx1IPpY67DwUujF1OfzgEyDsLrN6kE=
github.com/vektah/gqlparser/v2 v2.5.30/go.mod h1:D1/VCZtV3LPnQrcPBeR/q5jkSQIPti0uYCP/RI0gIeo=
github.com/xrash/smetrics v0.0.0-20250705151800-55b8f293f342 h1:FnBeRrxr7OU4VvAzt5X7s6266i6cSVkkFPS0TuXWbIg=
github.com/xrash/smetrics v0.0.0-20250705151800-55b8f293f342/go.mod h1:Ohn+xnUBiLI6FVj/9LpzZWtj1/D6lUovWYBkxHVV3aM=
github.com/swaggo/swag v1.16.4 h1:clWJtd9LStiG3VeijiCfOVODP6VpHtKdQy9ELFG3s1A=
github.com/swaggo/swag v1.16.4/go.mod h1:VBsHJRsDvfYvqoiMKnsdwhNV9LEMHgEDZcyVYX0sxPg=
github.com/urfave/cli/v2 v2.27.5 h1:WoHEJLdsXr6dDWoJgMq/CboDmyY/8HMMH1fTECbih+w=
github.com/urfave/cli/v2 v2.27.5/go.mod h1:3Sevf16NykTbInEnD0yKkjDAeZDS0A6bzhBH5hrMvTQ=
github.com/vektah/gqlparser/v2 v2.5.22 h1:yaaeJ0fu+nv1vUMW0Hl+aS1eiv1vMfapBNjpffAda1I=
github.com/vektah/gqlparser/v2 v2.5.22/go.mod h1:xMl+ta8a5M1Yo1A1Iwt/k7gSpscwSnHZdw7tfhEGfTM=
github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1 h1:gEOO8jv9F4OT7lGCjxCBTO/36wtF6j2nSip77qHd4x4=
github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1/go.mod h1:Ohn+xnUBiLI6FVj/9LpzZWtj1/D6lUovWYBkxHVV3aM=
github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.54.0 h1:TT4fX+nBOA/+LUkobKGW1ydGcn+G3vRw9+g5HwCphpk=
go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.54.0/go.mod h1:L7UH0GbB0p47T4Rri3uHjbpCFYrVrwc1I25QhNPiGK8=
go.opentelemetry.io/otel v1.29.0 h1:PdomN/Al4q/lN6iBJEN3AwPvUiHPMlt93c8bqTG5Llw=
go.opentelemetry.io/otel v1.29.0/go.mod h1:N/WtXPs1CNCUEx+Agz5uouwCba+i+bJGFicT8SR4NP8=
go.opentelemetry.io/otel/metric v1.29.0 h1:vPf/HFWTNkPu1aYeIsc98l4ktOQaL6LeSoeV2g+8YLc=
go.opentelemetry.io/otel/metric v1.29.0/go.mod h1:auu/QWieFVWx+DmQOUMgj0F8LHWdgalxXqvp7BII/W8=
go.opentelemetry.io/otel/trace v1.29.0 h1:J/8ZNK4XgR7a21DZUAsbF8pZ5Jcw1VhACmnYt39JTi4=
go.opentelemetry.io/otel/trace v1.29.0/go.mod h1:eHl3w0sp3paPkYstJOmAimxhiFXPg+MMTlEh3nsQgWQ=
go.uber.org/atomic v1.11.0 h1:ZvwS0R+56ePWxUNi+Atn9dWONBPp/AUETXlHW0DxSjE=
go.uber.org/atomic v1.11.0/go.mod h1:LUxbIzbOniOlMKjJjyPfpl4v+PKK2cNJn91OQbhoJI0=
go.uber.org/goleak v1.3.0 h1:2K3zAYmnTNqV73imy9J1T3WC+gmCePx2hEGkimedGto=
go.uber.org/goleak v1.3.0/go.mod h1:CoHD4mav9JJNrW/WLlf7HGZPjdw8EucARQHekz1X6bE=
go.yaml.in/yaml/v2 v2.4.3 h1:6gvOSjQoTB3vt1l+CU+tSyi/HOjfOjRLJ4YwYZGwRO0=
go.yaml.in/yaml/v2 v2.4.3/go.mod h1:zSxWcmIDjOzPXpjlTTbAsKokqkDNAVtZO0WOMiT90s8=
go.yaml.in/yaml/v3 v3.0.4 h1:tfq32ie2Jv2UxXFdLJdh3jXuOzWiL1fo0bu/FbuKpbc=
go.yaml.in/yaml/v3 v3.0.4/go.mod h1:DhzuOOF2ATzADvBadXxruRBLzYTpT36CKvDb3+aBEFg=
golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
golang.org/x/crypto v0.0.0-20210921155107-089bfa567519/go.mod h1:GvvjBRRGRdwPK5ydBHafDWAxML/pGHZbMvKqRZ5+Abc=
golang.org/x/crypto v0.6.0/go.mod h1:OFC/31mSvZgRz0V1QTNCzfAI1aIRzbiufJtkMIlEp58=
golang.org/x/crypto v0.13.0/go.mod h1:y6Z2r+Rw4iayiXXAIxJIDAJ1zMW4yaTpebo8fPOliYc=
golang.org/x/crypto v0.19.0/go.mod h1:Iy9bg/ha4yyC70EfRS8jz+B6ybOBKMaSxLj6P6oBDfU=
golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8=
golang.org/x/crypto v0.31.0/go.mod h1:kDsLvtWBEx7MV9tJOj9bnXsPbxwJQ6csT/x4KIN4Ssk=
golang.org/x/crypto v0.43.0 h1:dduJYIi3A3KOfdGOHX8AVZ/jGiyPa3IbBozJ5kNuE04=
golang.org/x/crypto v0.43.0/go.mod h1:BFbav4mRNlXJL4wNeejLpWxB7wMbc79PdRGhWKncxR0=
golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b h1:M2rDM6z3Fhozi9O7NWsxAkg/yqS/lQJ6PmkyIV3YP+o=
golang.org/x/exp v0.0.0-20250620022241-b7579e27df2b/go.mod h1:3//PLf8L/X+8b4vuAfHzxeRUl04Adcb341+IGKfnqS8=
golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
golang.org/x/crypto v0.33.0 h1:IOBPskki6Lysi0lo9qQvbxiQ+FvsCC/YWOecCHAixus=
golang.org/x/crypto v0.33.0/go.mod h1:bVdXmD7IV/4GdElGPozy6U7lWdRXA4qyRVGJV57uQ5M=
golang.org/x/exp v0.0.0-20250215185904-eff6e970281f h1:oFMYAjX0867ZD2jcNiLBrI9BdpmEkvPyi5YrBGXbamg=
golang.org/x/exp v0.0.0-20250215185904-eff6e970281f/go.mod h1:BHOTPb3L19zxehTsLoJXVaTktb06DFgmdW6Wb9s8jqk=
golang.org/x/mod v0.6.0-dev.0.20220419223038-86c51ed26bb4/go.mod h1:jJ57K6gSWd91VN4djpZkiMVwK6gcyfeH4XE8wZrZaV4=
golang.org/x/mod v0.8.0/go.mod h1:iBbtSCu2XBx23ZKBPSOrRkjjQPZFPuis4dIYUhu/chs=
golang.org/x/mod v0.12.0/go.mod h1:iBbtSCu2XBx23ZKBPSOrRkjjQPZFPuis4dIYUhu/chs=
golang.org/x/mod v0.15.0/go.mod h1:hTbmBsO62+eylJbnUtE2MGJUyE7QWk4xUqPFrRgJ+7c=
golang.org/x/mod v0.17.0/go.mod h1:hTbmBsO62+eylJbnUtE2MGJUyE7QWk4xUqPFrRgJ+7c=
golang.org/x/mod v0.29.0 h1:HV8lRxZC4l2cr3Zq1LvtOsi/ThTgWnUk/y64QSs8GwA=
golang.org/x/mod v0.29.0/go.mod h1:NyhrlYXJ2H4eJiRy/WDBO6HMqZQ6q9nk4JzS3NuCK+w=
golang.org/x/mod v0.23.0 h1:Zb7khfcRGKk+kqfxFaP5tZqCnDZMjC5VtUBs87Hr6QM=
golang.org/x/mod v0.23.0/go.mod h1:6SkKJ3Xj0I0BrPOZoBy3bdMptDDU9oJrpohJ3eWZ1fY=
golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
golang.org/x/net v0.0.0-20200114155413-6afb5195e5aa/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
@@ -324,23 +285,17 @@ golang.org/x/net v0.0.0-20220722155237-a158d28d115b/go.mod h1:XRhObCWvk6IyKnWLug
golang.org/x/net v0.6.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
golang.org/x/net v0.7.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
golang.org/x/net v0.10.0/go.mod h1:0qNGK6F8kojg2nk9dLZ2mShWaEBan6FAoqfSigmmuDg=
golang.org/x/net v0.15.0/go.mod h1:idbUs1IY1+zTqbi8yxTbhexhEEk5ur9LInksu6HrEpk=
golang.org/x/net v0.21.0/go.mod h1:bIjVDfnllIU7BJ2DNgfnXvpSvtn8VRwhlsaeUTyUS44=
golang.org/x/net v0.25.0/go.mod h1:JkAGAh7GEvH74S6FOH42FLoXpXbE/aqXSrIQjXgsiwM=
golang.org/x/net v0.33.0/go.mod h1:HXLR5J+9DxmrqMwG9qjGCxZ+zKXxBru04zlTvWlWuN4=
golang.org/x/net v0.46.0 h1:giFlY12I07fugqwPuWJi68oOnpfqFnJIJzaIIm2JVV4=
golang.org/x/net v0.46.0/go.mod h1:Q9BGdFy1y4nkUwiLvT5qtyhAnEHgnQ/zd8PfU6nc210=
golang.org/x/oauth2 v0.32.0 h1:jsCblLleRMDrxMN29H3z/k1KliIvpLgCkE6R8FXXNgY=
golang.org/x/oauth2 v0.32.0/go.mod h1:lzm5WQJQwKZ3nwavOZ3IS5Aulzxi68dUSgRHujetwEA=
golang.org/x/net v0.22.0/go.mod h1:JKghWKKOSdJwpW2GEx0Ja7fmaKnMsbu+MWVZTokSYmg=
golang.org/x/net v0.35.0 h1:T5GQRQb2y08kTAByq9L4/bz8cipCdA8FbRTXewonqY8=
golang.org/x/net v0.35.0/go.mod h1:EglIi67kWsHKlRzzVMUD93VMSWGFOMSZgxFjparz1Qk=
golang.org/x/oauth2 v0.21.0 h1:tsimM75w1tF/uws5rbeHzIWxEqElMehnc+iW793zsZs=
golang.org/x/oauth2 v0.21.0/go.mod h1:XYTD2NtWslqkgxebSiOHnXEap4TF09sJSc7H1sXbhtI=
golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
golang.org/x/sync v0.0.0-20220722155255-886fb9371eb4/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
golang.org/x/sync v0.1.0/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
golang.org/x/sync v0.3.0/go.mod h1:FU7BRWz2tNW+3quACPkgCx/L+uEAv1htQ0V83Z9Rj+Y=
golang.org/x/sync v0.6.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
golang.org/x/sync v0.7.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
golang.org/x/sync v0.10.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
golang.org/x/sync v0.17.0 h1:l60nONMj9l5drqw6jlhIELNv9I0A4OFgRsG9k2oT9Ug=
golang.org/x/sync v0.17.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
golang.org/x/sync v0.11.0 h1:GGz8+XQP4FvTTrjZPzNKTMFtSXH80RAzG+5ghFPgK9w=
golang.org/x/sync v0.11.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
golang.org/x/sys v0.0.0-20201119102817-f84b799fce68/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
golang.org/x/sys v0.0.0-20210615035016-665e8c7367d1/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
@@ -348,54 +303,46 @@ golang.org/x/sys v0.0.0-20220520151302-bc2c85ada10a/go.mod h1:oPkhp1MJrh7nUepCBc
golang.org/x/sys v0.0.0-20220722155257-8c9f86f7a55f/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.8.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.12.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.17.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
golang.org/x/sys v0.28.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
golang.org/x/sys v0.37.0 h1:fdNQudmxPjkdUTPnLn5mdQv7Zwvbvpaxqs831goi9kQ=
golang.org/x/sys v0.37.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
golang.org/x/telemetry v0.0.0-20240228155512-f48c80bd79b2/go.mod h1:TeRTkGYfJXctD9OcfyVLyj2J3IxLnKwHJR8f4D8a3YE=
golang.org/x/sys v0.18.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
golang.org/x/sys v0.30.0 h1:QjkSwP/36a20jFYWkSue1YwXzLmsV5Gfq7Eiy72C1uc=
golang.org/x/sys v0.30.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=
golang.org/x/term v0.8.0/go.mod h1:xPskH00ivmX89bAKVGSKKtLOWNx2+17Eiy94tnKShWo=
golang.org/x/term v0.12.0/go.mod h1:owVbMEjm3cBLCHdkQu9b1opXd4ETQWc3BhuQGKgXgvU=
golang.org/x/term v0.17.0/go.mod h1:lLRBjIVuehSbZlaOtGMbcMncT+aqLLLmKrsjNrUguwk=
golang.org/x/term v0.20.0/go.mod h1:8UkIAJTvZgivsXaD6/pH6U9ecQzZ45awqEOzuCvwpFY=
golang.org/x/term v0.27.0/go.mod h1:iMsnZpn0cago0GOrHO2+Y7u7JPn5AylBrcoWkElMTSM=
golang.org/x/term v0.18.0/go.mod h1:ILwASektA3OnRv7amZ1xhE/KTR+u50pbXfZ03+6Nx58=
golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
golang.org/x/text v0.3.7/go.mod h1:u+2+/6zg+i71rQMx5EYifcz6MCKuco9NR6JIITiCfzQ=
golang.org/x/text v0.7.0/go.mod h1:mrYo+phRRbMaCq/xk9113O4dZlRixOauAjOtrjsXDZ8=
golang.org/x/text v0.9.0/go.mod h1:e1OnstbJyHTd6l/uOt8jFFHp6TRDWZR/bV3emEE/zU8=
golang.org/x/text v0.13.0/go.mod h1:TvPlkZtksWOMsz7fbANvkp4WM8x/WCo/om8BMLbz+aE=
golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
golang.org/x/text v0.21.0/go.mod h1:4IBbMaMmOPCJ8SecivzSH54+73PCFmPWxNTLm+vZkEQ=
golang.org/x/text v0.30.0 h1:yznKA/E9zq54KzlzBEAWn1NXSQ8DIp/NYMy88xJjl4k=
golang.org/x/text v0.30.0/go.mod h1:yDdHFIX9t+tORqspjENWgzaCVXgk0yYnYuSZ8UzzBVM=
golang.org/x/time v0.13.0 h1:eUlYslOIt32DgYD6utsuUeHs4d7AsEYLuIAdg7FlYgI=
golang.org/x/time v0.13.0/go.mod h1:eL/Oa2bBBK0TkX57Fyni+NgnyQQN4LitPmob2Hjnqw4=
golang.org/x/text v0.22.0 h1:bofq7m3/HAFvbF51jz3Q9wLg3jkvSPuiZu/pD1XwgtM=
golang.org/x/text v0.22.0/go.mod h1:YRoo4H8PVmsu+E3Ou7cqLVH8oXWIHVoX0jqUWALQhfY=
golang.org/x/time v0.10.0 h1:3usCWA8tQn0L8+hFJQNgzpWbd89begxN66o1Ojdn5L4=
golang.org/x/time v0.10.0/go.mod h1:3BpzKBy/shNhVucY/MWOyx10tF3SFh9QdLuxbVysPQM=
golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
golang.org/x/tools v0.0.0-20191119224855-298f0cb1881e/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
golang.org/x/tools v0.1.12/go.mod h1:hNGJHUnrk76NpqgfD5Aqm5Crs+Hm0VOH/i9J2+nxYbc=
golang.org/x/tools v0.6.0/go.mod h1:Xwgl3UAJ/d3gWutnCtw505GrjyAbvKui8lOU390QaIU=
golang.org/x/tools v0.13.0/go.mod h1:HvlwmtVNQAhOuCjW7xxvovg8wbNq7LwfXh/k7wXUl58=
golang.org/x/tools v0.21.1-0.20240508182429-e35e4ccd0d2d/go.mod h1:aiJjzUbINMkxbQROHiO6hDPo2LHcIPhhQsa9DLh0yGk=
golang.org/x/tools v0.38.0 h1:Hx2Xv8hISq8Lm16jvBZ2VQf+RLmbd7wVUsALibYI/IQ=
golang.org/x/tools v0.38.0/go.mod h1:yEsQ/d/YK8cjh0L6rZlY8tgtlKiBNTL14pGDJPJpYQs=
golang.org/x/tools v0.30.0 h1:BgcpHewrV5AUp2G9MebG4XPFI1E2W41zU1SaqVA9vJY=
golang.org/x/tools v0.30.0/go.mod h1:c347cR/OJfw5TI+GfX7RUPNMdDRRbjvYTS0jPyvsVtY=
golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
google.golang.org/protobuf v1.36.9 h1:w2gp2mA27hUeUzj9Ex9FBjsBm40zfaDtEWow293U7Iw=
google.golang.org/protobuf v1.36.9/go.mod h1:fuxRtAxBytpl4zzqUh6/eyUujkJdNiuEkXntxiD/uRU=
google.golang.org/protobuf v1.36.1 h1:yBPeRvTftaleIgM3PZ/WBIZ7XM/eEYAaEyCwvyjq/gk=
google.golang.org/protobuf v1.36.1/go.mod h1:9fA7Ob0pmnwhb644+1+CVWFRbNajQ6iRojtC/QF5bRE=
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
gopkg.in/check.v1 v1.0.0-20200227125254-8fa46927fb4f/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c h1:Hei/4ADfdWqJk1ZMxUNpqntNwaWcugrBjAiHlqqRiVk=
gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c/go.mod h1:JHkPIbrfpd72SG/EVd6muEfDQjcINNoR0C8j2r3qZ4Q=
gopkg.in/yaml.v2 v2.2.2/go.mod h1:hI93XBmqTisBFMUTm0b8Fm+jr3Dg1NNxqwp+5A1VGuI=
gopkg.in/yaml.v2 v2.4.0 h1:D8xgwECY7CYvx+Y2n4sBz93Jn9JRvxdiyyo8CTfuKaY=
gopkg.in/yaml.v2 v2.4.0/go.mod h1:RDklbk79AGWmwhnvt/jBztapEOGDOx6ZbXqjP6csGnQ=
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
gopkg.in/yaml.v3 v3.0.0-20200615113413-eeeca48fe776/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
sigs.k8s.io/yaml v1.6.0 h1:G8fkbMSAFqgEFgh4b1wmtzDnioxFCUgTZhlbj5P9QYs=
sigs.k8s.io/yaml v1.6.0/go.mod h1:796bPqUfzR/0jLAl6XjHl3Ck7MiyVv8dbTdyT3/pMf4=
sigs.k8s.io/yaml v1.4.0 h1:Mk1wCc2gy/F0THH0TAp1QYyJNzRm2KCLy3o5ASXVI5E=
sigs.k8s.io/yaml v1.4.0/go.mod h1:Ejl7/uTz7PSA4eKMyQCUTnhZYNmLIl+5c2lQPGR2BPY=

View File

@@ -30,9 +30,7 @@ resolver:
# gqlgen will search for any type names in the schema in these go packages
# if they match it will use them, otherwise it will generate them.
autobind:
- "github.com/99designs/gqlgen/graphql/introspection"
- "github.com/ClusterCockpit/cc-backend/internal/graph/model"
- "github.com/ClusterCockpit/cc-backend/internal/config"
# This section declares type mapping between the GraphQL and go type systems
#
@@ -52,51 +50,61 @@ models:
- github.com/99designs/gqlgen/graphql.Int64
- github.com/99designs/gqlgen/graphql.Int32
Job:
model: "github.com/ClusterCockpit/cc-lib/schema.Job"
model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Job"
fields:
tags:
resolver: true
metaData:
resolver: true
Cluster:
model: "github.com/ClusterCockpit/cc-lib/schema.Cluster"
model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Cluster"
fields:
partitions:
resolver: true
# Node:
# model: "github.com/ClusterCockpit/cc-lib/schema.Node"
# fields:
# metaData:
# resolver: true
NullableFloat: { model: "github.com/ClusterCockpit/cc-lib/schema.Float" }
MetricScope: { model: "github.com/ClusterCockpit/cc-lib/schema.MetricScope" }
MetricValue: { model: "github.com/ClusterCockpit/cc-lib/schema.MetricValue" }
NullableFloat:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Float" }
MetricScope:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricScope" }
MetricValue:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricValue" }
JobStatistics:
{ model: "github.com/ClusterCockpit/cc-lib/schema.JobStatistics" }
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobStatistics" }
GlobalMetricListItem:
{ model: "github.com/ClusterCockpit/cc-lib/schema.GlobalMetricListItem" }
{
model: "github.com/ClusterCockpit/cc-backend/pkg/schema.GlobalMetricListItem",
}
ClusterSupport:
{ model: "github.com/ClusterCockpit/cc-lib/schema.ClusterSupport" }
Tag: { model: "github.com/ClusterCockpit/cc-lib/schema.Tag" }
Resource: { model: "github.com/ClusterCockpit/cc-lib/schema.Resource" }
JobState: { model: "github.com/ClusterCockpit/cc-lib/schema.JobState" }
Node: { model: "github.com/ClusterCockpit/cc-lib/schema.Node" }
SchedulerState:
{ model: "github.com/ClusterCockpit/cc-lib/schema.SchedulerState" }
HealthState:
{ model: "github.com/ClusterCockpit/cc-lib/schema.MonitoringState" }
JobMetric: { model: "github.com/ClusterCockpit/cc-lib/schema.JobMetric" }
Series: { model: "github.com/ClusterCockpit/cc-lib/schema.Series" }
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.ClusterSupport" }
Tag: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Tag" }
Resource:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Resource" }
JobState:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobState" }
TimeRange:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.TimeRange" }
IntRange:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.IntRange" }
JobMetric:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobMetric" }
Series: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Series" }
MetricStatistics:
{ model: "github.com/ClusterCockpit/cc-lib/schema.MetricStatistics" }
{
model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricStatistics",
}
MetricConfig:
{ model: "github.com/ClusterCockpit/cc-lib/schema.MetricConfig" }
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricConfig" }
SubClusterConfig:
{ model: "github.com/ClusterCockpit/cc-lib/schema.SubClusterConfig" }
Accelerator: { model: "github.com/ClusterCockpit/cc-lib/schema.Accelerator" }
Topology: { model: "github.com/ClusterCockpit/cc-lib/schema.Topology" }
{
model: "github.com/ClusterCockpit/cc-backend/pkg/schema.SubClusterConfig",
}
Accelerator:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Accelerator" }
Topology:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Topology" }
FilterRanges:
{ model: "github.com/ClusterCockpit/cc-lib/schema.FilterRanges" }
SubCluster: { model: "github.com/ClusterCockpit/cc-lib/schema.SubCluster" }
StatsSeries: { model: "github.com/ClusterCockpit/cc-lib/schema.StatsSeries" }
Unit: { model: "github.com/ClusterCockpit/cc-lib/schema.Unit" }
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.FilterRanges" }
SubCluster:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.SubCluster" }
StatsSeries:
{ model: "github.com/ClusterCockpit/cc-backend/pkg/schema.StatsSeries" }
Unit: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Unit" }

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api_test
@@ -27,9 +27,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/metricdata"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/gorilla/mux"
_ "github.com/mattn/go-sqlite3"
@@ -37,22 +36,15 @@ import (
func setup(t *testing.T) *api.RestApi {
const testconfig = `{
"main": {
"addr": "0.0.0.0:8080",
"validate": false,
"apiAllowedIPs": [
"*"
]
},
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"auth": {
"jwts": {
"max-age": "2m"
}
},
"jwts": {
"max-age": "2m"
},
"clusters": [
{
"name": "testcluster",
@@ -121,22 +113,22 @@ func setup(t *testing.T) *api.RestApi {
]
}`
cclog.Init("info", true)
log.Init("info", true)
tmpdir := t.TempDir()
jobarchive := filepath.Join(tmpdir, "job-archive")
if err := os.Mkdir(jobarchive, 0o777); err != nil {
if err := os.Mkdir(jobarchive, 0777); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), fmt.Appendf(nil, "%d", 2), 0o666); err != nil {
if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), []byte(fmt.Sprintf("%d", 2)), 0666); err != nil {
t.Fatal(err)
}
if err := os.Mkdir(filepath.Join(jobarchive, "testcluster"), 0o777); err != nil {
if err := os.Mkdir(filepath.Join(jobarchive, "testcluster"), 0777); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(filepath.Join(jobarchive, "testcluster", "cluster.json"), []byte(testclusterJson), 0o666); err != nil {
if err := os.WriteFile(filepath.Join(jobarchive, "testcluster", "cluster.json"), []byte(testclusterJson), 0666); err != nil {
t.Fatal(err)
}
@@ -147,22 +139,11 @@ func setup(t *testing.T) *api.RestApi {
}
cfgFilePath := filepath.Join(tmpdir, "config.json")
if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0o666); err != nil {
if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0666); err != nil {
t.Fatal(err)
}
ccconf.Init(cfgFilePath)
// Load and check main configuration
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
config.Init(cfg, clustercfg)
} else {
cclog.Abort("Cluster configuration must be present")
}
} else {
cclog.Abort("Main configuration must be present")
}
config.Init(cfgFilePath)
archiveCfg := fmt.Sprintf("{\"kind\": \"file\",\"path\": \"%s\"}", jobarchive)
repository.Connect("sqlite3", dbfilepath)
@@ -176,14 +157,7 @@ func setup(t *testing.T) *api.RestApi {
}
archiver.Start(repository.GetJobRepository())
if cfg := ccconf.GetPackageConfig("auth"); cfg != nil {
auth.Init(&cfg)
} else {
cclog.Warn("Authentication disabled due to missing configuration")
auth.Init(nil)
}
auth.Init()
graph.Init()
return api.New()
@@ -227,11 +201,11 @@ func TestRestApi(t *testing.T) {
restapi.MountApiRoutes(r)
var TestJobId int64 = 123
TestClusterName := "testcluster"
var TestClusterName string = "testcluster"
var TestStartTime int64 = 123456789
const startJobBody string = `{
"jobId": 123,
"jobId": 123,
"user": "testuser",
"project": "testproj",
"cluster": "testcluster",
@@ -241,9 +215,10 @@ func TestRestApi(t *testing.T) {
"numNodes": 1,
"numHwthreads": 8,
"numAcc": 0,
"shared": "none",
"exclusive": 1,
"monitoringStatus": 1,
"smt": 1,
"tags": [{ "type": "testTagType", "name": "testTagName", "scope": "testuser" }],
"resources": [
{
"hostname": "host123",
@@ -274,17 +249,16 @@ func TestRestApi(t *testing.T) {
if response.StatusCode != http.StatusCreated {
t.Fatal(response.Status, recorder.Body.String())
}
// resolver := graph.GetResolverInstance()
restapi.JobRepository.SyncJobs()
resolver := graph.GetResolverInstance()
job, err := restapi.JobRepository.Find(&TestJobId, &TestClusterName, &TestStartTime)
if err != nil {
t.Fatal(err)
}
// job.Tags, err = resolver.Job().Tags(ctx, job)
// if err != nil {
// t.Fatal(err)
// }
job.Tags, err = resolver.Job().Tags(ctx, job)
if err != nil {
t.Fatal(err)
}
if job.JobID != 123 ||
job.User != "testuser" ||
@@ -293,20 +267,21 @@ func TestRestApi(t *testing.T) {
job.SubCluster != "sc1" ||
job.Partition != "default" ||
job.Walltime != 3600 ||
job.ArrayJobID != 0 ||
job.ArrayJobId != 0 ||
job.NumNodes != 1 ||
job.NumHWThreads != 8 ||
job.NumAcc != 0 ||
job.Exclusive != 1 ||
job.MonitoringStatus != 1 ||
job.SMT != 1 ||
!reflect.DeepEqual(job.Resources, []*schema.Resource{{Hostname: "host123", HWThreads: []int{0, 1, 2, 3, 4, 5, 6, 7}}}) ||
job.StartTime != 123456789 {
job.StartTime.Unix() != 123456789 {
t.Fatalf("unexpected job properties: %#v", job)
}
// if len(job.Tags) != 1 || job.Tags[0].Type != "testTagType" || job.Tags[0].Name != "testTagName" || job.Tags[0].Scope != "testuser" {
// t.Fatalf("unexpected tags: %#v", job.Tags)
// }
if len(job.Tags) != 1 || job.Tags[0].Type != "testTagType" || job.Tags[0].Name != "testTagName" || job.Tags[0].Scope != "testuser" {
t.Fatalf("unexpected tags: %#v", job.Tags)
}
}); !ok {
return
}
@@ -374,7 +349,7 @@ func TestRestApi(t *testing.T) {
t.Run("CheckDoubleStart", func(t *testing.T) {
// Starting a job with the same jobId and cluster should only be allowed if the startTime is far appart!
body := strings.ReplaceAll(startJobBody, `"startTime": 123456789`, `"startTime": 123456790`)
body := strings.Replace(startJobBody, `"startTime": 123456789`, `"startTime": 123456790`, -1)
req := httptest.NewRequest(http.MethodPost, "/jobs/start_job/", bytes.NewBuffer([]byte(body)))
recorder := httptest.NewRecorder()
@@ -396,7 +371,7 @@ func TestRestApi(t *testing.T) {
"partition": "default",
"walltime": 3600,
"numNodes": 1,
"shared": "none",
"exclusive": 1,
"monitoringStatus": 1,
"smt": 1,
"resources": [
@@ -424,7 +399,6 @@ func TestRestApi(t *testing.T) {
}
time.Sleep(1 * time.Second)
restapi.JobRepository.SyncJobs()
const stopJobBodyFailed string = `{
"jobId": 12345,

View File

@@ -1,71 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"bufio"
"encoding/json"
"fmt"
"net/http"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
"github.com/ClusterCockpit/cc-lib/schema"
)
// GetClustersAPIResponse model
type GetClustersAPIResponse struct {
Clusters []*schema.Cluster `json:"clusters"` // Array of clusters
}
// getClusters godoc
// @summary Lists all cluster configs
// @tags Cluster query
// @description Get a list of all cluster configs. Specific cluster can be requested using query parameter.
// @produce json
// @param cluster query string false "Job Cluster"
// @success 200 {object} api.GetClustersApiResponse "Array of clusters"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/clusters/ [get]
func (api *RestApi) getClusters(rw http.ResponseWriter, r *http.Request) {
if user := repository.GetUserFromContext(r.Context()); user != nil &&
!user.HasRole(schema.RoleApi) {
handleError(fmt.Errorf("missing role: %v", schema.GetRoleString(schema.RoleApi)), http.StatusForbidden, rw)
return
}
rw.Header().Add("Content-Type", "application/json")
bw := bufio.NewWriter(rw)
defer bw.Flush()
var clusters []*schema.Cluster
if r.URL.Query().Has("cluster") {
name := r.URL.Query().Get("cluster")
cluster := archive.GetCluster(name)
if cluster == nil {
handleError(fmt.Errorf("unknown cluster: %s", name), http.StatusBadRequest, rw)
return
}
clusters = append(clusters, cluster)
} else {
clusters = archive.Clusters
}
payload := GetClustersAPIResponse{
Clusters: clusters,
}
if err := json.NewEncoder(bw).Encode(payload); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}

View File

@@ -23,7 +23,7 @@ const docTemplate = `{
"host": "{{.Host}}",
"basePath": "{{.BasePath}}",
"paths": {
"/api/clusters/": {
"/clusters/": {
"get": {
"security": [
{
@@ -80,7 +80,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/": {
"/jobs/": {
"get": {
"security": [
{
@@ -175,7 +175,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/delete_job/": {
"/jobs/delete_job/": {
"delete": {
"security": [
{
@@ -208,7 +208,7 @@ const docTemplate = `{
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -250,7 +250,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/delete_job/{id}": {
"/jobs/delete_job/{id}": {
"delete": {
"security": [
{
@@ -278,7 +278,7 @@ const docTemplate = `{
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -320,7 +320,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/delete_job_before/{ts}": {
"/jobs/delete_job_before/{ts}": {
"delete": {
"security": [
{
@@ -348,7 +348,7 @@ const docTemplate = `{
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.DeleteJobApiResponse"
}
},
"400": {
@@ -390,7 +390,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/edit_meta/{id}": {
"/jobs/edit_meta/{id}": {
"post": {
"security": [
{
@@ -460,14 +460,14 @@ const docTemplate = `{
}
}
},
"/api/jobs/start_job/": {
"/jobs/start_job/": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job specified in request body will be saved to database as \"running\" with new DB ID.\nJob specifications follow the 'Job' scheme, API will fail to execute if requirements are not met.",
"description": "Job specified in request body will be saved to database as \"running\" with new DB ID.\nJob specifications follow the 'JobMeta' scheme, API will fail to execute if requirements are not met.",
"consumes": [
"application/json"
],
@@ -485,7 +485,7 @@ const docTemplate = `{
"in": "body",
"required": true,
"schema": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
}
],
@@ -493,7 +493,7 @@ const docTemplate = `{
"201": {
"description": "Job added successfully",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"$ref": "#/definitions/api.StartJobApiResponse"
}
},
"400": {
@@ -529,14 +529,14 @@ const docTemplate = `{
}
}
},
"/api/jobs/stop_job/": {
"/jobs/stop_job/": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job to stop is specified by request body. All fields are required in this case.\nReturns full job resource information according to 'Job' scheme.",
"description": "Job to stop is specified by request body. All fields are required in this case.\nReturns full job resource information according to 'JobMeta' scheme.",
"produces": [
"application/json"
],
@@ -559,7 +559,7 @@ const docTemplate = `{
"200": {
"description": "Success message",
"schema": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
},
"400": {
@@ -587,7 +587,7 @@ const docTemplate = `{
}
},
"422": {
"description": "Unprocessable Entity: job has already been stopped",
"description": "Unprocessable Entity: finding job failed: sql: no rows in result set",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
@@ -601,7 +601,7 @@ const docTemplate = `{
}
}
},
"/api/jobs/tag_job/{id}": {
"/jobs/tag_job/{id}": {
"post": {
"security": [
{
@@ -674,14 +674,14 @@ const docTemplate = `{
}
}
},
"/api/jobs/{id}": {
"/jobs/{id}": {
"get": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.",
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.",
"produces": [
"application/json"
],
@@ -755,7 +755,7 @@ const docTemplate = `{
"ApiKeyAuth": []
}
],
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.",
"description": "Job to get is specified by database ID\nReturns full job resource information according to 'JobMeta' scheme and all metrics according to 'JobData'.",
"consumes": [
"application/json"
],
@@ -833,74 +833,119 @@ const docTemplate = `{
}
}
},
"/api/nodestats/": {
"/user/{id}": {
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.",
"description": "Modifies user defined by username (id) in one of four possible ways.\nIf more than one formValue is set then only the highest priority field is used.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"multipart/form-data"
],
"produces": [
"application/json"
"text/plain"
],
"tags": [
"Nodestates"
"User"
],
"summary": "Deliver updated Slurm node states",
"summary": "Updates an existing user",
"parameters": [
{
"description": "Request body containing nodes and their states",
"name": "request",
"in": "body",
"required": true,
"schema": {
"$ref": "#/definitions/api.UpdateNodeStatesRequest"
}
"type": "string",
"description": "Database ID of User",
"name": "id",
"in": "path",
"required": true
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "Priority 1: Role to add",
"name": "add-role",
"in": "formData"
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "Priority 2: Role to remove",
"name": "remove-role",
"in": "formData"
},
{
"type": "string",
"description": "Priority 3: Project to add",
"name": "add-project",
"in": "formData"
},
{
"type": "string",
"description": "Priority 4: Project to remove",
"name": "remove-project",
"in": "formData"
}
],
"responses": {
"200": {
"description": "Success message",
"description": "Success Response Message",
"schema": {
"$ref": "#/definitions/api.DefaultApiResponse"
"type": "string"
}
},
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"403": {
"description": "Forbidden",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: The user could not be updated",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
}
}
}
},
"/api/users/": {
"/users/": {
"get": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.",
"description": "Returns a JSON-encoded list of users.\nRequired query-parameter defines if all users or only users with additional special roles are returned.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"produces": [
"application/json"
],
@@ -952,111 +997,70 @@ const docTemplate = `{
}
}
}
}
},
"/jobs/tag_job/{id}": {
"delete": {
},
"post": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Removes tag(s) from a job specified by DB ID. Name and Type of Tag(s) must match.\nTag Scope is required for matching, options: \"global\", \"admin\". Private tags can not be deleted via API.\nIf tagged job is already finished: Tag will be removed from respective archive files.",
"description": "User specified in form data will be saved to database.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"application/json"
],
"produces": [
"application/json"
],
"tags": [
"Job add and modify"
],
"summary": "Removes one or more tags from a job",
"parameters": [
{
"type": "integer",
"description": "Job Database ID",
"name": "id",
"in": "path",
"required": true
},
{
"description": "Array of tag-objects to remove",
"name": "request",
"in": "body",
"required": true,
"schema": {
"type": "array",
"items": {
"$ref": "#/definitions/api.ApiTag"
}
}
}
],
"responses": {
"200": {
"description": "Updated job resource",
"schema": {
"$ref": "#/definitions/schema.Job"
}
},
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"404": {
"description": "Job or tag does not exist",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
}
}
}
}
},
"/tags/": {
"delete": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "Removes tags by type and name. Name and Type of Tag(s) must match.\nTag Scope is required for matching, options: \"global\", \"admin\". Private tags can not be deleted via API.\nTag wills be removed from respective archive files.",
"consumes": [
"application/json"
"multipart/form-data"
],
"produces": [
"text/plain"
],
"tags": [
"Tag remove"
"User"
],
"summary": "Removes all tags and job-relations for type:name tuple",
"summary": "Adds a new user",
"parameters": [
{
"description": "Array of tag-objects to remove",
"name": "request",
"in": "body",
"required": true,
"schema": {
"type": "array",
"items": {
"$ref": "#/definitions/api.ApiTag"
}
}
"type": "string",
"description": "Unique user ID",
"name": "username",
"in": "formData",
"required": true
},
{
"type": "string",
"description": "User password",
"name": "password",
"in": "formData",
"required": true
},
{
"enum": [
"admin",
"support",
"manager",
"user",
"api"
],
"type": "string",
"description": "User role",
"name": "role",
"in": "formData",
"required": true
},
{
"type": "string",
"description": "Managed project, required for new manager role user",
"name": "project",
"in": "formData"
},
{
"type": "string",
"description": "Users name",
"name": "name",
"in": "formData"
},
{
"type": "string",
"description": "Users email",
"name": "email",
"in": "formData"
}
],
"responses": {
@@ -1069,25 +1073,93 @@ const docTemplate = `{
"400": {
"description": "Bad Request",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"404": {
"description": "Job or tag does not exist",
"403": {
"description": "Forbidden",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: creating user failed",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"$ref": "#/definitions/api.ErrorResponse"
"type": "string"
}
}
}
},
"delete": {
"security": [
{
"ApiKeyAuth": []
}
],
"description": "User defined by username in form data will be deleted from database.\nOnly accessible from IPs registered with apiAllowedIPs configuration option.",
"consumes": [
"multipart/form-data"
],
"produces": [
"text/plain"
],
"tags": [
"User"
],
"summary": "Deletes a user",
"parameters": [
{
"type": "string",
"description": "User ID to delete",
"name": "username",
"in": "formData",
"required": true
}
],
"responses": {
"200": {
"description": "User deleted successfully"
},
"400": {
"description": "Bad Request",
"schema": {
"type": "string"
}
},
"401": {
"description": "Unauthorized",
"schema": {
"type": "string"
}
},
"403": {
"description": "Forbidden",
"schema": {
"type": "string"
}
},
"422": {
"description": "Unprocessable Entity: deleting user failed",
"schema": {
"type": "string"
}
},
"500": {
"description": "Internal Server Error",
"schema": {
"type": "string"
}
}
}
@@ -1141,14 +1213,6 @@ const docTemplate = `{
}
}
},
"api.DefaultApiResponse": {
"type": "object",
"properties": {
"msg": {
"type": "string"
}
}
},
"api.DeleteJobApiRequest": {
"type": "object",
"required": [
@@ -1172,6 +1236,14 @@ const docTemplate = `{
}
}
},
"api.DeleteJobApiResponse": {
"type": "object",
"properties": {
"msg": {
"type": "string"
}
}
},
"api.EditMetaRequest": {
"type": "object",
"properties": {
@@ -1235,7 +1307,7 @@ const docTemplate = `{
"description": "Array of jobs",
"type": "array",
"items": {
"$ref": "#/definitions/schema.Job"
"$ref": "#/definitions/schema.JobMeta"
}
},
"page": {
@@ -1258,35 +1330,11 @@ const docTemplate = `{
}
}
},
"api.Node": {
"api.StartJobApiResponse": {
"type": "object",
"properties": {
"cpusAllocated": {
"type": "integer"
},
"cpusTotal": {
"type": "integer"
},
"gpusAllocated": {
"type": "integer"
},
"gpusTotal": {
"type": "integer"
},
"hostname": {
"msg": {
"type": "string"
},
"memoryAllocated": {
"type": "integer"
},
"memoryTotal": {
"type": "integer"
},
"states": {
"type": "array",
"items": {
"type": "string"
}
}
}
},
@@ -1323,21 +1371,6 @@ const docTemplate = `{
}
}
},
"api.UpdateNodeStatesRequest": {
"type": "object",
"properties": {
"cluster": {
"type": "string",
"example": "fritz"
},
"nodes": {
"type": "array",
"items": {
"$ref": "#/definitions/api.Node"
}
}
}
},
"schema.Accelerator": {
"type": "object",
"properties": {
@@ -1373,6 +1406,7 @@ const docTemplate = `{
}
},
"schema.Job": {
"description": "Information of a HPC job.",
"type": "object",
"properties": {
"arrayJobId": {
@@ -1397,15 +1431,19 @@ const docTemplate = `{
"energyFootprint": {
"type": "object",
"additionalProperties": {
"type": "number",
"format": "float64"
"type": "number"
}
},
"exclusive": {
"type": "integer",
"maximum": 2,
"minimum": 0,
"example": 1
},
"footprint": {
"type": "object",
"additionalProperties": {
"type": "number",
"format": "float64"
"type": "number"
}
},
"id": {
@@ -1417,18 +1455,12 @@ const docTemplate = `{
},
"jobState": {
"enum": [
"boot_fail",
"cancelled",
"completed",
"deadline",
"failed",
"node_fail",
"out-of-memory",
"pending",
"preempted",
"running",
"suspended",
"timeout"
"cancelled",
"stopped",
"timeout",
"out_of_memory"
],
"allOf": [
{
@@ -1472,48 +1504,23 @@ const docTemplate = `{
"type": "string",
"example": "abcd200"
},
"requestedMemory": {
"description": "in MB",
"type": "integer",
"minimum": 1,
"example": 128000
},
"resources": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Resource"
}
},
"shared": {
"type": "string",
"enum": [
"none",
"single_user",
"multi_user"
]
},
"smt": {
"type": "integer",
"example": 4
},
"startTime": {
"type": "integer",
"example": 1649723812
},
"statistics": {
"type": "object",
"additionalProperties": {
"$ref": "#/definitions/schema.JobStatistics"
}
"type": "string"
},
"subCluster": {
"type": "string",
"example": "main"
},
"submitTime": {
"type": "integer",
"example": 1649723812
},
"tags": {
"type": "array",
"items": {
@@ -1556,6 +1563,147 @@ const docTemplate = `{
}
}
},
"schema.JobMeta": {
"description": "Meta data information of a HPC job.",
"type": "object",
"properties": {
"arrayJobId": {
"type": "integer",
"example": 123000
},
"cluster": {
"type": "string",
"example": "fritz"
},
"concurrentJobs": {
"$ref": "#/definitions/schema.JobLinkResultList"
},
"duration": {
"type": "integer",
"minimum": 1,
"example": 43200
},
"energy": {
"type": "number"
},
"energyFootprint": {
"type": "object",
"additionalProperties": {
"type": "number"
}
},
"exclusive": {
"type": "integer",
"maximum": 2,
"minimum": 0,
"example": 1
},
"footprint": {
"type": "object",
"additionalProperties": {
"type": "number"
}
},
"id": {
"type": "integer"
},
"jobId": {
"type": "integer",
"example": 123000
},
"jobState": {
"enum": [
"completed",
"failed",
"cancelled",
"stopped",
"timeout",
"out_of_memory"
],
"allOf": [
{
"$ref": "#/definitions/schema.JobState"
}
],
"example": "completed"
},
"metaData": {
"type": "object",
"additionalProperties": {
"type": "string"
}
},
"monitoringStatus": {
"type": "integer",
"maximum": 3,
"minimum": 0,
"example": 1
},
"numAcc": {
"type": "integer",
"minimum": 1,
"example": 2
},
"numHwthreads": {
"type": "integer",
"minimum": 1,
"example": 20
},
"numNodes": {
"type": "integer",
"minimum": 1,
"example": 2
},
"partition": {
"type": "string",
"example": "main"
},
"project": {
"type": "string",
"example": "abcd200"
},
"resources": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Resource"
}
},
"smt": {
"type": "integer",
"example": 4
},
"startTime": {
"type": "integer",
"minimum": 1,
"example": 1649723812
},
"statistics": {
"type": "object",
"additionalProperties": {
"$ref": "#/definitions/schema.JobStatistics"
}
},
"subCluster": {
"type": "string",
"example": "main"
},
"tags": {
"type": "array",
"items": {
"$ref": "#/definitions/schema.Tag"
}
},
"user": {
"type": "string",
"example": "abcd100h"
},
"walltime": {
"type": "integer",
"minimum": 1,
"example": 86400
}
}
},
"schema.JobMetric": {
"type": "object",
"properties": {
@@ -1579,32 +1727,24 @@ const docTemplate = `{
"schema.JobState": {
"type": "string",
"enum": [
"boot_fail",
"cancelled",
"completed",
"deadline",
"failed",
"node_fail",
"out_of_memory",
"pending",
"preempted",
"running",
"suspended",
"timeout"
"completed",
"failed",
"cancelled",
"stopped",
"timeout",
"preempted",
"out_of_memory"
],
"x-enum-varnames": [
"JobStateBootFail",
"JobStateCancelled",
"JobStateCompleted",
"JobStateDeadline",
"JobStateFailed",
"JobStateNodeFail",
"JobStateOutOfMemory",
"JobStatePending",
"JobStatePreempted",
"JobStateRunning",
"JobStateSuspended",
"JobStateTimeout"
"JobStateCompleted",
"JobStateFailed",
"JobStateCancelled",
"JobStateStopped",
"JobStateTimeout",
"JobStatePreempted",
"JobStateOutOfMemory"
]
},
"schema.JobStatistics": {
@@ -1803,8 +1943,7 @@ const docTemplate = `{
"additionalProperties": {
"type": "array",
"items": {
"type": "number",
"format": "float64"
"type": "number"
}
}
}
@@ -1892,9 +2031,6 @@ const docTemplate = `{
},
"remove": {
"type": "boolean"
},
"unit": {
"$ref": "#/definitions/schema.Unit"
}
}
},
@@ -1997,7 +2133,7 @@ const docTemplate = `{
var SwaggerInfo = &swag.Spec{
Version: "1.0.0",
Host: "localhost:8080",
BasePath: "",
BasePath: "/api",
Schemes: []string{},
Title: "ClusterCockpit REST API",
Description: "API for batch job control.",

View File

@@ -1,984 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"bufio"
"database/sql"
"encoding/json"
"errors"
"fmt"
"net/http"
"strconv"
"strings"
"sync"
"time"
"github.com/ClusterCockpit/cc-backend/internal/archiver"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/graph"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/internal/importer"
"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/gorilla/mux"
)
// StopJobApiRequest model
type StopJobApiRequest struct {
JobId *int64 `json:"jobId" example:"123000"`
Cluster *string `json:"cluster" example:"fritz"`
StartTime *int64 `json:"startTime" example:"1649723812"`
State schema.JobState `json:"jobState" validate:"required" example:"completed"`
StopTime int64 `json:"stopTime" validate:"required" example:"1649763839"`
}
// DeleteJobApiRequest model
type DeleteJobApiRequest struct {
JobId *int64 `json:"jobId" validate:"required" example:"123000"` // Cluster Job ID of job
Cluster *string `json:"cluster" example:"fritz"` // Cluster of job
StartTime *int64 `json:"startTime" example:"1649723812"` // Start Time of job as epoch
}
// GetJobsApiResponse model
type GetJobsApiResponse struct {
Jobs []*schema.Job `json:"jobs"` // Array of jobs
Items int `json:"items"` // Number of jobs returned
Page int `json:"page"` // Page id returned
}
// ApiTag model
type ApiTag struct {
// Tag Type
Type string `json:"type" example:"Debug"`
Name string `json:"name" example:"Testjob"` // Tag Name
Scope string `json:"scope" example:"global"` // Tag Scope for Frontend Display
}
// ApiMeta model
type EditMetaRequest struct {
Key string `json:"key" example:"jobScript"`
Value string `json:"value" example:"bash script"`
}
type TagJobApiRequest []*ApiTag
type GetJobApiRequest []string
type GetJobApiResponse struct {
Meta *schema.Job
Data []*JobMetricWithName
}
type GetCompleteJobApiResponse struct {
Meta *schema.Job
Data schema.JobData
}
type JobMetricWithName struct {
Metric *schema.JobMetric `json:"metric"`
Name string `json:"name"`
Scope schema.MetricScope `json:"scope"`
}
// getJobs godoc
// @summary Lists all jobs
// @tags Job query
// @description Get a list of all jobs. Filters can be applied using query parameters.
// @description Number of results can be limited by page. Results are sorted by descending startTime.
// @produce json
// @param state query string false "Job State" Enums(running, completed, failed, cancelled, stopped, timeout)
// @param cluster query string false "Job Cluster"
// @param start-time query string false "Syntax: '$from-$to', as unix epoch timestamps in seconds"
// @param items-per-page query int false "Items per page (Default: 25)"
// @param page query int false "Page Number (Default: 1)"
// @param with-metadata query bool false "Include metadata (e.g. jobScript) in response"
// @success 200 {object} api.GetJobsApiResponse "Job array and page info"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/ [get]
func (api *RestApi) getJobs(rw http.ResponseWriter, r *http.Request) {
withMetadata := false
filter := &model.JobFilter{}
page := &model.PageRequest{ItemsPerPage: 25, Page: 1}
order := &model.OrderByInput{Field: "startTime", Type: "col", Order: model.SortDirectionEnumDesc}
for key, vals := range r.URL.Query() {
switch key {
// TODO: add project filter
case "state":
for _, s := range vals {
state := schema.JobState(s)
if !state.Valid() {
handleError(fmt.Errorf("invalid query parameter value: state"),
http.StatusBadRequest, rw)
return
}
filter.State = append(filter.State, state)
}
case "cluster":
filter.Cluster = &model.StringInput{Eq: &vals[0]}
case "start-time": // ?startTime=1753707480-1754053139
st := strings.Split(vals[0], "-")
if len(st) != 2 {
handleError(fmt.Errorf("invalid query parameter value: startTime"),
http.StatusBadRequest, rw)
return
}
from, err := strconv.ParseInt(st[0], 10, 64)
if err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
to, err := strconv.ParseInt(st[1], 10, 64)
if err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
ufrom, uto := time.Unix(from, 0), time.Unix(to, 0)
filter.StartTime = &config.TimeRange{From: &ufrom, To: &uto}
case "page":
x, err := strconv.Atoi(vals[0])
if err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
page.Page = x
case "items-per-page":
x, err := strconv.Atoi(vals[0])
if err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
page.ItemsPerPage = x
case "with-metadata":
withMetadata = true
default:
handleError(fmt.Errorf("invalid query parameter: %s", key),
http.StatusBadRequest, rw)
return
}
}
jobs, err := api.JobRepository.QueryJobs(r.Context(), []*model.JobFilter{filter}, page, order)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
results := make([]*schema.Job, 0, len(jobs))
for _, job := range jobs {
if withMetadata {
if _, err = api.JobRepository.FetchMetadata(job); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}
job.Tags, err = api.JobRepository.GetTags(repository.GetUserFromContext(r.Context()), job.ID)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
if job.MonitoringStatus == schema.MonitoringStatusArchivingSuccessful {
job.Statistics, err = archive.GetStatistics(job)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}
results = append(results, job)
}
cclog.Debugf("/api/jobs: %d jobs returned", len(results))
rw.Header().Add("Content-Type", "application/json")
bw := bufio.NewWriter(rw)
defer bw.Flush()
payload := GetJobsApiResponse{
Jobs: results,
Items: page.ItemsPerPage,
Page: page.Page,
}
if err := json.NewEncoder(bw).Encode(payload); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}
// getCompleteJobById godoc
// @summary Get job meta and optional all metric data
// @tags Job query
// @description Job to get is specified by database ID
// @description Returns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.
// @produce json
// @param id path int true "Database ID of Job"
// @param all-metrics query bool false "Include all available metrics"
// @success 200 {object} api.GetJobApiResponse "Job resource"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: finding job failed: sql: no rows in result set"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/{id} [get]
func (api *RestApi) getCompleteJobById(rw http.ResponseWriter, r *http.Request) {
// Fetch job from db
id, ok := mux.Vars(r)["id"]
var job *schema.Job
var err error
if ok {
id, e := strconv.ParseInt(id, 10, 64)
if e != nil {
handleError(fmt.Errorf("integer expected in path for id: %w", e), http.StatusBadRequest, rw)
return
}
job, err = api.JobRepository.FindById(r.Context(), id) // Get Job from Repo by ID
} else {
handleError(fmt.Errorf("the parameter 'id' is required"), http.StatusBadRequest, rw)
return
}
if err != nil {
handleError(fmt.Errorf("finding job with db id %s failed: %w", id, err), http.StatusUnprocessableEntity, rw)
return
}
job.Tags, err = api.JobRepository.GetTags(repository.GetUserFromContext(r.Context()), job.ID)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
if _, err = api.JobRepository.FetchMetadata(job); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
var scopes []schema.MetricScope
if job.NumNodes == 1 {
scopes = []schema.MetricScope{"core"}
} else {
scopes = []schema.MetricScope{"node"}
}
var data schema.JobData
metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
resolution := 0
for _, mc := range metricConfigs {
resolution = max(resolution, mc.Timestep)
}
if r.URL.Query().Get("all-metrics") == "true" {
data, err = metricDataDispatcher.LoadData(job, nil, scopes, r.Context(), resolution)
if err != nil {
cclog.Warnf("REST: error while loading all-metrics job data for JobID %d on %s", job.JobID, job.Cluster)
return
}
}
cclog.Debugf("/api/job/%s: get job %d", id, job.JobID)
rw.Header().Add("Content-Type", "application/json")
bw := bufio.NewWriter(rw)
defer bw.Flush()
payload := GetCompleteJobApiResponse{
Meta: job,
Data: data,
}
if err := json.NewEncoder(bw).Encode(payload); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}
// getJobById godoc
// @summary Get job meta and configurable metric data
// @tags Job query
// @description Job to get is specified by database ID
// @description Returns full job resource information according to 'Job' scheme and all metrics according to 'JobData'.
// @accept json
// @produce json
// @param id path int true "Database ID of Job"
// @param request body api.GetJobApiRequest true "Array of metric names"
// @success 200 {object} api.GetJobApiResponse "Job resource"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: finding job failed: sql: no rows in result set"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/{id} [post]
func (api *RestApi) getJobById(rw http.ResponseWriter, r *http.Request) {
// Fetch job from db
id, ok := mux.Vars(r)["id"]
var job *schema.Job
var err error
if ok {
id, e := strconv.ParseInt(id, 10, 64)
if e != nil {
handleError(fmt.Errorf("integer expected in path for id: %w", e), http.StatusBadRequest, rw)
return
}
job, err = api.JobRepository.FindById(r.Context(), id)
} else {
handleError(errors.New("the parameter 'id' is required"), http.StatusBadRequest, rw)
return
}
if err != nil {
handleError(fmt.Errorf("finding job with db id %s failed: %w", id, err), http.StatusUnprocessableEntity, rw)
return
}
job.Tags, err = api.JobRepository.GetTags(repository.GetUserFromContext(r.Context()), job.ID)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
if _, err = api.JobRepository.FetchMetadata(job); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
var metrics GetJobApiRequest
if err = decode(r.Body, &metrics); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
var scopes []schema.MetricScope
if job.NumNodes == 1 {
scopes = []schema.MetricScope{"core"}
} else {
scopes = []schema.MetricScope{"node"}
}
metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
resolution := 0
for _, mc := range metricConfigs {
resolution = max(resolution, mc.Timestep)
}
data, err := metricDataDispatcher.LoadData(job, metrics, scopes, r.Context(), resolution)
if err != nil {
cclog.Warnf("REST: error while loading job data for JobID %d on %s", job.JobID, job.Cluster)
return
}
res := []*JobMetricWithName{}
for name, md := range data {
for scope, metric := range md {
res = append(res, &JobMetricWithName{
Name: name,
Scope: scope,
Metric: metric,
})
}
}
cclog.Debugf("/api/job/%s: get job %d", id, job.JobID)
rw.Header().Add("Content-Type", "application/json")
bw := bufio.NewWriter(rw)
defer bw.Flush()
payload := GetJobApiResponse{
Meta: job,
Data: res,
}
if err := json.NewEncoder(bw).Encode(payload); err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
}
// editMeta godoc
// @summary Edit meta-data json
// @tags Job add and modify
// @description Edit key value pairs in job metadata json
// @description If a key already exists its content will be overwritten
// @accept json
// @produce json
// @param id path int true "Job Database ID"
// @param request body api.EditMetaRequest true "Kay value pair to add"
// @success 200 {object} schema.Job "Updated job resource"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 404 {object} api.ErrorResponse "Job does not exist"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/edit_meta/{id} [post]
func (api *RestApi) editMeta(rw http.ResponseWriter, r *http.Request) {
id, err := strconv.ParseInt(mux.Vars(r)["id"], 10, 64)
if err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
job, err := api.JobRepository.FindById(r.Context(), id)
if err != nil {
http.Error(rw, err.Error(), http.StatusNotFound)
return
}
var req EditMetaRequest
if err := decode(r.Body, &req); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
if err := api.JobRepository.UpdateMetadata(job, req.Key, req.Value); err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(job)
}
// tagJob godoc
// @summary Adds one or more tags to a job
// @tags Job add and modify
// @description Adds tag(s) to a job specified by DB ID. Name and Type of Tag(s) can be chosen freely.
// @description Tag Scope for frontend visibility will default to "global" if none entered, other options: "admin" or specific username.
// @description If tagged job is already finished: Tag will be written directly to respective archive files.
// @accept json
// @produce json
// @param id path int true "Job Database ID"
// @param request body api.TagJobApiRequest true "Array of tag-objects to add"
// @success 200 {object} schema.Job "Updated job resource"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 404 {object} api.ErrorResponse "Job or tag does not exist"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/tag_job/{id} [post]
func (api *RestApi) tagJob(rw http.ResponseWriter, r *http.Request) {
id, err := strconv.ParseInt(mux.Vars(r)["id"], 10, 64)
if err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
job, err := api.JobRepository.FindById(r.Context(), id)
if err != nil {
http.Error(rw, err.Error(), http.StatusNotFound)
return
}
job.Tags, err = api.JobRepository.GetTags(repository.GetUserFromContext(r.Context()), job.ID)
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
var req TagJobApiRequest
if err := decode(r.Body, &req); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
for _, tag := range req {
tagId, err := api.JobRepository.AddTagOrCreate(repository.GetUserFromContext(r.Context()), *job.ID, tag.Type, tag.Name, tag.Scope)
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
job.Tags = append(job.Tags, &schema.Tag{
ID: tagId,
Type: tag.Type,
Name: tag.Name,
Scope: tag.Scope,
})
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(job)
}
// removeTagJob godoc
// @summary Removes one or more tags from a job
// @tags Job add and modify
// @description Removes tag(s) from a job specified by DB ID. Name and Type of Tag(s) must match.
// @description Tag Scope is required for matching, options: "global", "admin". Private tags can not be deleted via API.
// @description If tagged job is already finished: Tag will be removed from respective archive files.
// @accept json
// @produce json
// @param id path int true "Job Database ID"
// @param request body api.TagJobApiRequest true "Array of tag-objects to remove"
// @success 200 {object} schema.Job "Updated job resource"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 404 {object} api.ErrorResponse "Job or tag does not exist"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /jobs/tag_job/{id} [delete]
func (api *RestApi) removeTagJob(rw http.ResponseWriter, r *http.Request) {
id, err := strconv.ParseInt(mux.Vars(r)["id"], 10, 64)
if err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
job, err := api.JobRepository.FindById(r.Context(), id)
if err != nil {
http.Error(rw, err.Error(), http.StatusNotFound)
return
}
job.Tags, err = api.JobRepository.GetTags(repository.GetUserFromContext(r.Context()), job.ID)
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
var req TagJobApiRequest
if err := decode(r.Body, &req); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
for _, rtag := range req {
// Only Global and Admin Tags
if rtag.Scope != "global" && rtag.Scope != "admin" {
cclog.Warnf("Cannot delete private tag for job %d: Skip", job.JobID)
continue
}
remainingTags, err := api.JobRepository.RemoveJobTagByRequest(repository.GetUserFromContext(r.Context()), *job.ID, rtag.Type, rtag.Name, rtag.Scope)
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
job.Tags = remainingTags
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(job)
}
// removeTags godoc
// @summary Removes all tags and job-relations for type:name tuple
// @tags Tag remove
// @description Removes tags by type and name. Name and Type of Tag(s) must match.
// @description Tag Scope is required for matching, options: "global", "admin". Private tags can not be deleted via API.
// @description Tag wills be removed from respective archive files.
// @accept json
// @produce plain
// @param request body api.TagJobApiRequest true "Array of tag-objects to remove"
// @success 200 {string} string "Success Response"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 404 {object} api.ErrorResponse "Job or tag does not exist"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /tags/ [delete]
func (api *RestApi) removeTags(rw http.ResponseWriter, r *http.Request) {
var req TagJobApiRequest
if err := decode(r.Body, &req); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
targetCount := len(req)
currentCount := 0
for _, rtag := range req {
// Only Global and Admin Tags
if rtag.Scope != "global" && rtag.Scope != "admin" {
cclog.Warn("Cannot delete private tag: Skip")
continue
}
err := api.JobRepository.RemoveTagByRequest(rtag.Type, rtag.Name, rtag.Scope)
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
} else {
currentCount++
}
}
rw.WriteHeader(http.StatusOK)
fmt.Fprintf(rw, "Deleted Tags from DB: %d successfull of %d requested\n", currentCount, targetCount)
}
// startJob godoc
// @summary Adds a new job as "running"
// @tags Job add and modify
// @description Job specified in request body will be saved to database as "running" with new DB ID.
// @description Job specifications follow the 'Job' scheme, API will fail to execute if requirements are not met.
// @accept json
// @produce json
// @param request body schema.Job true "Job to add"
// @success 201 {object} api.DefaultApiResponse "Job added successfully"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: The combination of jobId, clusterId and startTime does already exist"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/start_job/ [post]
func (api *RestApi) startJob(rw http.ResponseWriter, r *http.Request) {
req := schema.Job{
Shared: "none",
MonitoringStatus: schema.MonitoringStatusRunningOrArchiving,
}
if err := decode(r.Body, &req); err != nil {
handleError(fmt.Errorf("parsing request body failed: %w", err), http.StatusBadRequest, rw)
return
}
cclog.Printf("REST: %s\n", req.GoString())
req.State = schema.JobStateRunning
if err := importer.SanityChecks(&req); err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
// aquire lock to avoid race condition between API calls
var unlockOnce sync.Once
api.RepositoryMutex.Lock()
defer unlockOnce.Do(api.RepositoryMutex.Unlock)
// Check if combination of (job_id, cluster_id, start_time) already exists:
jobs, err := api.JobRepository.FindAll(&req.JobID, &req.Cluster, nil)
if err != nil && err != sql.ErrNoRows {
handleError(fmt.Errorf("checking for duplicate failed: %w", err), http.StatusInternalServerError, rw)
return
} else if err == nil {
for _, job := range jobs {
if (req.StartTime - job.StartTime) < 86400 {
handleError(fmt.Errorf("a job with that jobId, cluster and startTime already exists: dbid: %d, jobid: %d", job.ID, job.JobID), http.StatusUnprocessableEntity, rw)
return
}
}
}
id, err := api.JobRepository.Start(&req)
if err != nil {
handleError(fmt.Errorf("insert into database failed: %w", err), http.StatusInternalServerError, rw)
return
}
// unlock here, adding Tags can be async
unlockOnce.Do(api.RepositoryMutex.Unlock)
for _, tag := range req.Tags {
if _, err := api.JobRepository.AddTagOrCreate(repository.GetUserFromContext(r.Context()), id, tag.Type, tag.Name, tag.Scope); err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
handleError(fmt.Errorf("adding tag to new job %d failed: %w", id, err), http.StatusInternalServerError, rw)
return
}
}
cclog.Printf("new job (id: %d): cluster=%s, jobId=%d, user=%s, startTime=%d", id, req.Cluster, req.JobID, req.User, req.StartTime)
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusCreated)
json.NewEncoder(rw).Encode(DefaultApiResponse{
Message: "success",
})
}
// stopJobByRequest godoc
// @summary Marks job as completed and triggers archiving
// @tags Job add and modify
// @description Job to stop is specified by request body. All fields are required in this case.
// @description Returns full job resource information according to 'Job' scheme.
// @produce json
// @param request body api.StopJobApiRequest true "All fields required"
// @success 200 {object} schema.Job "Success message"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: job has already been stopped"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/stop_job/ [post]
func (api *RestApi) stopJobByRequest(rw http.ResponseWriter, r *http.Request) {
// Parse request body
req := StopJobApiRequest{}
if err := decode(r.Body, &req); err != nil {
handleError(fmt.Errorf("parsing request body failed: %w", err), http.StatusBadRequest, rw)
return
}
// Fetch job (that will be stopped) from db
var job *schema.Job
var err error
if req.JobId == nil {
handleError(errors.New("the field 'jobId' is required"), http.StatusBadRequest, rw)
return
}
// cclog.Printf("loading db job for stopJobByRequest... : stopJobApiRequest=%v", req)
job, err = api.JobRepository.Find(req.JobId, req.Cluster, req.StartTime)
if err != nil {
job, err = api.JobRepository.FindCached(req.JobId, req.Cluster, req.StartTime)
// FIXME: Previous error is hidden
if err != nil {
handleError(fmt.Errorf("finding job failed: %w", err), http.StatusUnprocessableEntity, rw)
return
}
}
api.checkAndHandleStopJob(rw, job, req)
}
// deleteJobById godoc
// @summary Remove a job from the sql database
// @tags Job remove
// @description Job to remove is specified by database ID. This will not remove the job from the job archive.
// @produce json
// @param id path int true "Database ID of Job"
// @success 200 {object} api.DefaultApiResponse "Success message"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: finding job failed: sql: no rows in result set"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/delete_job/{id} [delete]
func (api *RestApi) deleteJobById(rw http.ResponseWriter, r *http.Request) {
// Fetch job (that will be stopped) from db
id, ok := mux.Vars(r)["id"]
var err error
if ok {
id, e := strconv.ParseInt(id, 10, 64)
if e != nil {
handleError(fmt.Errorf("integer expected in path for id: %w", e), http.StatusBadRequest, rw)
return
}
err = api.JobRepository.DeleteJobById(id)
} else {
handleError(errors.New("the parameter 'id' is required"), http.StatusBadRequest, rw)
return
}
if err != nil {
handleError(fmt.Errorf("deleting job failed: %w", err), http.StatusUnprocessableEntity, rw)
return
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(DefaultApiResponse{
Message: fmt.Sprintf("Successfully deleted job %s", id),
})
}
// deleteJobByRequest godoc
// @summary Remove a job from the sql database
// @tags Job remove
// @description Job to delete is specified by request body. All fields are required in this case.
// @accept json
// @produce json
// @param request body api.DeleteJobApiRequest true "All fields required"
// @success 200 {object} api.DefaultApiResponse "Success message"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: finding job failed: sql: no rows in result set"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/delete_job/ [delete]
func (api *RestApi) deleteJobByRequest(rw http.ResponseWriter, r *http.Request) {
// Parse request body
req := DeleteJobApiRequest{}
if err := decode(r.Body, &req); err != nil {
handleError(fmt.Errorf("parsing request body failed: %w", err), http.StatusBadRequest, rw)
return
}
// Fetch job (that will be deleted) from db
var job *schema.Job
var err error
if req.JobId == nil {
handleError(errors.New("the field 'jobId' is required"), http.StatusBadRequest, rw)
return
}
job, err = api.JobRepository.Find(req.JobId, req.Cluster, req.StartTime)
if err != nil {
handleError(fmt.Errorf("finding job failed: %w", err), http.StatusUnprocessableEntity, rw)
return
}
err = api.JobRepository.DeleteJobById(*job.ID)
if err != nil {
handleError(fmt.Errorf("deleting job failed: %w", err), http.StatusUnprocessableEntity, rw)
return
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(DefaultApiResponse{
Message: fmt.Sprintf("Successfully deleted job %d", job.ID),
})
}
// deleteJobBefore godoc
// @summary Remove a job from the sql database
// @tags Job remove
// @description Remove all jobs with start time before timestamp. The jobs will not be removed from the job archive.
// @produce json
// @param ts path int true "Unix epoch timestamp"
// @success 200 {object} api.DefaultApiResponse "Success message"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 404 {object} api.ErrorResponse "Resource not found"
// @failure 422 {object} api.ErrorResponse "Unprocessable Entity: finding job failed: sql: no rows in result set"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/jobs/delete_job_before/{ts} [delete]
func (api *RestApi) deleteJobBefore(rw http.ResponseWriter, r *http.Request) {
var cnt int
// Fetch job (that will be stopped) from db
id, ok := mux.Vars(r)["ts"]
var err error
if ok {
ts, e := strconv.ParseInt(id, 10, 64)
if e != nil {
handleError(fmt.Errorf("integer expected in path for ts: %w", e), http.StatusBadRequest, rw)
return
}
cnt, err = api.JobRepository.DeleteJobsBefore(ts)
} else {
handleError(errors.New("the parameter 'ts' is required"), http.StatusBadRequest, rw)
return
}
if err != nil {
handleError(fmt.Errorf("deleting jobs failed: %w", err), http.StatusUnprocessableEntity, rw)
return
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(DefaultApiResponse{
Message: fmt.Sprintf("Successfully deleted %d jobs", cnt),
})
}
func (api *RestApi) checkAndHandleStopJob(rw http.ResponseWriter, job *schema.Job, req StopJobApiRequest) {
// Sanity checks
if job.State != schema.JobStateRunning {
handleError(fmt.Errorf("jobId %d (id %d) on %s : job has already been stopped (state is: %s)", job.JobID, job.ID, job.Cluster, job.State), http.StatusUnprocessableEntity, rw)
return
}
if job == nil || job.StartTime > req.StopTime {
handleError(fmt.Errorf("jobId %d (id %d) on %s : stopTime %d must be larger/equal than startTime %d", job.JobID, job.ID, job.Cluster, req.StopTime, job.StartTime), http.StatusBadRequest, rw)
return
}
if req.State != "" && !req.State.Valid() {
handleError(fmt.Errorf("jobId %d (id %d) on %s : invalid requested job state: %#v", job.JobID, job.ID, job.Cluster, req.State), http.StatusBadRequest, rw)
return
} else if req.State == "" {
req.State = schema.JobStateCompleted
}
// Mark job as stopped in the database (update state and duration)
job.Duration = int32(req.StopTime - job.StartTime)
job.State = req.State
api.JobRepository.Mutex.Lock()
if err := api.JobRepository.Stop(*job.ID, job.Duration, job.State, job.MonitoringStatus); err != nil {
if err := api.JobRepository.StopCached(*job.ID, job.Duration, job.State, job.MonitoringStatus); err != nil {
api.JobRepository.Mutex.Unlock()
handleError(fmt.Errorf("jobId %d (id %d) on %s : marking job as '%s' (duration: %d) in DB failed: %w", job.JobID, job.ID, job.Cluster, job.State, job.Duration, err), http.StatusInternalServerError, rw)
return
}
}
api.JobRepository.Mutex.Unlock()
cclog.Printf("archiving job... (dbid: %d): cluster=%s, jobId=%d, user=%s, startTime=%d, duration=%d, state=%s", job.ID, job.Cluster, job.JobID, job.User, job.StartTime, job.Duration, job.State)
// Send a response (with status OK). This means that errors that happen from here on forward
// can *NOT* be communicated to the client. If reading from a MetricDataRepository or
// writing to the filesystem fails, the client will not know.
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
json.NewEncoder(rw).Encode(job)
// Monitoring is disabled...
if job.MonitoringStatus == schema.MonitoringStatusDisabled {
return
}
// Trigger async archiving
archiver.TriggerArchiving(job)
}
func (api *RestApi) getJobMetrics(rw http.ResponseWriter, r *http.Request) {
id := mux.Vars(r)["id"]
metrics := r.URL.Query()["metric"]
var scopes []schema.MetricScope
for _, scope := range r.URL.Query()["scope"] {
var s schema.MetricScope
if err := s.UnmarshalGQL(scope); err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
scopes = append(scopes, s)
}
rw.Header().Add("Content-Type", "application/json")
rw.WriteHeader(http.StatusOK)
type Respone struct {
Data *struct {
JobMetrics []*model.JobMetricWithName `json:"jobMetrics"`
} `json:"data"`
Error *struct {
Message string `json:"message"`
} `json:"error"`
}
resolver := graph.GetResolverInstance()
data, err := resolver.Query().JobMetrics(r.Context(), id, metrics, scopes, nil)
if err != nil {
json.NewEncoder(rw).Encode(Respone{
Error: &struct {
Message string "json:\"message\""
}{Message: err.Error()},
})
return
}
json.NewEncoder(rw).Encode(Respone{
Data: &struct {
JobMetrics []*model.JobMetricWithName "json:\"jobMetrics\""
}{JobMetrics: data},
})
}

View File

@@ -1,177 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"bufio"
"encoding/json"
"errors"
"fmt"
"io"
"net/http"
"strconv"
"strings"
"github.com/ClusterCockpit/cc-backend/internal/memorystore"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/influxdata/line-protocol/v2/lineprotocol"
)
// handleFree godoc
// @summary
// @tags free
// @description This endpoint allows the users to free the Buffers from the
// metric store. This endpoint offers the users to remove then systematically
// and also allows then to prune the data under node, if they do not want to
// remove the whole node.
// @produce json
// @param to query string false "up to timestamp"
// @success 200 {string} string "ok"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /free/ [post]
func freeMetrics(rw http.ResponseWriter, r *http.Request) {
rawTo := r.URL.Query().Get("to")
if rawTo == "" {
handleError(errors.New("'to' is a required query parameter"), http.StatusBadRequest, rw)
return
}
to, err := strconv.ParseInt(rawTo, 10, 64)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
// // TODO: lastCheckpoint might be modified by different go-routines.
// // Load it using the sync/atomic package?
// freeUpTo := lastCheckpoint.Unix()
// if to < freeUpTo {
// freeUpTo = to
// }
bodyDec := json.NewDecoder(r.Body)
var selectors [][]string
err = bodyDec.Decode(&selectors)
if err != nil {
http.Error(rw, err.Error(), http.StatusBadRequest)
return
}
ms := memorystore.GetMemoryStore()
n := 0
for _, sel := range selectors {
bn, err := ms.Free(sel, to)
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
n += bn
}
rw.WriteHeader(http.StatusOK)
fmt.Fprintf(rw, "buffers freed: %d\n", n)
}
// handleWrite godoc
// @summary Receive metrics in InfluxDB line-protocol
// @tags write
// @description Write data to the in-memory store in the InfluxDB line-protocol using [this format](https://github.com/ClusterCockpit/cc-specifications/blob/master/metrics/lineprotocol_alternative.md)
// @accept plain
// @produce json
// @param cluster query string false "If the lines in the body do not have a cluster tag, use this value instead."
// @success 200 {string} string "ok"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /write/ [post]
func writeMetrics(rw http.ResponseWriter, r *http.Request) {
bytes, err := io.ReadAll(r.Body)
rw.Header().Add("Content-Type", "application/json")
if err != nil {
handleError(err, http.StatusInternalServerError, rw)
return
}
ms := memorystore.GetMemoryStore()
dec := lineprotocol.NewDecoderWithBytes(bytes)
if err := memorystore.DecodeLine(dec, ms, r.URL.Query().Get("cluster")); err != nil {
cclog.Errorf("/api/write error: %s", err.Error())
handleError(err, http.StatusBadRequest, rw)
return
}
rw.WriteHeader(http.StatusOK)
}
// handleDebug godoc
// @summary Debug endpoint
// @tags debug
// @description This endpoint allows the users to print the content of
// nodes/clusters/metrics to review the state of the data.
// @produce json
// @param selector query string false "Selector"
// @success 200 {string} string "Debug dump"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /debug/ [post]
func debugMetrics(rw http.ResponseWriter, r *http.Request) {
raw := r.URL.Query().Get("selector")
rw.Header().Add("Content-Type", "application/json")
selector := []string{}
if len(raw) != 0 {
selector = strings.Split(raw, ":")
}
ms := memorystore.GetMemoryStore()
if err := ms.DebugDump(bufio.NewWriter(rw), selector); err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
}
// handleHealthCheck godoc
// @summary HealthCheck endpoint
// @tags healthcheck
// @description This endpoint allows the users to check if a node is healthy
// @produce json
// @param selector query string false "Selector"
// @success 200 {string} string "Debug dump"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /healthcheck/ [get]
func metricsHealth(rw http.ResponseWriter, r *http.Request) {
rawCluster := r.URL.Query().Get("cluster")
rawNode := r.URL.Query().Get("node")
if rawCluster == "" || rawNode == "" {
handleError(errors.New("'cluster' and 'node' are required query parameter"), http.StatusBadRequest, rw)
return
}
rw.Header().Add("Content-Type", "application/json")
selector := []string{rawCluster, rawNode}
ms := memorystore.GetMemoryStore()
if err := ms.HealthCheck(bufio.NewWriter(rw), selector); err != nil {
handleError(err, http.StatusBadRequest, rw)
return
}
}

57
internal/api/nats.go Normal file
View File

@@ -0,0 +1,57 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"sync"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
lp "github.com/ClusterCockpit/cc-lib/ccMessage"
"github.com/ClusterCockpit/cc-lib/sinks"
)
type NatsClient struct {
SinkManager sinks.SinkManager
SinkChannel chan lp.CCMessage
}
var (
initOnce sync.Once
ni *NatsClient
)
func Init(wg *sync.WaitGroup) {
initOnce.Do(func() {
ni = &NatsClient{}
var err error
if len(config.Keys.SinkConfigFile) == 0 {
log.Error("Sink configuration file must be set")
return
}
ni.SinkManager, err = sinks.New(wg, config.Keys.SinkConfigFile)
if err != nil {
log.Error(err.Error())
return
}
ni.SinkChannel = make(chan lp.CCMessage, 200)
ni.SinkManager.AddInput(ni.SinkChannel)
ni.SinkManager.Start()
})
}
func Shutdown() {
if ni.SinkManager != nil {
log.Debug("Shutdown SinkManager...")
ni.SinkManager.Close()
}
}
func forwardJob(job schema.BaseJob) {
payload := lp.NewEvent("start_job", nil , meta map[string]string, event string, tm time.Time)
}

View File

@@ -1,80 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"fmt"
"net/http"
"strings"
"time"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-lib/schema"
)
type UpdateNodeStatesRequest struct {
Nodes []schema.NodePayload `json:"nodes"`
Cluster string `json:"cluster" example:"fritz"`
}
// this routine assumes that only one of them exists per node
func determineState(states []string) schema.SchedulerState {
for _, state := range states {
switch strings.ToLower(state) {
case "allocated":
return schema.NodeStateAllocated
case "reserved":
return schema.NodeStateReserved
case "idle":
return schema.NodeStateIdle
case "down":
return schema.NodeStateDown
case "mixed":
return schema.NodeStateMixed
}
}
return schema.NodeStateUnknown
}
// updateNodeStates godoc
// @summary Deliver updated Slurm node states
// @tags Nodestates
// @description Returns a JSON-encoded list of users.
// @description Required query-parameter defines if all users or only users with additional special roles are returned.
// @produce json
// @param request body UpdateNodeStatesRequest true "Request body containing nodes and their states"
// @success 200 {object} api.DefaultApiResponse "Success message"
// @failure 400 {object} api.ErrorResponse "Bad Request"
// @failure 401 {object} api.ErrorResponse "Unauthorized"
// @failure 403 {object} api.ErrorResponse "Forbidden"
// @failure 500 {object} api.ErrorResponse "Internal Server Error"
// @security ApiKeyAuth
// @router /api/nodestats/ [post]
func (api *RestApi) updateNodeStates(rw http.ResponseWriter, r *http.Request) {
// Parse request body
req := UpdateNodeStatesRequest{}
if err := decode(r.Body, &req); err != nil {
handleError(fmt.Errorf("parsing request body failed: %w", err),
http.StatusBadRequest, rw)
return
}
repo := repository.GetNodeRepository()
for _, node := range req.Nodes {
state := determineState(node.States)
nodeState := schema.NodeStateDB{
TimeStamp: time.Now().Unix(), NodeState: state,
CpusAllocated: node.CpusAllocated,
MemoryAllocated: node.MemoryAllocated,
GpusAllocated: node.GpusAllocated,
HealthState: schema.MonitoringStateFull,
JobsRunning: node.JobsRunning,
}
repo.UpdateNodeState(node.Hostname, req.Cluster, &nodeState)
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,159 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package api
import (
"encoding/json"
"fmt"
"net/http"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/gorilla/mux"
)
type ApiReturnedUser struct {
Username string `json:"username"`
Name string `json:"name"`
Roles []string `json:"roles"`
Email string `json:"email"`
Projects []string `json:"projects"`
}
// getUsers godoc
// @summary Returns a list of users
// @tags User
// @description Returns a JSON-encoded list of users.
// @description Required query-parameter defines if all users or only users with additional special roles are returned.
// @produce json
// @param not-just-user query bool true "If returned list should contain all users or only users with additional special roles"
// @success 200 {array} api.ApiReturnedUser "List of users returned successfully"
// @failure 400 {string} string "Bad Request"
// @failure 401 {string} string "Unauthorized"
// @failure 403 {string} string "Forbidden"
// @failure 500 {string} string "Internal Server Error"
// @security ApiKeyAuth
// @router /api/users/ [get]
func (api *RestApi) getUsers(rw http.ResponseWriter, r *http.Request) {
// SecuredCheck() only worked with TokenAuth: Removed
if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
http.Error(rw, "Only admins are allowed to fetch a list of users", http.StatusForbidden)
return
}
users, err := repository.GetUserRepository().ListUsers(r.URL.Query().Get("not-just-user") == "true")
if err != nil {
http.Error(rw, err.Error(), http.StatusInternalServerError)
return
}
json.NewEncoder(rw).Encode(users)
}
func (api *RestApi) updateUser(rw http.ResponseWriter, r *http.Request) {
// SecuredCheck() only worked with TokenAuth: Removed
if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
http.Error(rw, "Only admins are allowed to update a user", http.StatusForbidden)
return
}
// Get Values
newrole := r.FormValue("add-role")
delrole := r.FormValue("remove-role")
newproj := r.FormValue("add-project")
delproj := r.FormValue("remove-project")
// TODO: Handle anything but roles...
if newrole != "" {
if err := repository.GetUserRepository().AddRole(r.Context(), mux.Vars(r)["id"], newrole); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
rw.Write([]byte("Add Role Success"))
} else if delrole != "" {
if err := repository.GetUserRepository().RemoveRole(r.Context(), mux.Vars(r)["id"], delrole); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
rw.Write([]byte("Remove Role Success"))
} else if newproj != "" {
if err := repository.GetUserRepository().AddProject(r.Context(), mux.Vars(r)["id"], newproj); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
rw.Write([]byte("Add Project Success"))
} else if delproj != "" {
if err := repository.GetUserRepository().RemoveProject(r.Context(), mux.Vars(r)["id"], delproj); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
rw.Write([]byte("Remove Project Success"))
} else {
http.Error(rw, "Not Add or Del [role|project]?", http.StatusInternalServerError)
}
}
func (api *RestApi) createUser(rw http.ResponseWriter, r *http.Request) {
// SecuredCheck() only worked with TokenAuth: Removed
rw.Header().Set("Content-Type", "text/plain")
me := repository.GetUserFromContext(r.Context())
if !me.HasRole(schema.RoleAdmin) {
http.Error(rw, "Only admins are allowed to create new users", http.StatusForbidden)
return
}
username, password, role, name, email, project := r.FormValue("username"),
r.FormValue("password"), r.FormValue("role"), r.FormValue("name"),
r.FormValue("email"), r.FormValue("project")
if len(password) == 0 && role != schema.GetRoleString(schema.RoleApi) {
http.Error(rw, "Only API users are allowed to have a blank password (login will be impossible)", http.StatusBadRequest)
return
}
if len(project) != 0 && role != schema.GetRoleString(schema.RoleManager) {
http.Error(rw, "only managers require a project (can be changed later)",
http.StatusBadRequest)
return
} else if len(project) == 0 && role == schema.GetRoleString(schema.RoleManager) {
http.Error(rw, "managers require a project to manage (can be changed later)",
http.StatusBadRequest)
return
}
if err := repository.GetUserRepository().AddUser(&schema.User{
Username: username,
Name: name,
Password: password,
Email: email,
Projects: []string{project},
Roles: []string{role},
}); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
fmt.Fprintf(rw, "User %v successfully created!\n", username)
}
func (api *RestApi) deleteUser(rw http.ResponseWriter, r *http.Request) {
// SecuredCheck() only worked with TokenAuth: Removed
if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
http.Error(rw, "Only admins are allowed to delete a user", http.StatusForbidden)
return
}
username := r.FormValue("username")
if err := repository.GetUserRepository().DelUser(username); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return
}
rw.WriteHeader(http.StatusOK)
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package archiver
@@ -10,8 +10,8 @@ import (
"time"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
@@ -40,8 +40,8 @@ func archivingWorker() {
// not using meta data, called to load JobMeta into Cache?
// will fail if job meta not in repository
if _, err := jobRepo.FetchMetadata(job); err != nil {
cclog.Errorf("archiving job (dbid: %d) failed at check metadata step: %s", job.ID, err.Error())
jobRepo.UpdateMonitoringStatus(*job.ID, schema.MonitoringStatusArchivingFailed)
log.Errorf("archiving job (dbid: %d) failed at check metadata step: %s", job.ID, err.Error())
jobRepo.UpdateMonitoringStatus(job.ID, schema.MonitoringStatusArchivingFailed)
continue
}
@@ -49,34 +49,30 @@ func archivingWorker() {
// TODO: Maybe use context with cancel/timeout here
jobMeta, err := ArchiveJob(job, context.Background())
if err != nil {
cclog.Errorf("archiving job (dbid: %d) failed at archiving job step: %s", job.ID, err.Error())
jobRepo.UpdateMonitoringStatus(*job.ID, schema.MonitoringStatusArchivingFailed)
log.Errorf("archiving job (dbid: %d) failed at archiving job step: %s", job.ID, err.Error())
jobRepo.UpdateMonitoringStatus(job.ID, schema.MonitoringStatusArchivingFailed)
continue
}
stmt := sq.Update("job").Where("job.id = ?", job.ID)
if stmt, err = jobRepo.UpdateFootprint(stmt, jobMeta); err != nil {
cclog.Errorf("archiving job (dbid: %d) failed at update Footprint step: %s", job.ID, err.Error())
log.Errorf("archiving job (dbid: %d) failed at update Footprint step: %s", job.ID, err.Error())
continue
}
if stmt, err = jobRepo.UpdateEnergy(stmt, jobMeta); err != nil {
cclog.Errorf("archiving job (dbid: %d) failed at update Energy step: %s", job.ID, err.Error())
log.Errorf("archiving job (dbid: %d) failed at update Energy step: %s", job.ID, err.Error())
continue
}
// Update the jobs database entry one last time:
stmt = jobRepo.MarkArchived(stmt, schema.MonitoringStatusArchivingSuccessful)
if err := jobRepo.Execute(stmt); err != nil {
cclog.Errorf("archiving job (dbid: %d) failed at db execute: %s", job.ID, err.Error())
log.Errorf("archiving job (dbid: %d) failed at db execute: %s", job.ID, err.Error())
continue
}
cclog.Debugf("archiving job %d took %s", job.JobID, time.Since(start))
cclog.Printf("archiving job (dbid: %d) successful", job.ID)
repository.CallJobStopHooks(job)
log.Debugf("archiving job %d took %s", job.JobID, time.Since(start))
log.Printf("archiving job (dbid: %d) successful", job.ID)
archivePending.Done()
default:
continue
}
}
}
@@ -84,7 +80,7 @@ func archivingWorker() {
// Trigger async archiving
func TriggerArchiving(job *schema.Job) {
if archiveChannel == nil {
cclog.Fatal("Cannot archive without archiving channel. Did you Start the archiver?")
log.Fatal("Cannot archive without archiving channel. Did you Start the archiver?")
}
archivePending.Add(1)

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package archiver
@@ -11,12 +11,12 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
// Writes a running job to the job-archive
func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.Job, error) {
func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.JobMeta, error) {
allMetrics := make([]string, 0)
metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
for _, mc := range metricConfigs {
@@ -36,11 +36,15 @@ func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.Job, error) {
jobData, err := metricDataDispatcher.LoadData(job, allMetrics, scopes, ctx, 0) // 0 Resulotion-Value retrieves highest res (60s)
if err != nil {
cclog.Error("Error wile loading job data for archiving")
log.Error("Error wile loading job data for archiving")
return nil, err
}
job.Statistics = make(map[string]schema.JobStatistics)
jobMeta := &schema.JobMeta{
BaseJob: job.BaseJob,
StartTime: job.StartTime.Unix(),
Statistics: make(map[string]schema.JobStatistics),
}
for metric, data := range jobData {
avg, min, max := 0.0, math.MaxFloat32, -math.MaxFloat32
@@ -56,13 +60,12 @@ func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.Job, error) {
max = math.Max(max, series.Statistics.Max)
}
// Round AVG Result to 2 Digits
job.Statistics[metric] = schema.JobStatistics{
jobMeta.Statistics[metric] = schema.JobStatistics{
Unit: schema.Unit{
Prefix: archive.GetMetricConfig(job.Cluster, metric).Unit.Prefix,
Base: archive.GetMetricConfig(job.Cluster, metric).Unit.Base,
},
Avg: (math.Round((avg/float64(job.NumNodes))*100) / 100),
Avg: avg / float64(job.NumNodes),
Min: min,
Max: max,
}
@@ -72,8 +75,8 @@ func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.Job, error) {
// only return the JobMeta structure as the
// statistics in there are needed.
if config.Keys.DisableArchive {
return job, nil
return jobMeta, nil
}
return job, archive.GetHandle().ImportJob(job, &jobData)
return jobMeta, archive.GetHandle().ImportJob(jobMeta, &jobData)
}

View File

@@ -1,24 +1,18 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
// Package auth implements various authentication methods
package auth
import (
"bytes"
"context"
"crypto/rand"
"database/sql"
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"net"
"net/http"
"os"
"strings"
"sync"
"time"
@@ -26,9 +20,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-lib/util"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/gorilla/sessions"
)
@@ -55,14 +48,6 @@ func getIPUserLimiter(ip, username string) *rate.Limiter {
return limiter.(*rate.Limiter)
}
type AuthConfig struct {
LdapConfig *LdapConfig `json:"ldap"`
JwtConfig *JWTAuthConfig `json:"jwts"`
OpenIDConfig *OpenIDConfig `json:"oidc"`
}
var Keys AuthConfig
type Authentication struct {
sessionStore *sessions.CookieStore
LdapAuth *LdapAuthenticator
@@ -78,7 +63,7 @@ func (auth *Authentication) AuthViaSession(
) (*schema.User, error) {
session, err := auth.sessionStore.Get(r, "session")
if err != nil {
cclog.Error("Error while getting session store")
log.Error("Error while getting session store")
return nil, err
}
@@ -99,22 +84,22 @@ func (auth *Authentication) AuthViaSession(
}, nil
}
func Init(authCfg *json.RawMessage) {
func Init() {
initOnce.Do(func() {
authInstance = &Authentication{}
sessKey := os.Getenv("SESSION_KEY")
if sessKey == "" {
cclog.Warn("environment variable 'SESSION_KEY' not set (will use non-persistent random key)")
log.Warn("environment variable 'SESSION_KEY' not set (will use non-persistent random key)")
bytes := make([]byte, 32)
if _, err := rand.Read(bytes); err != nil {
cclog.Fatal("Error while initializing authentication -> failed to generate random bytes for session key")
log.Fatal("Error while initializing authentication -> failed to generate random bytes for session key")
}
authInstance.sessionStore = sessions.NewCookieStore(bytes)
} else {
bytes, err := base64.StdEncoding.DecodeString(sessKey)
if err != nil {
cclog.Fatal("Error while initializing authentication -> decoding session key failed")
log.Fatal("Error while initializing authentication -> decoding session key failed")
}
authInstance.sessionStore = sessions.NewCookieStore(bytes)
}
@@ -123,55 +108,44 @@ func Init(authCfg *json.RawMessage) {
authInstance.SessionMaxAge = d
}
if authCfg == nil {
return
}
config.Validate(configSchema, *authCfg)
dec := json.NewDecoder(bytes.NewReader(*authCfg))
dec.DisallowUnknownFields()
if err := dec.Decode(&Keys); err != nil {
cclog.Errorf("error while decoding ldap config: %v", err)
}
if Keys.LdapConfig != nil {
if config.Keys.LdapConfig != nil {
ldapAuth := &LdapAuthenticator{}
if err := ldapAuth.Init(); err != nil {
cclog.Warn("Error while initializing authentication -> ldapAuth init failed")
log.Warn("Error while initializing authentication -> ldapAuth init failed")
} else {
authInstance.LdapAuth = ldapAuth
authInstance.authenticators = append(authInstance.authenticators, authInstance.LdapAuth)
}
} else {
cclog.Info("Missing LDAP configuration: No LDAP support!")
log.Info("Missing LDAP configuration: No LDAP support!")
}
if Keys.JwtConfig != nil {
if config.Keys.JwtConfig != nil {
authInstance.JwtAuth = &JWTAuthenticator{}
if err := authInstance.JwtAuth.Init(); err != nil {
cclog.Fatal("Error while initializing authentication -> jwtAuth init failed")
log.Fatal("Error while initializing authentication -> jwtAuth init failed")
}
jwtSessionAuth := &JWTSessionAuthenticator{}
if err := jwtSessionAuth.Init(); err != nil {
cclog.Info("jwtSessionAuth init failed: No JWT login support!")
log.Info("jwtSessionAuth init failed: No JWT login support!")
} else {
authInstance.authenticators = append(authInstance.authenticators, jwtSessionAuth)
}
jwtCookieSessionAuth := &JWTCookieSessionAuthenticator{}
if err := jwtCookieSessionAuth.Init(); err != nil {
cclog.Info("jwtCookieSessionAuth init failed: No JWT cookie login support!")
log.Info("jwtCookieSessionAuth init failed: No JWT cookie login support!")
} else {
authInstance.authenticators = append(authInstance.authenticators, jwtCookieSessionAuth)
}
} else {
cclog.Info("Missing JWT configuration: No JWT token support!")
log.Info("Missing JWT configuration: No JWT token support!")
}
authInstance.LocalAuth = &LocalAuthenticator{}
if err := authInstance.LocalAuth.Init(); err != nil {
cclog.Fatal("Error while initializing authentication -> localAuth init failed")
log.Fatal("Error while initializing authentication -> localAuth init failed")
}
authInstance.authenticators = append(authInstance.authenticators, authInstance.LocalAuth)
})
@@ -179,7 +153,7 @@ func Init(authCfg *json.RawMessage) {
func GetAuthInstance() *Authentication {
if authInstance == nil {
cclog.Fatal("Authentication module not initialized!")
log.Fatal("Authentication module not initialized!")
}
return authInstance
@@ -190,14 +164,14 @@ func handleTokenUser(tokenUser *schema.User) {
dbUser, err := r.GetUser(tokenUser.Username)
if err != nil && err != sql.ErrNoRows {
cclog.Errorf("Error while loading user '%s': %v", tokenUser.Username, err)
} else if err == sql.ErrNoRows && Keys.JwtConfig.SyncUserOnLogin { // Adds New User
log.Errorf("Error while loading user '%s': %v", tokenUser.Username, err)
} else if err == sql.ErrNoRows && config.Keys.JwtConfig.SyncUserOnLogin { // Adds New User
if err := r.AddUser(tokenUser); err != nil {
cclog.Errorf("Error while adding user '%s' to DB: %v", tokenUser.Username, err)
log.Errorf("Error while adding user '%s' to DB: %v", tokenUser.Username, err)
}
} else if err == nil && Keys.JwtConfig.UpdateUserOnLogin { // Update Existing User
} else if err == nil && config.Keys.JwtConfig.UpdateUserOnLogin { // Update Existing User
if err := r.UpdateUser(dbUser, tokenUser); err != nil {
cclog.Errorf("Error while updating user '%s' to DB: %v", dbUser.Username, err)
log.Errorf("Error while updating user '%s' to DB: %v", dbUser.Username, err)
}
}
}
@@ -207,14 +181,14 @@ func handleOIDCUser(OIDCUser *schema.User) {
dbUser, err := r.GetUser(OIDCUser.Username)
if err != nil && err != sql.ErrNoRows {
cclog.Errorf("Error while loading user '%s': %v", OIDCUser.Username, err)
} else if err == sql.ErrNoRows && Keys.OpenIDConfig.SyncUserOnLogin { // Adds New User
log.Errorf("Error while loading user '%s': %v", OIDCUser.Username, err)
} else if err == sql.ErrNoRows && config.Keys.OpenIDConfig.SyncUserOnLogin { // Adds New User
if err := r.AddUser(OIDCUser); err != nil {
cclog.Errorf("Error while adding user '%s' to DB: %v", OIDCUser.Username, err)
log.Errorf("Error while adding user '%s' to DB: %v", OIDCUser.Username, err)
}
} else if err == nil && Keys.OpenIDConfig.UpdateUserOnLogin { // Update Existing User
} else if err == nil && config.Keys.OpenIDConfig.UpdateUserOnLogin { // Update Existing User
if err := r.UpdateUser(dbUser, OIDCUser); err != nil {
cclog.Errorf("Error while updating user '%s' to DB: %v", dbUser.Username, err)
log.Errorf("Error while updating user '%s' to DB: %v", dbUser.Username, err)
}
}
}
@@ -222,7 +196,7 @@ func handleOIDCUser(OIDCUser *schema.User) {
func (auth *Authentication) SaveSession(rw http.ResponseWriter, r *http.Request, user *schema.User) error {
session, err := auth.sessionStore.New(r, "session")
if err != nil {
cclog.Errorf("session creation failed: %s", err.Error())
log.Errorf("session creation failed: %s", err.Error())
http.Error(rw, err.Error(), http.StatusInternalServerError)
return err
}
@@ -230,7 +204,7 @@ func (auth *Authentication) SaveSession(rw http.ResponseWriter, r *http.Request,
if auth.SessionMaxAge != 0 {
session.Options.MaxAge = int(auth.SessionMaxAge.Seconds())
}
if config.Keys.HTTPSCertFile == "" && config.Keys.HTTPSKeyFile == "" {
if config.Keys.HttpsCertFile == "" && config.Keys.HttpsKeyFile == "" {
session.Options.Secure = false
}
session.Options.SameSite = http.SameSiteStrictMode
@@ -238,7 +212,7 @@ func (auth *Authentication) SaveSession(rw http.ResponseWriter, r *http.Request,
session.Values["projects"] = user.Projects
session.Values["roles"] = user.Roles
if err := auth.sessionStore.Save(r, rw, session); err != nil {
cclog.Warnf("session save failed: %s", err.Error())
log.Warnf("session save failed: %s", err.Error())
http.Error(rw, err.Error(), http.StatusInternalServerError)
return err
}
@@ -259,9 +233,9 @@ func (auth *Authentication) Login(
limiter := getIPUserLimiter(ip, username)
if !limiter.Allow() {
cclog.Warnf("AUTH/RATE > Too many login attempts for combination IP: %s, Username: %s", ip, username)
onfailure(rw, r, errors.New("too many login attempts, try again in a few minutes"))
return
log.Warnf("AUTH/RATE > Too many login attempts for combination IP: %s, Username: %s", ip, username)
onfailure(rw, r, errors.New("Too many login attempts, try again in a few minutes."))
return
}
var dbUser *schema.User
@@ -269,7 +243,7 @@ func (auth *Authentication) Login(
var err error
dbUser, err = repository.GetUserRepository().GetUser(username)
if err != nil && err != sql.ErrNoRows {
cclog.Errorf("Error while loading user '%v'", username)
log.Errorf("Error while loading user '%v'", username)
}
}
@@ -279,12 +253,12 @@ func (auth *Authentication) Login(
if user, ok = authenticator.CanLogin(dbUser, username, rw, r); !ok {
continue
} else {
cclog.Debugf("Can login with user %v", user)
log.Debugf("Can login with user %v", user)
}
user, err := authenticator.Login(user, rw, r)
if err != nil {
cclog.Warnf("user login failed: %s", err.Error())
log.Warnf("user login failed: %s", err.Error())
onfailure(rw, r, err)
return
}
@@ -293,7 +267,7 @@ func (auth *Authentication) Login(
return
}
cclog.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
log.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
if r.FormValue("redirect") != "" {
@@ -305,7 +279,7 @@ func (auth *Authentication) Login(
return
}
cclog.Debugf("login failed: no authenticator applied")
log.Debugf("login failed: no authenticator applied")
onfailure(rw, r, errors.New("no authenticator applied"))
})
}
@@ -317,14 +291,14 @@ func (auth *Authentication) Auth(
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.JwtAuth.AuthViaJWT(rw, r)
if err != nil {
cclog.Infof("auth -> authentication failed: %s", err.Error())
log.Infof("auth -> authentication failed: %s", err.Error())
http.Error(rw, err.Error(), http.StatusUnauthorized)
return
}
if user == nil {
user, err = auth.AuthViaSession(rw, r)
if err != nil {
cclog.Infof("auth -> authentication failed: %s", err.Error())
log.Infof("auth -> authentication failed: %s", err.Error())
http.Error(rw, err.Error(), http.StatusUnauthorized)
return
}
@@ -335,30 +309,22 @@ func (auth *Authentication) Auth(
return
}
cclog.Info("auth -> authentication failed")
log.Info("auth -> authentication failed")
onfailure(rw, r, errors.New("unauthorized (please login first)"))
})
}
func (auth *Authentication) AuthAPI(
func (auth *Authentication) AuthApi(
onsuccess http.Handler,
onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
) http.Handler {
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.JwtAuth.AuthViaJWT(rw, r)
if err != nil {
cclog.Infof("auth api -> authentication failed: %s", err.Error())
log.Infof("auth api -> authentication failed: %s", err.Error())
onfailure(rw, r, err)
return
}
ipErr := securedCheck(user, r)
if ipErr != nil {
cclog.Infof("auth api -> secured check failed: %s", ipErr.Error())
onfailure(rw, r, ipErr)
return
}
if user != nil {
switch {
case len(user.Roles) == 1:
@@ -374,63 +340,26 @@ func (auth *Authentication) AuthAPI(
return
}
default:
cclog.Info("auth api -> authentication failed: missing role")
log.Info("auth api -> authentication failed: missing role")
onfailure(rw, r, errors.New("unauthorized"))
}
}
cclog.Info("auth api -> authentication failed: no auth")
log.Info("auth api -> authentication failed: no auth")
onfailure(rw, r, errors.New("unauthorized"))
})
}
func (auth *Authentication) AuthUserAPI(
func (auth *Authentication) AuthUserApi(
onsuccess http.Handler,
onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
) http.Handler {
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.JwtAuth.AuthViaJWT(rw, r)
if err != nil {
cclog.Infof("auth user api -> authentication failed: %s", err.Error())
log.Infof("auth user api -> authentication failed: %s", err.Error())
onfailure(rw, r, err)
return
}
if user != nil {
switch {
case len(user.Roles) == 1:
if user.HasRole(schema.RoleApi) {
ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
onsuccess.ServeHTTP(rw, r.WithContext(ctx))
return
}
case len(user.Roles) >= 2:
if user.HasRole(schema.RoleApi) && user.HasAnyRole([]schema.Role{schema.RoleUser, schema.RoleManager, schema.RoleSupport, schema.RoleAdmin}) {
ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
onsuccess.ServeHTTP(rw, r.WithContext(ctx))
return
}
default:
cclog.Info("auth user api -> authentication failed: missing role")
onfailure(rw, r, errors.New("unauthorized"))
}
}
cclog.Info("auth user api -> authentication failed: no auth")
onfailure(rw, r, errors.New("unauthorized"))
})
}
func (auth *Authentication) AuthMetricStoreAPI(
onsuccess http.Handler,
onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
) http.Handler {
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.JwtAuth.AuthViaJWT(rw, r)
if err != nil {
cclog.Infof("auth metricstore api -> authentication failed: %s", err.Error())
onfailure(rw, r, err)
return
}
if user != nil {
switch {
case len(user.Roles) == 1:
@@ -446,23 +375,23 @@ func (auth *Authentication) AuthMetricStoreAPI(
return
}
default:
cclog.Info("auth metricstore api -> authentication failed: missing role")
log.Info("auth user api -> authentication failed: missing role")
onfailure(rw, r, errors.New("unauthorized"))
}
}
cclog.Info("auth metricstore api -> authentication failed: no auth")
log.Info("auth user api -> authentication failed: no auth")
onfailure(rw, r, errors.New("unauthorized"))
})
}
func (auth *Authentication) AuthConfigAPI(
func (auth *Authentication) AuthConfigApi(
onsuccess http.Handler,
onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
) http.Handler {
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.AuthViaSession(rw, r)
if err != nil {
cclog.Infof("auth config api -> authentication failed: %s", err.Error())
log.Infof("auth config api -> authentication failed: %s", err.Error())
onfailure(rw, r, err)
return
}
@@ -471,19 +400,19 @@ func (auth *Authentication) AuthConfigAPI(
onsuccess.ServeHTTP(rw, r.WithContext(ctx))
return
}
cclog.Info("auth config api -> authentication failed: no auth")
log.Info("auth config api -> authentication failed: no auth")
onfailure(rw, r, errors.New("unauthorized"))
})
}
func (auth *Authentication) AuthFrontendAPI(
func (auth *Authentication) AuthFrontendApi(
onsuccess http.Handler,
onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
) http.Handler {
return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
user, err := auth.AuthViaSession(rw, r)
if err != nil {
cclog.Infof("auth frontend api -> authentication failed: %s", err.Error())
log.Infof("auth frontend api -> authentication failed: %s", err.Error())
onfailure(rw, r, err)
return
}
@@ -492,7 +421,7 @@ func (auth *Authentication) AuthFrontendAPI(
onsuccess.ServeHTTP(rw, r.WithContext(ctx))
return
}
cclog.Info("auth frontend api -> authentication failed: no auth")
log.Info("auth frontend api -> authentication failed: no auth")
onfailure(rw, r, errors.New("unauthorized"))
})
}
@@ -516,39 +445,3 @@ func (auth *Authentication) Logout(onsuccess http.Handler) http.Handler {
onsuccess.ServeHTTP(rw, r)
})
}
// Helper Moved To MiddleWare Auth Handlers
func securedCheck(user *schema.User, r *http.Request) error {
if user == nil {
return fmt.Errorf("no user for secured check")
}
// extract IP address for checking
IPAddress := r.Header.Get("X-Real-Ip")
if IPAddress == "" {
IPAddress = r.Header.Get("X-Forwarded-For")
}
if IPAddress == "" {
IPAddress = r.RemoteAddr
}
// FIXME: IPV6 not handled
if strings.Contains(IPAddress, ":") {
IPAddress = strings.Split(IPAddress, ":")[0]
}
// If nothing declared in config: deny all request to this api endpoint
if len(config.Keys.APIAllowedIPs) == 0 {
return fmt.Errorf("missing configuration key ApiAllowedIPs")
}
// If wildcard declared in config: Continue
if config.Keys.APIAllowedIPs[0] == "*" {
return nil
}
// check if IP is allowed
if !util.Contains(config.Keys.APIAllowedIPs, IPAddress) {
return fmt.Errorf("unknown ip: %v", IPAddress)
}
return nil
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
@@ -14,34 +13,13 @@ import (
"strings"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/golang-jwt/jwt/v5"
)
type JWTAuthConfig struct {
// Specifies for how long a JWT token shall be valid
// as a string parsable by time.ParseDuration().
MaxAge string `json:"max-age"`
// Specifies which cookie should be checked for a JWT token (if no authorization header is present)
CookieName string `json:"cookieName"`
// Deny login for users not in database (but defined in JWT).
// Ignore user roles defined in JWTs ('roles' claim), get them from db.
ValidateUser bool `json:"validateUser"`
// Specifies which issuer should be accepted when validating external JWTs ('iss' claim)
TrustedIssuer string `json:"trustedIssuer"`
// Should an non-existent user be added to the DB based on the information in the token
SyncUserOnLogin bool `json:"syncUserOnLogin"`
// Should an existent user be updated in the DB based on the information in the token
UpdateUserOnLogin bool `json:"updateUserOnLogin"`
}
type JWTAuthenticator struct {
publicKey ed25519.PublicKey
privateKey ed25519.PrivateKey
@@ -50,17 +28,17 @@ type JWTAuthenticator struct {
func (ja *JWTAuthenticator) Init() error {
pubKey, privKey := os.Getenv("JWT_PUBLIC_KEY"), os.Getenv("JWT_PRIVATE_KEY")
if pubKey == "" || privKey == "" {
cclog.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
log.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
} else {
bytes, err := base64.StdEncoding.DecodeString(pubKey)
if err != nil {
cclog.Warn("Could not decode JWT public key")
log.Warn("Could not decode JWT public key")
return err
}
ja.publicKey = ed25519.PublicKey(bytes)
bytes, err = base64.StdEncoding.DecodeString(privKey)
if err != nil {
cclog.Warn("Could not decode JWT private key")
log.Warn("Could not decode JWT private key")
return err
}
ja.privateKey = ed25519.PrivateKey(bytes)
@@ -84,7 +62,7 @@ func (ja *JWTAuthenticator) AuthViaJWT(
return nil, nil
}
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (interface{}, error) {
if t.Method != jwt.SigningMethodEdDSA {
return nil, errors.New("only Ed25519/EdDSA supported")
}
@@ -92,11 +70,11 @@ func (ja *JWTAuthenticator) AuthViaJWT(
return ja.publicKey, nil
})
if err != nil {
cclog.Warn("Error while parsing JWT token")
log.Warn("Error while parsing JWT token")
return nil, err
}
if !token.Valid {
cclog.Warn("jwt token claims are not valid")
log.Warn("jwt token claims are not valid")
return nil, errors.New("jwt token claims are not valid")
}
@@ -107,19 +85,19 @@ func (ja *JWTAuthenticator) AuthViaJWT(
var roles []string
// Validate user + roles from JWT against database?
if Keys.JwtConfig.ValidateUser {
if config.Keys.JwtConfig.ValidateUser {
ur := repository.GetUserRepository()
user, err := ur.GetUser(sub)
// Deny any logins for unknown usernames
if err != nil {
cclog.Warn("Could not find user from JWT in internal database.")
log.Warn("Could not find user from JWT in internal database.")
return nil, errors.New("unknown user")
}
// Take user roles from database instead of trusting the JWT
roles = user.Roles
} else {
// Extract roles from JWT (if present)
if rawroles, ok := claims["roles"].([]any); ok {
if rawroles, ok := claims["roles"].([]interface{}); ok {
for _, rr := range rawroles {
if r, ok := rr.(string); ok {
roles = append(roles, r)
@@ -136,7 +114,7 @@ func (ja *JWTAuthenticator) AuthViaJWT(
}, nil
}
// ProvideJWT generates a new JWT that can be used for authentication
// Generate a new JWT that can be used for authentication
func (ja *JWTAuthenticator) ProvideJWT(user *schema.User) (string, error) {
if ja.privateKey == nil {
return "", errors.New("environment variable 'JWT_PRIVATE_KEY' not set")
@@ -148,8 +126,8 @@ func (ja *JWTAuthenticator) ProvideJWT(user *schema.User) (string, error) {
"roles": user.Roles,
"iat": now.Unix(),
}
if Keys.JwtConfig.MaxAge != "" {
d, err := time.ParseDuration(Keys.JwtConfig.MaxAge)
if config.Keys.JwtConfig.MaxAge != "" {
d, err := time.ParseDuration(config.Keys.JwtConfig.MaxAge)
if err != nil {
return "", errors.New("cannot parse max-age config key")
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
@@ -14,9 +13,10 @@ import (
"net/http"
"os"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/golang-jwt/jwt/v5"
)
@@ -31,18 +31,18 @@ var _ Authenticator = (*JWTCookieSessionAuthenticator)(nil)
func (ja *JWTCookieSessionAuthenticator) Init() error {
pubKey, privKey := os.Getenv("JWT_PUBLIC_KEY"), os.Getenv("JWT_PRIVATE_KEY")
if pubKey == "" || privKey == "" {
cclog.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
log.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
return errors.New("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
} else {
bytes, err := base64.StdEncoding.DecodeString(pubKey)
if err != nil {
cclog.Warn("Could not decode JWT public key")
log.Warn("Could not decode JWT public key")
return err
}
ja.publicKey = ed25519.PublicKey(bytes)
bytes, err = base64.StdEncoding.DecodeString(privKey)
if err != nil {
cclog.Warn("Could not decode JWT private key")
log.Warn("Could not decode JWT private key")
return err
}
ja.privateKey = ed25519.PrivateKey(bytes)
@@ -53,35 +53,36 @@ func (ja *JWTCookieSessionAuthenticator) Init() error {
if keyFound && pubKeyCrossLogin != "" {
bytes, err := base64.StdEncoding.DecodeString(pubKeyCrossLogin)
if err != nil {
cclog.Warn("Could not decode cross login JWT public key")
log.Warn("Could not decode cross login JWT public key")
return err
}
ja.publicKeyCrossLogin = ed25519.PublicKey(bytes)
} else {
ja.publicKeyCrossLogin = nil
cclog.Debug("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
log.Debug("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
return errors.New("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
}
jc := config.Keys.JwtConfig
// Warn if other necessary settings are not configured
if Keys.JwtConfig != nil {
if Keys.JwtConfig.CookieName == "" {
cclog.Info("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
if jc != nil {
if jc.CookieName == "" {
log.Info("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
return errors.New("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
}
if !Keys.JwtConfig.ValidateUser {
cclog.Info("forceJWTValidationViaDatabase not set to true: CC will accept users and roles defined in JWTs regardless of its own database!")
if !jc.ValidateUser {
log.Info("forceJWTValidationViaDatabase not set to true: CC will accept users and roles defined in JWTs regardless of its own database!")
}
if Keys.JwtConfig.TrustedIssuer == "" {
cclog.Info("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
if jc.TrustedIssuer == "" {
log.Info("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
return errors.New("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
}
} else {
cclog.Warn("config for JWTs not configured (cross login via JWT cookie will fail)")
log.Warn("config for JWTs not configured (cross login via JWT cookie will fail)")
return errors.New("config for JWTs not configured (cross login via JWT cookie will fail)")
}
cclog.Info("JWT Cookie Session authenticator successfully registered")
log.Info("JWT Cookie Session authenticator successfully registered")
return nil
}
@@ -91,7 +92,7 @@ func (ja *JWTCookieSessionAuthenticator) CanLogin(
rw http.ResponseWriter,
r *http.Request,
) (*schema.User, bool) {
jc := Keys.JwtConfig
jc := config.Keys.JwtConfig
cookieName := ""
if jc.CookieName != "" {
cookieName = jc.CookieName
@@ -114,7 +115,7 @@ func (ja *JWTCookieSessionAuthenticator) Login(
rw http.ResponseWriter,
r *http.Request,
) (*schema.User, error) {
jc := Keys.JwtConfig
jc := config.Keys.JwtConfig
jwtCookie, err := r.Cookie(jc.CookieName)
var rawtoken string
@@ -122,7 +123,7 @@ func (ja *JWTCookieSessionAuthenticator) Login(
rawtoken = jwtCookie.Value
}
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (interface{}, error) {
if t.Method != jwt.SigningMethodEdDSA {
return nil, errors.New("only Ed25519/EdDSA supported")
}
@@ -139,12 +140,12 @@ func (ja *JWTCookieSessionAuthenticator) Login(
return ja.publicKey, nil
})
if err != nil {
cclog.Warn("JWT cookie session: error while parsing token")
log.Warn("JWT cookie session: error while parsing token")
return nil, err
}
if !token.Valid {
cclog.Warn("jwt token claims are not valid")
log.Warn("jwt token claims are not valid")
return nil, errors.New("jwt token claims are not valid")
}
@@ -158,18 +159,18 @@ func (ja *JWTCookieSessionAuthenticator) Login(
var err error
user, err = repository.GetUserRepository().GetUser(sub)
if err != nil && err != sql.ErrNoRows {
cclog.Errorf("Error while loading user '%v'", sub)
log.Errorf("Error while loading user '%v'", sub)
}
// Deny any logins for unknown usernames
if user == nil {
cclog.Warn("Could not find user from JWT in internal database.")
log.Warn("Could not find user from JWT in internal database.")
return nil, errors.New("unknown user")
}
} else {
var name string
if wrap, ok := claims["name"].(map[string]any); ok {
if vals, ok := wrap["values"].([]any); ok {
if wrap, ok := claims["name"].(map[string]interface{}); ok {
if vals, ok := wrap["values"].([]interface{}); ok {
if len(vals) != 0 {
name = fmt.Sprintf("%v", vals[0])
@@ -181,7 +182,7 @@ func (ja *JWTCookieSessionAuthenticator) Login(
}
// Extract roles from JWT (if present)
if rawroles, ok := claims["roles"].([]any); ok {
if rawroles, ok := claims["roles"].([]interface{}); ok {
for _, rr := range rawroles {
if r, ok := rr.(string); ok {
roles = append(roles, r)

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
@@ -14,9 +13,10 @@ import (
"os"
"strings"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/golang-jwt/jwt/v5"
)
@@ -30,13 +30,13 @@ func (ja *JWTSessionAuthenticator) Init() error {
if pubKey := os.Getenv("CROSS_LOGIN_JWT_HS512_KEY"); pubKey != "" {
bytes, err := base64.StdEncoding.DecodeString(pubKey)
if err != nil {
cclog.Warn("Could not decode cross login JWT HS512 key")
log.Warn("Could not decode cross login JWT HS512 key")
return err
}
ja.loginTokenKey = bytes
}
cclog.Info("JWT Session authenticator successfully registered")
log.Info("JWT Session authenticator successfully registered")
return nil
}
@@ -60,19 +60,19 @@ func (ja *JWTSessionAuthenticator) Login(
rawtoken = r.URL.Query().Get("login-token")
}
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (interface{}, error) {
if t.Method == jwt.SigningMethodHS256 || t.Method == jwt.SigningMethodHS512 {
return ja.loginTokenKey, nil
}
return nil, fmt.Errorf("unkown signing method for login token: %s (known: HS256, HS512, EdDSA)", t.Method.Alg())
})
if err != nil {
cclog.Warn("Error while parsing jwt token")
log.Warn("Error while parsing jwt token")
return nil, err
}
if !token.Valid {
cclog.Warn("jwt token claims are not valid")
log.Warn("jwt token claims are not valid")
return nil, errors.New("jwt token claims are not valid")
}
@@ -82,22 +82,22 @@ func (ja *JWTSessionAuthenticator) Login(
var roles []string
projects := make([]string, 0)
if Keys.JwtConfig.ValidateUser {
if config.Keys.JwtConfig.ValidateUser {
var err error
user, err = repository.GetUserRepository().GetUser(sub)
if err != nil && err != sql.ErrNoRows {
cclog.Errorf("Error while loading user '%v'", sub)
log.Errorf("Error while loading user '%v'", sub)
}
// Deny any logins for unknown usernames
if user == nil {
cclog.Warn("Could not find user from JWT in internal database.")
log.Warn("Could not find user from JWT in internal database.")
return nil, errors.New("unknown user")
}
} else {
var name string
if wrap, ok := claims["name"].(map[string]any); ok {
if vals, ok := wrap["values"].([]any); ok {
if wrap, ok := claims["name"].(map[string]interface{}); ok {
if vals, ok := wrap["values"].([]interface{}); ok {
if len(vals) != 0 {
name = fmt.Sprintf("%v", vals[0])
@@ -109,7 +109,7 @@ func (ja *JWTSessionAuthenticator) Login(
}
// Extract roles from JWT (if present)
if rawroles, ok := claims["roles"].([]any); ok {
if rawroles, ok := claims["roles"].([]interface{}); ok {
for _, rr := range rawroles {
if r, ok := rr.(string); ok {
if schema.IsValidRole(r) {
@@ -119,7 +119,7 @@ func (ja *JWTSessionAuthenticator) Login(
}
}
if rawprojs, ok := claims["projects"].([]any); ok {
if rawprojs, ok := claims["projects"].([]interface{}); ok {
for _, pp := range rawprojs {
if p, ok := pp.(string); ok {
projects = append(projects, p)
@@ -138,7 +138,7 @@ func (ja *JWTSessionAuthenticator) Login(
AuthSource: schema.AuthViaToken,
}
if Keys.JwtConfig.SyncUserOnLogin || Keys.JwtConfig.UpdateUserOnLogin {
if config.Keys.JwtConfig.SyncUserOnLogin || config.Keys.JwtConfig.UpdateUserOnLogin {
handleTokenUser(user)
}
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
@@ -12,26 +11,13 @@ import (
"os"
"strings"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/go-ldap/ldap/v3"
)
type LdapConfig struct {
URL string `json:"url"`
UserBase string `json:"user_base"`
SearchDN string `json:"search_dn"`
UserBind string `json:"user_bind"`
UserFilter string `json:"user_filter"`
UserAttr string `json:"username_attr"`
SyncInterval string `json:"sync_interval"` // Parsed using time.ParseDuration.
SyncDelOldUsers bool `json:"sync_del_old_users"`
// Should an non-existent user be added to the DB if user exists in ldap directory
SyncUserOnLogin bool `json:"syncUserOnLogin"`
}
type LdapAuthenticator struct {
syncPassword string
UserAttr string
@@ -42,11 +28,13 @@ var _ Authenticator = (*LdapAuthenticator)(nil)
func (la *LdapAuthenticator) Init() error {
la.syncPassword = os.Getenv("LDAP_ADMIN_PASSWORD")
if la.syncPassword == "" {
cclog.Warn("environment variable 'LDAP_ADMIN_PASSWORD' not set (ldap sync will not work)")
log.Warn("environment variable 'LDAP_ADMIN_PASSWORD' not set (ldap sync will not work)")
}
if Keys.LdapConfig.UserAttr != "" {
la.UserAttr = Keys.LdapConfig.UserAttr
lc := config.Keys.LdapConfig
if lc.UserAttr != "" {
la.UserAttr = lc.UserAttr
} else {
la.UserAttr = "gecos"
}
@@ -60,7 +48,7 @@ func (la *LdapAuthenticator) CanLogin(
rw http.ResponseWriter,
r *http.Request,
) (*schema.User, bool) {
lc := Keys.LdapConfig
lc := config.Keys.LdapConfig
if user != nil {
if user.AuthSource == schema.AuthViaLDAP {
@@ -70,7 +58,7 @@ func (la *LdapAuthenticator) CanLogin(
if lc.SyncUserOnLogin {
l, err := la.getLdapConnection(true)
if err != nil {
cclog.Error("LDAP connection error")
log.Error("LDAP connection error")
}
defer l.Close()
@@ -83,12 +71,12 @@ func (la *LdapAuthenticator) CanLogin(
sr, err := l.Search(searchRequest)
if err != nil {
cclog.Warn(err)
log.Warn(err)
return nil, false
}
if len(sr.Entries) != 1 {
cclog.Warn("LDAP: User does not exist or too many entries returned")
log.Warn("LDAP: User does not exist or too many entries returned")
return nil, false
}
@@ -108,7 +96,7 @@ func (la *LdapAuthenticator) CanLogin(
}
if err := repository.GetUserRepository().AddUser(user); err != nil {
cclog.Errorf("User '%s' LDAP: Insert into DB failed", username)
log.Errorf("User '%s' LDAP: Insert into DB failed", username)
return nil, false
}
@@ -126,14 +114,14 @@ func (la *LdapAuthenticator) Login(
) (*schema.User, error) {
l, err := la.getLdapConnection(false)
if err != nil {
cclog.Warn("Error while getting ldap connection")
log.Warn("Error while getting ldap connection")
return nil, err
}
defer l.Close()
userDn := strings.ReplaceAll(Keys.LdapConfig.UserBind, "{username}", user.Username)
userDn := strings.Replace(config.Keys.LdapConfig.UserBind, "{username}", user.Username, -1)
if err := l.Bind(userDn, r.FormValue("password")); err != nil {
cclog.Errorf("AUTH/LDAP > Authentication for user %s failed: %v",
log.Errorf("AUTH/LDAP > Authentication for user %s failed: %v",
user.Username, err)
return nil, fmt.Errorf("Authentication failed")
}
@@ -142,11 +130,11 @@ func (la *LdapAuthenticator) Login(
}
func (la *LdapAuthenticator) Sync() error {
const InDB int = 1
const InLdap int = 2
const InBoth int = 3
const IN_DB int = 1
const IN_LDAP int = 2
const IN_BOTH int = 3
ur := repository.GetUserRepository()
lc := Keys.LdapConfig
lc := config.Keys.LdapConfig
users := map[string]int{}
usernames, err := ur.GetLdapUsernames()
@@ -155,12 +143,12 @@ func (la *LdapAuthenticator) Sync() error {
}
for _, username := range usernames {
users[username] = InDB
users[username] = IN_DB
}
l, err := la.getLdapConnection(true)
if err != nil {
cclog.Error("LDAP connection error")
log.Error("LDAP connection error")
return err
}
defer l.Close()
@@ -171,7 +159,7 @@ func (la *LdapAuthenticator) Sync() error {
lc.UserFilter,
[]string{"dn", "uid", la.UserAttr}, nil))
if err != nil {
cclog.Warn("LDAP search error")
log.Warn("LDAP search error")
return err
}
@@ -184,18 +172,18 @@ func (la *LdapAuthenticator) Sync() error {
_, ok := users[username]
if !ok {
users[username] = InLdap
users[username] = IN_LDAP
newnames[username] = entry.GetAttributeValue(la.UserAttr)
} else {
users[username] = InBoth
users[username] = IN_BOTH
}
}
for username, where := range users {
if where == InDB && lc.SyncDelOldUsers {
if where == IN_DB && lc.SyncDelOldUsers {
ur.DelUser(username)
cclog.Debugf("sync: remove %v (does not show up in LDAP anymore)", username)
} else if where == InLdap {
log.Debugf("sync: remove %v (does not show up in LDAP anymore)", username)
} else if where == IN_LDAP {
name := newnames[username]
var roles []string
@@ -210,9 +198,9 @@ func (la *LdapAuthenticator) Sync() error {
AuthSource: schema.AuthViaLDAP,
}
cclog.Debugf("sync: add %v (name: %v, roles: [user], ldap: true)", username, name)
log.Debugf("sync: add %v (name: %v, roles: [user], ldap: true)", username, name)
if err := ur.AddUser(user); err != nil {
cclog.Errorf("User '%s' LDAP: Insert into DB failed", username)
log.Errorf("User '%s' LDAP: Insert into DB failed", username)
return err
}
}
@@ -222,17 +210,17 @@ func (la *LdapAuthenticator) Sync() error {
}
func (la *LdapAuthenticator) getLdapConnection(admin bool) (*ldap.Conn, error) {
lc := Keys.LdapConfig
conn, err := ldap.DialURL(lc.URL)
lc := config.Keys.LdapConfig
conn, err := ldap.DialURL(lc.Url)
if err != nil {
cclog.Warn("LDAP URL dial failed")
log.Warn("LDAP URL dial failed")
return nil, err
}
if admin {
if err := conn.Bind(lc.SearchDN, la.syncPassword); err != nil {
conn.Close()
cclog.Warn("LDAP connection bind failed")
log.Warn("LDAP connection bind failed")
return nil, err
}
}

View File

@@ -1,16 +1,15 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
"fmt"
"net/http"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"golang.org/x/crypto/bcrypt"
)
@@ -28,19 +27,19 @@ func (la *LocalAuthenticator) CanLogin(
user *schema.User,
username string,
rw http.ResponseWriter,
r *http.Request,
) (*schema.User, bool) {
r *http.Request) (*schema.User, bool) {
return user, user != nil && user.AuthSource == schema.AuthViaLocalPassword
}
func (la *LocalAuthenticator) Login(
user *schema.User,
rw http.ResponseWriter,
r *http.Request,
) (*schema.User, error) {
r *http.Request) (*schema.User, error) {
if e := bcrypt.CompareHashAndPassword([]byte(user.Password),
[]byte(r.FormValue("password"))); e != nil {
cclog.Errorf("AUTH/LOCAL > Authentication for user %s failed!", user.Username)
log.Errorf("AUTH/LOCAL > Authentication for user %s failed!", user.Username)
return nil, fmt.Errorf("Authentication failed")
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
import (
@@ -14,20 +13,15 @@ import (
"os"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
"github.com/coreos/go-oidc/v3/oidc"
"github.com/gorilla/mux"
"golang.org/x/oauth2"
)
type OpenIDConfig struct {
Provider string `json:"provider"`
SyncUserOnLogin bool `json:"syncUserOnLogin"`
UpdateUserOnLogin bool `json:"updateUserOnLogin"`
}
type OIDC struct {
client *oauth2.Config
provider *oidc.Provider
@@ -55,17 +49,17 @@ func setCallbackCookie(w http.ResponseWriter, r *http.Request, name, value strin
}
func NewOIDC(a *Authentication) *OIDC {
provider, err := oidc.NewProvider(context.Background(), Keys.OpenIDConfig.Provider)
provider, err := oidc.NewProvider(context.Background(), config.Keys.OpenIDConfig.Provider)
if err != nil {
cclog.Fatal(err)
log.Fatal(err)
}
clientID := os.Getenv("OID_CLIENT_ID")
if clientID == "" {
cclog.Warn("environment variable 'OID_CLIENT_ID' not set (Open ID connect auth will not work)")
log.Warn("environment variable 'OID_CLIENT_ID' not set (Open ID connect auth will not work)")
}
clientSecret := os.Getenv("OID_CLIENT_SECRET")
if clientSecret == "" {
cclog.Warn("environment variable 'OID_CLIENT_SECRET' not set (Open ID connect auth will not work)")
log.Warn("environment variable 'OID_CLIENT_SECRET' not set (Open ID connect auth will not work)")
}
client := &oauth2.Config{
@@ -174,12 +168,12 @@ func (oa *OIDC) OAuth2Callback(rw http.ResponseWriter, r *http.Request) {
AuthSource: schema.AuthViaOIDC,
}
if Keys.OpenIDConfig.SyncUserOnLogin || Keys.OpenIDConfig.UpdateUserOnLogin {
if config.Keys.OpenIDConfig.SyncUserOnLogin || config.Keys.OpenIDConfig.UpdateUserOnLogin {
handleOIDCUser(user)
}
oa.authentication.SaveSession(rw, r, user)
cclog.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
log.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
http.RedirectHandler("/", http.StatusTemporaryRedirect).ServeHTTP(rw, r.WithContext(ctx))
}

View File

@@ -1,96 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package auth
var configSchema = `
{
"jwts": {
"description": "For JWT token authentication.",
"type": "object",
"properties": {
"max-age": {
"description": "Configure how long a token is valid. As string parsable by time.ParseDuration()",
"type": "string"
},
"cookieName": {
"description": "Cookie that should be checked for a JWT token.",
"type": "string"
},
"validateUser": {
"description": "Deny login for users not in database (but defined in JWT). Overwrite roles in JWT with database roles.",
"type": "boolean"
},
"trustedIssuer": {
"description": "Issuer that should be accepted when validating external JWTs ",
"type": "string"
},
"syncUserOnLogin": {
"description": "Add non-existent user to DB at login attempt with values provided in JWT.",
"type": "boolean"
}
},
"required": ["max-age"]
},
"oidc": {
"provider": {
"description": "",
"type": "string"
},
"syncUserOnLogin": {
"description": "",
"type": "boolean"
},
"updateUserOnLogin": {
"description": "",
"type": "boolean"
},
"required": ["provider"]
},
"ldap": {
"description": "For LDAP Authentication and user synchronisation.",
"type": "object",
"properties": {
"url": {
"description": "URL of LDAP directory server.",
"type": "string"
},
"user_base": {
"description": "Base DN of user tree root.",
"type": "string"
},
"search_dn": {
"description": "DN for authenticating LDAP admin account with general read rights.",
"type": "string"
},
"user_bind": {
"description": "Expression used to authenticate users via LDAP bind. Must contain uid={username}.",
"type": "string"
},
"user_filter": {
"description": "Filter to extract users for syncing.",
"type": "string"
},
"username_attr": {
"description": "Attribute with full username. Default: gecos",
"type": "string"
},
"sync_interval": {
"description": "Interval used for syncing local user table with LDAP directory. Parsed using time.ParseDuration.",
"type": "string"
},
"sync_del_old_users": {
"description": "Delete obsolete users in database.",
"type": "boolean"
},
"syncUserOnLogin": {
"description": "Add non-existent user to DB at login attempt if user exists in Ldap directory",
"type": "boolean"
}
},
"required": ["url", "user_base", "search_dn", "user_bind", "user_filter"]
},
"required": ["jwts"]
}`

View File

@@ -1,143 +1,72 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
// Package config implements the program configuration data structures, validation and parsing
package config
import (
"bytes"
"encoding/json"
"time"
"log"
"os"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
type ProgramConfig struct {
// Address where the http (or https) server will listen on (for example: 'localhost:80').
Addr string `json:"addr"`
// Addresses from which secured admin API endpoints can be reached, can be wildcard "*"
APIAllowedIPs []string `json:"apiAllowedIPs"`
// Drop root permissions once .env was read and the port was taken.
User string `json:"user"`
Group string `json:"group"`
// Disable authentication (for everything: API, Web-UI, ...)
DisableAuthentication bool `json:"disable-authentication"`
// If `embed-static-files` is true (default), the frontend files are directly
// embeded into the go binary and expected to be in web/frontend. Only if
// it is false the files in `static-files` are served instead.
EmbedStaticFiles bool `json:"embed-static-files"`
StaticFiles string `json:"static-files"`
// 'sqlite3' or 'mysql' (mysql will work for mariadb as well)
DBDriver string `json:"db-driver"`
// For sqlite3 a filename, for mysql a DSN in this format: https://github.com/go-sql-driver/mysql#dsn-data-source-name (Without query parameters!).
DB string `json:"db"`
// Keep all metric data in the metric data repositories,
// do not write to the job-archive.
DisableArchive bool `json:"disable-archive"`
EnableJobTaggers bool `json:"enable-job-taggers"`
// Validate json input against schema
Validate bool `json:"validate"`
// If 0 or empty, the session does not expire!
SessionMaxAge string `json:"session-max-age"`
// If both those options are not empty, use HTTPS using those certificates.
HTTPSCertFile string `json:"https-cert-file"`
HTTPSKeyFile string `json:"https-key-file"`
// If not the empty string and `addr` does not end in ":80",
// redirect every request incoming at port 80 to that url.
RedirectHTTPTo string `json:"redirect-http-to"`
// Where to store MachineState files
MachineStateDir string `json:"machine-state-dir"`
// If not zero, automatically mark jobs as stopped running X seconds longer than their walltime.
StopJobsExceedingWalltime int `json:"stop-jobs-exceeding-walltime"`
// Defines time X in seconds in which jobs are considered to be "short" and will be filtered in specific views.
ShortRunningJobsDuration int `json:"short-running-jobs-duration"`
// Energy Mix CO2 Emission Constant [g/kWh]
// If entered, displays estimated CO2 emission for job based on jobs totalEnergy
EmissionConstant int `json:"emission-constant"`
// If exists, will enable dynamic zoom in frontend metric plots using the configured values
EnableResampling *ResampleConfig `json:"resampling"`
}
type ResampleConfig struct {
// Array of resampling target resolutions, in seconds; Example: [600,300,60]
Resolutions []int `json:"resolutions"`
// Trigger next zoom level at less than this many visible datapoints
Trigger int `json:"trigger"`
}
type IntRange struct {
From int `json:"from"`
To int `json:"to"`
}
type TimeRange struct {
From *time.Time `json:"from"`
To *time.Time `json:"to"`
Range string `json:"range,omitempty"`
}
type FilterRanges struct {
Duration *IntRange `json:"duration"`
NumNodes *IntRange `json:"numNodes"`
StartTime *TimeRange `json:"startTime"`
}
type ClusterConfig struct {
Name string `json:"name"`
FilterRanges *FilterRanges `json:"filterRanges"`
MetricDataRepository json.RawMessage `json:"metricDataRepository"`
}
var Clusters []*ClusterConfig
var Keys ProgramConfig = ProgramConfig{
var Keys schema.ProgramConfig = schema.ProgramConfig{
Addr: "localhost:8080",
DisableAuthentication: false,
EmbedStaticFiles: true,
DBDriver: "sqlite3",
DB: "./var/job.db",
Archive: json.RawMessage(`{\"kind\":\"file\",\"path\":\"./var/job-archive\"}`),
DisableArchive: false,
Validate: false,
SessionMaxAge: "168h",
StopJobsExceedingWalltime: 0,
ShortRunningJobsDuration: 5 * 60,
UiDefaults: map[string]interface{}{
"analysis_view_histogramMetrics": []string{"flops_any", "mem_bw", "mem_used"},
"analysis_view_scatterPlotMetrics": [][]string{{"flops_any", "mem_bw"}, {"flops_any", "cpu_load"}, {"cpu_load", "mem_bw"}},
"job_view_nodestats_selectedMetrics": []string{"flops_any", "mem_bw", "mem_used"},
"job_view_selectedMetrics": []string{"flops_any", "mem_bw", "mem_used"},
"job_view_showFootprint": true,
"job_list_usePaging": false,
"plot_general_colorBackground": true,
"plot_general_colorscheme": []string{"#00bfff", "#0000ff", "#ff00ff", "#ff0000", "#ff8000", "#ffff00", "#80ff00"},
"plot_general_lineWidth": 3,
"plot_list_jobsPerPage": 50,
"plot_list_selectedMetrics": []string{"cpu_load", "mem_used", "flops_any", "mem_bw"},
"plot_view_plotsPerRow": 3,
"plot_view_showPolarplot": true,
"plot_view_showRoofline": true,
"plot_view_showStatTable": true,
"system_view_selectedMetric": "cpu_load",
"analysis_view_selectedTopEntity": "user",
"analysis_view_selectedTopCategory": "totalWalltime",
"status_view_selectedTopUserCategory": "totalJobs",
"status_view_selectedTopProjectCategory": "totalJobs",
},
}
func Init(mainConfig json.RawMessage, clusterConfig json.RawMessage) {
Validate(configSchema, mainConfig)
dec := json.NewDecoder(bytes.NewReader(mainConfig))
dec.DisallowUnknownFields()
if err := dec.Decode(&Keys); err != nil {
cclog.Abortf("Config Init: Could not decode config file '%s'.\nError: %s\n", mainConfig, err.Error())
}
func Init(flagConfigFile string) {
raw, err := os.ReadFile(flagConfigFile)
if err != nil {
if !os.IsNotExist(err) {
log.Fatalf("CONFIG ERROR: %v", err)
}
} else {
if err := schema.Validate(schema.Config, bytes.NewReader(raw)); err != nil {
log.Fatalf("Validate config: %v\n", err)
}
dec := json.NewDecoder(bytes.NewReader(raw))
dec.DisallowUnknownFields()
if err := dec.Decode(&Keys); err != nil {
log.Fatalf("could not decode: %v", err)
}
Validate(clustersSchema, clusterConfig)
dec = json.NewDecoder(bytes.NewReader(clusterConfig))
dec.DisallowUnknownFields()
if err := dec.Decode(&Clusters); err != nil {
cclog.Abortf("Config Init: Could not decode config file '%s'.\nError: %s\n", mainConfig, err.Error())
}
if len(Clusters) < 1 {
cclog.Abort("Config Init: At least one cluster required in config. Exited with error.")
if Keys.Clusters == nil || len(Keys.Clusters) < 1 {
log.Fatal("At least one cluster required in config!")
}
}
}

View File

@@ -1,30 +1,16 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package config
import (
"testing"
ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
)
func TestInit(t *testing.T) {
fp := "../../configs/config.json"
ccconf.Init(fp)
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
Init(cfg, clustercfg)
} else {
cclog.Abort("Cluster configuration must be present")
}
} else {
cclog.Abort("Main configuration must be present")
}
Init(fp)
if Keys.Addr != "0.0.0.0:443" {
t.Errorf("wrong addr\ngot: %s \nwant: 0.0.0.0:443", Keys.Addr)
}
@@ -32,17 +18,7 @@ func TestInit(t *testing.T) {
func TestInitMinimal(t *testing.T) {
fp := "../../configs/config-demo.json"
ccconf.Init(fp)
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
Init(cfg, clustercfg)
} else {
cclog.Abort("Cluster configuration must be present")
}
} else {
cclog.Abort("Main configuration must be present")
}
Init(fp)
if Keys.Addr != "127.0.0.1:8080" {
t.Errorf("wrong addr\ngot: %s \nwant: 127.0.0.1:8080", Keys.Addr)
}

View File

@@ -1,8 +1,3 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package config
import (
@@ -11,8 +6,6 @@ import (
"strings"
)
// DEPRECATED: SUPERSEDED BY NEW USER CONFIG - userConfig.go / web.go
type DefaultMetricsCluster struct {
Name string `json:"name"`
DefaultMetrics string `json:"default_metrics"`
@@ -23,7 +16,7 @@ type DefaultMetricsConfig struct {
}
func LoadDefaultMetricsConfig() (*DefaultMetricsConfig, error) {
filePath := "default_metrics.json"
filePath := "configs/default_metrics.json"
if _, err := os.Stat(filePath); os.IsNotExist(err) {
return nil, nil
}

View File

@@ -1,201 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package config
var configSchema = `
{
"type": "object",
"properties": {
"addr": {
"description": "Address where the http (or https) server will listen on (for example: 'localhost:80').",
"type": "string"
},
"apiAllowedIPs": {
"description": "Addresses from which secured API endpoints can be reached",
"type": "array",
"items": {
"type": "string"
}
},
"user": {
"description": "Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.",
"type": "string"
},
"group": {
"description": "Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.",
"type": "string"
},
"disable-authentication": {
"description": "Disable authentication (for everything: API, Web-UI, ...).",
"type": "boolean"
},
"embed-static-files": {
"description": "If all files in web/frontend/public should be served from within the binary itself (they are embedded) or not.",
"type": "boolean"
},
"static-files": {
"description": "Folder where static assets can be found, if embed-static-files is false.",
"type": "string"
},
"db": {
"description": "For sqlite3 a filename, for mysql a DSN in this format: https://github.com/go-sql-driver/mysql#dsn-data-source-name (Without query parameters!).",
"type": "string"
},
"disable-archive": {
"description": "Keep all metric data in the metric data repositories, do not write to the job-archive.",
"type": "boolean"
},
"enable-job-taggers": {
"description": "Turn on automatic application and jobclass taggers",
"type": "boolean"
},
"validate": {
"description": "Validate all input json documents against json schema.",
"type": "boolean"
},
"session-max-age": {
"description": "Specifies for how long a session shall be valid as a string parsable by time.ParseDuration(). If 0 or empty, the session/token does not expire!",
"type": "string"
},
"https-cert-file": {
"description": "Filepath to SSL certificate. If also https-key-file is set use HTTPS using those certificates.",
"type": "string"
},
"https-key-file": {
"description": "Filepath to SSL key file. If also https-cert-file is set use HTTPS using those certificates.",
"type": "string"
},
"redirect-http-to": {
"description": "If not the empty string and addr does not end in :80, redirect every request incoming at port 80 to that url.",
"type": "string"
},
"stop-jobs-exceeding-walltime": {
"description": "If not zero, automatically mark jobs as stopped running X seconds longer than their walltime. Only applies if walltime is set for job.",
"type": "integer"
},
"short-running-jobs-duration": {
"description": "Do not show running jobs shorter than X seconds.",
"type": "integer"
},
"emission-constant": {
"description": ".",
"type": "integer"
},
"cron-frequency": {
"description": "Frequency of cron job workers.",
"type": "object",
"properties": {
"duration-worker": {
"description": "Duration Update Worker [Defaults to '5m']",
"type": "string"
},
"footprint-worker": {
"description": "Metric-Footprint Update Worker [Defaults to '10m']",
"type": "string"
}
}
},
"enable-resampling": {
"description": "Enable dynamic zoom in frontend metric plots.",
"type": "object",
"properties": {
"trigger": {
"description": "Trigger next zoom level at less than this many visible datapoints.",
"type": "integer"
},
"resolutions": {
"description": "Array of resampling target resolutions, in seconds.",
"type": "array",
"items": {
"type": "integer"
}
}
},
"required": ["trigger", "resolutions"]
}
},
"required": ["apiAllowedIPs"]
}`
var clustersSchema = `
{
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"description": "The name of the cluster.",
"type": "string"
},
"metricDataRepository": {
"description": "Type of the metric data repository for this cluster",
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": ["influxdb", "prometheus", "cc-metric-store", "cc-metric-store-internal", "test"]
},
"url": {
"type": "string"
},
"token": {
"type": "string"
}
},
"required": ["kind"]
},
"filterRanges": {
"description": "This option controls the slider ranges for the UI controls of numNodes, duration, and startTime.",
"type": "object",
"properties": {
"numNodes": {
"description": "UI slider range for number of nodes",
"type": "object",
"properties": {
"from": {
"type": "integer"
},
"to": {
"type": "integer"
}
},
"required": ["from", "to"]
},
"duration": {
"description": "UI slider range for duration",
"type": "object",
"properties": {
"from": {
"type": "integer"
},
"to": {
"type": "integer"
}
},
"required": ["from", "to"]
},
"startTime": {
"description": "UI slider range for start time",
"type": "object",
"properties": {
"from": {
"type": "string",
"format": "date-time"
},
"to": {
"type": "null"
}
},
"required": ["from", "to"]
}
},
"required": ["numNodes", "duration", "startTime"]
}
},
"required": ["name", "metricDataRepository", "filterRanges"],
"minItems": 1
}
}`

View File

@@ -1,29 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package config
import (
"encoding/json"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/santhosh-tekuri/jsonschema/v5"
)
func Validate(schema string, instance json.RawMessage) {
sch, err := jsonschema.CompileString("schema.json", schema)
if err != nil {
cclog.Fatalf("%#v", err)
}
var v any
if err := json.Unmarshal([]byte(instance), &v); err != nil {
cclog.Fatal(err)
}
if err = sch.Validate(v); err != nil {
cclog.Fatalf("%#v", err)
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,6 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package model

View File

@@ -3,14 +3,12 @@
package model
import (
"bytes"
"fmt"
"io"
"strconv"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
type Count struct {
@@ -52,7 +50,6 @@ type IntRangeOutput struct {
type JobFilter struct {
Tags []string `json:"tags,omitempty"`
DbID []string `json:"dbId,omitempty"`
JobID *StringInput `json:"jobId,omitempty"`
ArrayJobID *int `json:"arrayJobId,omitempty"`
User *StringInput `json:"user,omitempty"`
@@ -60,16 +57,16 @@ type JobFilter struct {
JobName *StringInput `json:"jobName,omitempty"`
Cluster *StringInput `json:"cluster,omitempty"`
Partition *StringInput `json:"partition,omitempty"`
Duration *config.IntRange `json:"duration,omitempty"`
Duration *schema.IntRange `json:"duration,omitempty"`
Energy *FloatRange `json:"energy,omitempty"`
MinRunningFor *int `json:"minRunningFor,omitempty"`
NumNodes *config.IntRange `json:"numNodes,omitempty"`
NumAccelerators *config.IntRange `json:"numAccelerators,omitempty"`
NumHWThreads *config.IntRange `json:"numHWThreads,omitempty"`
StartTime *config.TimeRange `json:"startTime,omitempty"`
NumNodes *schema.IntRange `json:"numNodes,omitempty"`
NumAccelerators *schema.IntRange `json:"numAccelerators,omitempty"`
NumHWThreads *schema.IntRange `json:"numHWThreads,omitempty"`
StartTime *schema.TimeRange `json:"startTime,omitempty"`
State []schema.JobState `json:"state,omitempty"`
MetricStats []*MetricStatItem `json:"metricStats,omitempty"`
Shared *string `json:"shared,omitempty"`
Exclusive *int `json:"exclusive,omitempty"`
Node *StringInput `json:"node,omitempty"`
}
@@ -98,23 +95,9 @@ type JobResultList struct {
HasNextPage *bool `json:"hasNextPage,omitempty"`
}
type JobStats struct {
ID int `json:"id"`
JobID string `json:"jobId"`
StartTime int `json:"startTime"`
Duration int `json:"duration"`
Cluster string `json:"cluster"`
SubCluster string `json:"subCluster"`
NumNodes int `json:"numNodes"`
NumHWThreads *int `json:"numHWThreads,omitempty"`
NumAccelerators *int `json:"numAccelerators,omitempty"`
Stats []*NamedStats `json:"stats"`
}
type JobsStatistics struct {
ID string `json:"id"`
Name string `json:"name"`
TotalUsers int `json:"totalUsers"`
TotalJobs int `json:"totalJobs"`
RunningJobs int `json:"runningJobs"`
ShortJobs int `json:"shortJobs"`
@@ -159,49 +142,12 @@ type MetricStatItem struct {
type Mutation struct {
}
type NamedStats struct {
Name string `json:"name"`
Data *schema.MetricStatistics `json:"data"`
}
type NamedStatsWithScope struct {
Name string `json:"name"`
Scope schema.MetricScope `json:"scope"`
Stats []*ScopedStats `json:"stats"`
}
type NodeFilter struct {
Hostname *StringInput `json:"hostname,omitempty"`
Cluster *StringInput `json:"cluster,omitempty"`
Subcluster *StringInput `json:"subcluster,omitempty"`
SchedulerState *schema.SchedulerState `json:"schedulerState,omitempty"`
HealthState *string `json:"healthState,omitempty"`
TimeStart *int `json:"timeStart,omitempty"`
}
type NodeMetrics struct {
Host string `json:"host"`
SubCluster string `json:"subCluster"`
Metrics []*JobMetricWithName `json:"metrics"`
}
type NodeStateResultList struct {
Items []*schema.Node `json:"items"`
Count *int `json:"count,omitempty"`
}
type NodeStates struct {
State string `json:"state"`
Count int `json:"count"`
}
type NodeStatesTimed struct {
State string `json:"state"`
Type string `json:"type"`
Count int `json:"count"`
Time int `json:"time"`
}
type NodesResultList struct {
Items []*NodeMetrics `json:"items"`
Offset *int `json:"offset,omitempty"`
@@ -222,10 +168,7 @@ type PageRequest struct {
Page int `json:"page"`
}
type ScopedStats struct {
Hostname string `json:"hostname"`
ID *string `json:"id,omitempty"`
Data *schema.MetricStatistics `json:"data"`
type Query struct {
}
type StringInput struct {
@@ -258,22 +201,20 @@ type User struct {
type Aggregate string
const (
AggregateUser Aggregate = "USER"
AggregateProject Aggregate = "PROJECT"
AggregateCluster Aggregate = "CLUSTER"
AggregateSubcluster Aggregate = "SUBCLUSTER"
AggregateUser Aggregate = "USER"
AggregateProject Aggregate = "PROJECT"
AggregateCluster Aggregate = "CLUSTER"
)
var AllAggregate = []Aggregate{
AggregateUser,
AggregateProject,
AggregateCluster,
AggregateSubcluster,
}
func (e Aggregate) IsValid() bool {
switch e {
case AggregateUser, AggregateProject, AggregateCluster, AggregateSubcluster:
case AggregateUser, AggregateProject, AggregateCluster:
return true
}
return false
@@ -283,7 +224,7 @@ func (e Aggregate) String() string {
return string(e)
}
func (e *Aggregate) UnmarshalGQL(v any) error {
func (e *Aggregate) UnmarshalGQL(v interface{}) error {
str, ok := v.(string)
if !ok {
return fmt.Errorf("enums must be strings")
@@ -300,26 +241,11 @@ func (e Aggregate) MarshalGQL(w io.Writer) {
fmt.Fprint(w, strconv.Quote(e.String()))
}
func (e *Aggregate) UnmarshalJSON(b []byte) error {
s, err := strconv.Unquote(string(b))
if err != nil {
return err
}
return e.UnmarshalGQL(s)
}
func (e Aggregate) MarshalJSON() ([]byte, error) {
var buf bytes.Buffer
e.MarshalGQL(&buf)
return buf.Bytes(), nil
}
type SortByAggregate string
const (
SortByAggregateTotalwalltime SortByAggregate = "TOTALWALLTIME"
SortByAggregateTotaljobs SortByAggregate = "TOTALJOBS"
SortByAggregateTotalusers SortByAggregate = "TOTALUSERS"
SortByAggregateTotalnodes SortByAggregate = "TOTALNODES"
SortByAggregateTotalnodehours SortByAggregate = "TOTALNODEHOURS"
SortByAggregateTotalcores SortByAggregate = "TOTALCORES"
@@ -331,7 +257,6 @@ const (
var AllSortByAggregate = []SortByAggregate{
SortByAggregateTotalwalltime,
SortByAggregateTotaljobs,
SortByAggregateTotalusers,
SortByAggregateTotalnodes,
SortByAggregateTotalnodehours,
SortByAggregateTotalcores,
@@ -342,7 +267,7 @@ var AllSortByAggregate = []SortByAggregate{
func (e SortByAggregate) IsValid() bool {
switch e {
case SortByAggregateTotalwalltime, SortByAggregateTotaljobs, SortByAggregateTotalusers, SortByAggregateTotalnodes, SortByAggregateTotalnodehours, SortByAggregateTotalcores, SortByAggregateTotalcorehours, SortByAggregateTotalaccs, SortByAggregateTotalacchours:
case SortByAggregateTotalwalltime, SortByAggregateTotaljobs, SortByAggregateTotalnodes, SortByAggregateTotalnodehours, SortByAggregateTotalcores, SortByAggregateTotalcorehours, SortByAggregateTotalaccs, SortByAggregateTotalacchours:
return true
}
return false
@@ -352,7 +277,7 @@ func (e SortByAggregate) String() string {
return string(e)
}
func (e *SortByAggregate) UnmarshalGQL(v any) error {
func (e *SortByAggregate) UnmarshalGQL(v interface{}) error {
str, ok := v.(string)
if !ok {
return fmt.Errorf("enums must be strings")
@@ -369,20 +294,6 @@ func (e SortByAggregate) MarshalGQL(w io.Writer) {
fmt.Fprint(w, strconv.Quote(e.String()))
}
func (e *SortByAggregate) UnmarshalJSON(b []byte) error {
s, err := strconv.Unquote(string(b))
if err != nil {
return err
}
return e.UnmarshalGQL(s)
}
func (e SortByAggregate) MarshalJSON() ([]byte, error) {
var buf bytes.Buffer
e.MarshalGQL(&buf)
return buf.Bytes(), nil
}
type SortDirectionEnum string
const (
@@ -407,7 +318,7 @@ func (e SortDirectionEnum) String() string {
return string(e)
}
func (e *SortDirectionEnum) UnmarshalGQL(v any) error {
func (e *SortDirectionEnum) UnmarshalGQL(v interface{}) error {
str, ok := v.(string)
if !ok {
return fmt.Errorf("enums must be strings")
@@ -423,17 +334,3 @@ func (e *SortDirectionEnum) UnmarshalGQL(v any) error {
func (e SortDirectionEnum) MarshalGQL(w io.Writer) {
fmt.Fprint(w, strconv.Quote(e.String()))
}
func (e *SortDirectionEnum) UnmarshalJSON(b []byte) error {
s, err := strconv.Unquote(string(b))
if err != nil {
return err
}
return e.UnmarshalGQL(s)
}
func (e SortDirectionEnum) MarshalJSON() ([]byte, error) {
var buf bytes.Buffer
e.MarshalGQL(&buf)
return buf.Bytes(), nil
}

View File

@@ -4,7 +4,7 @@ import (
"sync"
"github.com/ClusterCockpit/cc-backend/internal/repository"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/jmoiron/sqlx"
)
@@ -32,7 +32,7 @@ func Init() {
func GetResolverInstance() *Resolver {
if resolverInstance == nil {
cclog.Fatal("Authentication module not initialized!")
log.Fatal("Authentication module not initialized!")
}
return resolverInstance

View File

@@ -2,7 +2,7 @@ package graph
// This file will be automatically regenerated based on the schema, any resolver implementations
// will be copied through when generating and any unknown code will be moved to the end.
// Code generated by github.com/99designs/gqlgen version v0.17.81
// Code generated by github.com/99designs/gqlgen version v0.17.57
import (
"context"
@@ -20,8 +20,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
// Partitions is the resolver for the partitions field.
@@ -29,21 +29,15 @@ func (r *clusterResolver) Partitions(ctx context.Context, obj *schema.Cluster) (
return r.Repo.Partitions(obj.Name)
}
// StartTime is the resolver for the startTime field.
func (r *jobResolver) StartTime(ctx context.Context, obj *schema.Job) (*time.Time, error) {
timestamp := time.Unix(obj.StartTime, 0)
return &timestamp, nil
}
// Tags is the resolver for the tags field.
func (r *jobResolver) Tags(ctx context.Context, obj *schema.Job) ([]*schema.Tag, error) {
return r.Repo.GetTags(repository.GetUserFromContext(ctx), obj.ID)
return r.Repo.GetTags(repository.GetUserFromContext(ctx), &obj.ID)
}
// ConcurrentJobs is the resolver for the concurrentJobs field.
func (r *jobResolver) ConcurrentJobs(ctx context.Context, obj *schema.Job) (*model.JobLinkResultList, error) {
// FIXME: Make the hardcoded duration configurable
if obj.Shared != "none" && obj.Duration > 600 {
if obj.Exclusive != 1 && obj.Duration > 600 {
return r.Repo.FindConcurrentJobs(ctx, obj)
}
@@ -54,7 +48,7 @@ func (r *jobResolver) ConcurrentJobs(ctx context.Context, obj *schema.Job) (*mod
func (r *jobResolver) Footprint(ctx context.Context, obj *schema.Job) ([]*model.FootprintValue, error) {
rawFootprint, err := r.Repo.FetchFootprint(obj)
if err != nil {
cclog.Warn("Error while fetching job footprint data")
log.Warn("Error while fetching job footprint data")
return nil, err
}
@@ -79,7 +73,7 @@ func (r *jobResolver) Footprint(ctx context.Context, obj *schema.Job) ([]*model.
func (r *jobResolver) EnergyFootprint(ctx context.Context, obj *schema.Job) ([]*model.EnergyFootprintValue, error) {
rawEnergyFootprint, err := r.Repo.FetchEnergyFootprint(obj)
if err != nil {
cclog.Warn("Error while fetching job energy footprint data")
log.Warn("Error while fetching job energy footprint data")
return nil, err
}
@@ -131,75 +125,40 @@ func (r *metricValueResolver) Name(ctx context.Context, obj *schema.MetricValue)
// CreateTag is the resolver for the createTag field.
func (r *mutationResolver) CreateTag(ctx context.Context, typeArg string, name string, scope string) (*schema.Tag, error) {
user := repository.GetUserFromContext(ctx)
if user == nil {
return nil, fmt.Errorf("no user in context")
id, err := r.Repo.CreateTag(typeArg, name, scope)
if err != nil {
log.Warn("Error while creating tag")
return nil, err
}
// Test Access: Admins && Admin Tag OR Support/Admin and Global Tag OR Everyone && Private Tag
if user.HasRole(schema.RoleAdmin) && scope == "admin" ||
user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport}) && scope == "global" ||
user.Username == scope {
// Create in DB
id, err := r.Repo.CreateTag(typeArg, name, scope)
if err != nil {
cclog.Warn("Error while creating tag")
return nil, err
}
return &schema.Tag{ID: id, Type: typeArg, Name: name, Scope: scope}, nil
} else {
cclog.Warnf("Not authorized to create tag with scope: %s", scope)
return nil, fmt.Errorf("not authorized to create tag with scope: %s", scope)
}
return &schema.Tag{ID: id, Type: typeArg, Name: name, Scope: scope}, nil
}
// DeleteTag is the resolver for the deleteTag field.
func (r *mutationResolver) DeleteTag(ctx context.Context, id string) (string, error) {
// This Uses ID string <-> ID string, removeTagFromList uses []string <-> []int
panic(fmt.Errorf("not implemented: DeleteTag - deleteTag"))
}
// AddTagsToJob is the resolver for the addTagsToJob field.
func (r *mutationResolver) AddTagsToJob(ctx context.Context, job string, tagIds []string) ([]*schema.Tag, error) {
user := repository.GetUserFromContext(ctx)
if user == nil {
return nil, fmt.Errorf("no user in context")
}
// Selectable Tags Pre-Filtered by Scope in Frontend: No backend check required
jid, err := strconv.ParseInt(job, 10, 64)
if err != nil {
cclog.Warn("Error while adding tag to job")
log.Warn("Error while adding tag to job")
return nil, err
}
tags := []*schema.Tag{}
for _, tagId := range tagIds {
// Get ID
tid, err := strconv.ParseInt(tagId, 10, 64)
if err != nil {
cclog.Warn("Error while parsing tag id")
log.Warn("Error while parsing tag id")
return nil, err
}
// Test Exists
_, _, tscope, exists := r.Repo.TagInfo(tid)
if !exists {
cclog.Warnf("Tag does not exist (ID): %d", tid)
return nil, fmt.Errorf("tag does not exist (ID): %d", tid)
}
// Test Access: Admins && Admin Tag OR Support/Admin and Global Tag OR Everyone && Private Tag
if user.HasRole(schema.RoleAdmin) && tscope == "admin" ||
user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport}) && tscope == "global" ||
user.Username == tscope {
// Add to Job
if tags, err = r.Repo.AddTag(user, jid, tid); err != nil {
cclog.Warn("Error while adding tag")
return nil, err
}
} else {
cclog.Warnf("Not authorized to add tag: %d", tid)
return nil, fmt.Errorf("not authorized to add tag: %d", tid)
if tags, err = r.Repo.AddTag(repository.GetUserFromContext(ctx), jid, tid); err != nil {
log.Warn("Error while adding tag")
return nil, err
}
}
@@ -208,123 +167,40 @@ func (r *mutationResolver) AddTagsToJob(ctx context.Context, job string, tagIds
// RemoveTagsFromJob is the resolver for the removeTagsFromJob field.
func (r *mutationResolver) RemoveTagsFromJob(ctx context.Context, job string, tagIds []string) ([]*schema.Tag, error) {
user := repository.GetUserFromContext(ctx)
if user == nil {
return nil, fmt.Errorf("no user in context")
}
// Removable Tags Pre-Filtered by Scope in Frontend: No backend check required
jid, err := strconv.ParseInt(job, 10, 64)
if err != nil {
cclog.Warn("Error while parsing job id")
log.Warn("Error while parsing job id")
return nil, err
}
tags := []*schema.Tag{}
for _, tagId := range tagIds {
// Get ID
tid, err := strconv.ParseInt(tagId, 10, 64)
if err != nil {
cclog.Warn("Error while parsing tag id")
log.Warn("Error while parsing tag id")
return nil, err
}
// Test Exists
_, _, tscope, exists := r.Repo.TagInfo(tid)
if !exists {
cclog.Warnf("Tag does not exist (ID): %d", tid)
return nil, fmt.Errorf("tag does not exist (ID): %d", tid)
}
// Test Access: Admins && Admin Tag OR Support/Admin and Global Tag OR Everyone && Private Tag
if user.HasRole(schema.RoleAdmin) && tscope == "admin" ||
user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport}) && tscope == "global" ||
user.Username == tscope {
// Remove from Job
if tags, err = r.Repo.RemoveTag(user, jid, tid); err != nil {
cclog.Warn("Error while removing tag")
return nil, err
}
} else {
cclog.Warnf("Not authorized to remove tag: %d", tid)
return nil, fmt.Errorf("not authorized to remove tag: %d", tid)
}
}
return tags, nil
}
// RemoveTagFromList is the resolver for the removeTagFromList field.
func (r *mutationResolver) RemoveTagFromList(ctx context.Context, tagIds []string) ([]int, error) {
// Needs Contextuser
user := repository.GetUserFromContext(ctx)
if user == nil {
return nil, fmt.Errorf("no user in context")
}
tags := []int{}
for _, tagId := range tagIds {
// Get ID
tid, err := strconv.ParseInt(tagId, 10, 64)
if err != nil {
cclog.Warn("Error while parsing tag id for removal")
if tags, err = r.Repo.RemoveTag(repository.GetUserFromContext(ctx), jid, tid); err != nil {
log.Warn("Error while removing tag")
return nil, err
}
// Test Exists
_, _, tscope, exists := r.Repo.TagInfo(tid)
if !exists {
cclog.Warnf("Tag does not exist (ID): %d", tid)
return nil, fmt.Errorf("tag does not exist (ID): %d", tid)
}
// Test Access: Admins && Admin Tag OR Everyone && Private Tag
if user.HasRole(schema.RoleAdmin) && (tscope == "global" || tscope == "admin") || user.Username == tscope {
// Remove from DB
if err = r.Repo.RemoveTagById(tid); err != nil {
cclog.Warn("Error while removing tag")
return nil, err
} else {
tags = append(tags, int(tid))
}
} else {
cclog.Warnf("Not authorized to remove tag: %d", tid)
return nil, fmt.Errorf("not authorized to remove tag: %d", tid)
}
}
return tags, nil
}
// UpdateConfiguration is the resolver for the updateConfiguration field.
func (r *mutationResolver) UpdateConfiguration(ctx context.Context, name string, value string) (*string, error) {
if err := repository.GetUserCfgRepo().UpdateConfig(name, value, repository.GetUserFromContext(ctx)); err != nil {
cclog.Warn("Error while updating user config")
log.Warn("Error while updating user config")
return nil, err
}
return nil, nil
}
// ID is the resolver for the id field.
func (r *nodeResolver) ID(ctx context.Context, obj *schema.Node) (string, error) {
panic(fmt.Errorf("not implemented: ID - id"))
}
// SchedulerState is the resolver for the schedulerState field.
func (r *nodeResolver) SchedulerState(ctx context.Context, obj *schema.Node) (schema.SchedulerState, error) {
panic(fmt.Errorf("not implemented: SchedulerState - schedulerState"))
}
// HealthState is the resolver for the healthState field.
func (r *nodeResolver) HealthState(ctx context.Context, obj *schema.Node) (string, error) {
panic(fmt.Errorf("not implemented: HealthState - healthState"))
}
// MetaData is the resolver for the metaData field.
func (r *nodeResolver) MetaData(ctx context.Context, obj *schema.Node) (any, error) {
panic(fmt.Errorf("not implemented: MetaData - metaData"))
}
// Clusters is the resolver for the clusters field.
func (r *queryResolver) Clusters(ctx context.Context) ([]*schema.Cluster, error) {
return archive.Clusters, nil
@@ -349,7 +225,7 @@ func (r *queryResolver) User(ctx context.Context, username string) (*model.User,
func (r *queryResolver) AllocatedNodes(ctx context.Context, cluster string) ([]*model.Count, error) {
data, err := r.Repo.AllocatedNodes(cluster)
if err != nil {
cclog.Warn("Error while fetching allocated nodes")
log.Warn("Error while fetching allocated nodes")
return nil, err
}
@@ -364,81 +240,17 @@ func (r *queryResolver) AllocatedNodes(ctx context.Context, cluster string) ([]*
return counts, nil
}
// Node is the resolver for the node field.
func (r *queryResolver) Node(ctx context.Context, id string) (*schema.Node, error) {
repo := repository.GetNodeRepository()
numericId, err := strconv.ParseInt(id, 10, 64)
if err != nil {
cclog.Warn("Error while parsing job id")
return nil, err
}
return repo.GetNodeById(numericId, false)
}
// Nodes is the resolver for the nodes field.
func (r *queryResolver) Nodes(ctx context.Context, filter []*model.NodeFilter, order *model.OrderByInput) (*model.NodeStateResultList, error) {
repo := repository.GetNodeRepository()
nodes, err := repo.QueryNodes(ctx, filter, order)
count := len(nodes)
return &model.NodeStateResultList{Items: nodes, Count: &count}, err
}
// NodeStates is the resolver for the nodeStates field.
func (r *queryResolver) NodeStates(ctx context.Context, filter []*model.NodeFilter) ([]*model.NodeStates, error) {
repo := repository.GetNodeRepository()
stateCounts, serr := repo.CountNodeStates(ctx, filter)
if serr != nil {
cclog.Warnf("Error while counting nodeStates: %s", serr.Error())
return nil, serr
}
healthCounts, herr := repo.CountHealthStates(ctx, filter)
if herr != nil {
cclog.Warnf("Error while counting healthStates: %s", herr.Error())
return nil, herr
}
allCounts := make([]*model.NodeStates, 0)
allCounts = append(stateCounts, healthCounts...)
return allCounts, nil
}
// NodeStatesTimed is the resolver for the nodeStatesTimed field.
func (r *queryResolver) NodeStatesTimed(ctx context.Context, filter []*model.NodeFilter) ([]*model.NodeStatesTimed, error) {
panic(fmt.Errorf("not implemented: NodeStatesTimed - NodeStatesTimed"))
// repo := repository.GetNodeRepository()
// stateCounts, serr := repo.CountNodeStates(ctx, filter)
// if serr != nil {
// cclog.Warnf("Error while counting nodeStates: %s", serr.Error())
// return nil, serr
// }
// healthCounts, herr := repo.CountHealthStates(ctx, filter)
// if herr != nil {
// cclog.Warnf("Error while counting healthStates: %s", herr.Error())
// return nil, herr
// }
// allCounts := make([]*model.NodeStates, 0)
// allCounts = append(stateCounts, healthCounts...)
// return allCounts, nil
}
// Job is the resolver for the job field.
func (r *queryResolver) Job(ctx context.Context, id string) (*schema.Job, error) {
numericId, err := strconv.ParseInt(id, 10, 64)
if err != nil {
cclog.Warn("Error while parsing job id")
log.Warn("Error while parsing job id")
return nil, err
}
job, err := r.Repo.FindById(ctx, numericId)
if err != nil {
cclog.Warn("Error while finding job by id")
log.Warn("Error while finding job by id")
return nil, err
}
@@ -465,13 +277,13 @@ func (r *queryResolver) JobMetrics(ctx context.Context, id string, metrics []str
job, err := r.Query().Job(ctx, id)
if err != nil {
cclog.Warn("Error while querying job for metrics")
log.Warn("Error while querying job for metrics")
return nil, err
}
data, err := metricDataDispatcher.LoadData(job, metrics, scopes, ctx, *resolution)
if err != nil {
cclog.Warn("Error while loading job data")
log.Warn("Error while loading job data")
return nil, err
}
@@ -489,67 +301,9 @@ func (r *queryResolver) JobMetrics(ctx context.Context, id string, metrics []str
return res, err
}
// JobStats is the resolver for the jobStats field.
func (r *queryResolver) JobStats(ctx context.Context, id string, metrics []string) ([]*model.NamedStats, error) {
job, err := r.Query().Job(ctx, id)
if err != nil {
cclog.Warnf("Error while querying job %s for metadata", id)
return nil, err
}
data, err := metricDataDispatcher.LoadJobStats(job, metrics, ctx)
if err != nil {
cclog.Warnf("Error while loading jobStats data for job id %s", id)
return nil, err
}
res := []*model.NamedStats{}
for name, md := range data {
res = append(res, &model.NamedStats{
Name: name,
Data: &md,
})
}
return res, err
}
// ScopedJobStats is the resolver for the scopedJobStats field.
func (r *queryResolver) ScopedJobStats(ctx context.Context, id string, metrics []string, scopes []schema.MetricScope) ([]*model.NamedStatsWithScope, error) {
job, err := r.Query().Job(ctx, id)
if err != nil {
cclog.Warnf("Error while querying job %s for metadata", id)
return nil, err
}
data, err := metricDataDispatcher.LoadScopedJobStats(job, metrics, scopes, ctx)
if err != nil {
cclog.Warnf("Error while loading scopedJobStats data for job id %s", id)
return nil, err
}
res := make([]*model.NamedStatsWithScope, 0)
for name, scoped := range data {
for scope, stats := range scoped {
mdlStats := make([]*model.ScopedStats, 0)
for _, stat := range stats {
mdlStats = append(mdlStats, &model.ScopedStats{
Hostname: stat.Hostname,
ID: stat.Id,
Data: stat.Data,
})
}
res = append(res, &model.NamedStatsWithScope{
Name: name,
Scope: scope,
Stats: mdlStats,
})
}
}
return res, nil
// JobsFootprints is the resolver for the jobsFootprints field.
func (r *queryResolver) JobsFootprints(ctx context.Context, filter []*model.JobFilter, metrics []string) (*model.Footprints, error) {
return r.jobsFootprints(ctx, filter, metrics)
}
// Jobs is the resolver for the jobs field.
@@ -563,35 +317,40 @@ func (r *queryResolver) Jobs(ctx context.Context, filter []*model.JobFilter, pag
jobs, err := r.Repo.QueryJobs(ctx, filter, page, order)
if err != nil {
cclog.Warn("Error while querying jobs")
log.Warn("Error while querying jobs")
return nil, err
}
count, err := r.Repo.CountJobs(ctx, filter)
if err != nil {
cclog.Warn("Error while counting jobs")
log.Warn("Error while counting jobs")
return nil, err
}
// Note: Even if App-Default 'config.Keys.UiDefaults["job_list_usePaging"]' is set, always return hasNextPage boolean.
// Users can decide in frontend to use continuous scroll, even if app-default is paging!
/*
Example Page 4 @ 10 IpP : Does item 41 exist?
Minimal Page 41 @ 1 IpP : If len(result) is 1, Page 5 @ 10 IpP exists.
*/
nextPage := &model.PageRequest{
ItemsPerPage: 1,
Page: ((page.Page * page.ItemsPerPage) + 1),
}
nextJobs, err := r.Repo.QueryJobs(ctx, filter, nextPage, order)
if err != nil {
cclog.Warn("Error while querying next jobs")
return nil, err
}
if !config.Keys.UiDefaults["job_list_usePaging"].(bool) {
hasNextPage := false
// page.Page += 1 : Simple, but expensive
// Example Page 4 @ 10 IpP : Does item 41 exist?
// Minimal Page 41 @ 1 IpP : If len(result) is 1, Page 5 @ 10 IpP exists.
nextPage := &model.PageRequest{
ItemsPerPage: 1,
Page: ((page.Page * page.ItemsPerPage) + 1),
}
hasNextPage := len(nextJobs) == 1
nextJobs, err := r.Repo.QueryJobs(ctx, filter, nextPage, order)
if err != nil {
log.Warn("Error while querying next jobs")
return nil, err
}
return &model.JobResultList{Items: jobs, Count: &count, HasNextPage: &hasNextPage}, nil
if len(nextJobs) == 1 {
hasNextPage = true
}
return &model.JobResultList{Items: jobs, Count: &count, HasNextPage: &hasNextPage}, nil
} else {
return &model.JobResultList{Items: jobs, Count: &count}, nil
}
}
// JobsStatistics is the resolver for the jobsStatistics field.
@@ -600,10 +359,10 @@ func (r *queryResolver) JobsStatistics(ctx context.Context, filter []*model.JobF
var stats []*model.JobsStatistics
// Top Level Defaults
defaultDurationBins := "1h"
defaultMetricBins := 10
var defaultDurationBins string = "1h"
var defaultMetricBins int = 10
if requireField(ctx, "totalJobs") || requireField(ctx, "totalUsers") || requireField(ctx, "totalWalltime") || requireField(ctx, "totalNodes") || requireField(ctx, "totalCores") ||
if requireField(ctx, "totalJobs") || requireField(ctx, "totalWalltime") || requireField(ctx, "totalNodes") || requireField(ctx, "totalCores") ||
requireField(ctx, "totalAccs") || requireField(ctx, "totalNodeHours") || requireField(ctx, "totalCoreHours") || requireField(ctx, "totalAccHours") {
if groupBy == nil {
stats, err = r.Repo.JobsStats(ctx, filter)
@@ -670,62 +429,6 @@ func (r *queryResolver) JobsStatistics(ctx context.Context, filter []*model.JobF
return stats, nil
}
// JobsMetricStats is the resolver for the jobsMetricStats field.
func (r *queryResolver) JobsMetricStats(ctx context.Context, filter []*model.JobFilter, metrics []string) ([]*model.JobStats, error) {
// No Paging, Fixed Order by StartTime ASC
order := &model.OrderByInput{
Field: "startTime",
Type: "col",
Order: "ASC",
}
jobs, err := r.Repo.QueryJobs(ctx, filter, nil, order)
if err != nil {
cclog.Warn("Error while querying jobs for comparison")
return nil, err
}
res := []*model.JobStats{}
for _, job := range jobs {
data, err := metricDataDispatcher.LoadJobStats(job, metrics, ctx)
if err != nil {
cclog.Warnf("Error while loading comparison jobStats data for job id %d", job.JobID)
continue
// return nil, err
}
sres := []*model.NamedStats{}
for name, md := range data {
sres = append(sres, &model.NamedStats{
Name: name,
Data: &md,
})
}
numThreadsInt := int(job.NumHWThreads)
numAccsInt := int(job.NumAcc)
res = append(res, &model.JobStats{
ID: int(*job.ID),
JobID: strconv.Itoa(int(job.JobID)),
StartTime: int(job.StartTime),
Duration: int(job.Duration),
Cluster: job.Cluster,
SubCluster: job.SubCluster,
NumNodes: int(job.NumNodes),
NumHWThreads: &numThreadsInt,
NumAccelerators: &numAccsInt,
Stats: sres,
})
}
return res, err
}
// JobsFootprints is the resolver for the jobsFootprints field.
func (r *queryResolver) JobsFootprints(ctx context.Context, filter []*model.JobFilter, metrics []string) (*model.Footprints, error) {
// NOTE: Legacy Naming! This resolver is for normalized histograms in analysis view only - *Not* related to DB "footprint" column!
return r.jobsFootprints(ctx, filter, metrics)
}
// RooflineHeatmap is the resolver for the rooflineHeatmap field.
func (r *queryResolver) RooflineHeatmap(ctx context.Context, filter []*model.JobFilter, rows int, cols int, minX float64, minY float64, maxX float64, maxY float64) ([][]float64, error) {
return r.rooflineHeatmap(ctx, filter, rows, cols, minX, minY, maxX, maxY)
@@ -746,7 +449,7 @@ func (r *queryResolver) NodeMetrics(ctx context.Context, cluster string, nodes [
data, err := metricDataDispatcher.LoadNodeData(cluster, metrics, nodes, scopes, from, to, ctx)
if err != nil {
cclog.Warn("error while loading node data")
log.Warn("error while loading node data")
return nil, err
}
@@ -758,7 +461,7 @@ func (r *queryResolver) NodeMetrics(ctx context.Context, cluster string, nodes [
}
host.SubCluster, err = archive.GetSubClusterByNode(cluster, hostname)
if err != nil {
cclog.Warnf("error in nodeMetrics resolver: %s", err)
log.Warnf("error in nodeMetrics resolver: %s", err)
}
for metric, scopedMetrics := range metrics {
@@ -802,7 +505,7 @@ func (r *queryResolver) NodeMetricsList(ctx context.Context, cluster string, sub
data, totalNodes, hasNextPage, err := metricDataDispatcher.LoadNodeListData(cluster, subCluster, nodeFilter, metrics, scopes, *resolution, from, to, page, ctx)
if err != nil {
cclog.Warn("error while loading node data")
log.Warn("error while loading node data")
return nil, err
}
@@ -814,7 +517,7 @@ func (r *queryResolver) NodeMetricsList(ctx context.Context, cluster string, sub
}
host.SubCluster, err = archive.GetSubClusterByNode(cluster, hostname)
if err != nil {
cclog.Warnf("error in nodeMetrics resolver: %s", err)
log.Warnf("error in nodeMetrics resolver: %s", err)
}
for metric, scopedMetrics := range metrics {
@@ -860,9 +563,6 @@ func (r *Resolver) MetricValue() generated.MetricValueResolver { return &metricV
// Mutation returns generated.MutationResolver implementation.
func (r *Resolver) Mutation() generated.MutationResolver { return &mutationResolver{r} }
// Node returns generated.NodeResolver implementation.
func (r *Resolver) Node() generated.NodeResolver { return &nodeResolver{r} }
// Query returns generated.QueryResolver implementation.
func (r *Resolver) Query() generated.QueryResolver { return &queryResolver{r} }
@@ -873,6 +573,5 @@ type clusterResolver struct{ *Resolver }
type jobResolver struct{ *Resolver }
type metricValueResolver struct{ *Resolver }
type mutationResolver struct{ *Resolver }
type nodeResolver struct{ *Resolver }
type queryResolver struct{ *Resolver }
type subClusterResolver struct{ *Resolver }

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package graph
@@ -12,8 +12,9 @@ import (
"github.com/99designs/gqlgen/graphql"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
// "github.com/ClusterCockpit/cc-backend/pkg/archive"
)
const MAX_JOBS_FOR_ANALYSIS = 500
@@ -27,7 +28,7 @@ func (r *queryResolver) rooflineHeatmap(
) ([][]float64, error) {
jobs, err := r.Repo.QueryJobs(ctx, filter, &model.PageRequest{Page: 1, ItemsPerPage: MAX_JOBS_FOR_ANALYSIS + 1}, nil)
if err != nil {
cclog.Error("Error while querying jobs for roofline")
log.Error("Error while querying jobs for roofline")
return nil, err
}
if len(jobs) > MAX_JOBS_FOR_ANALYSIS {
@@ -55,13 +56,13 @@ func (r *queryResolver) rooflineHeatmap(
jobdata, err := metricDataDispatcher.LoadData(job, []string{"flops_any", "mem_bw"}, []schema.MetricScope{schema.MetricScopeNode}, ctx, 0)
if err != nil {
cclog.Errorf("Error while loading roofline metrics for job %d", job.ID)
log.Errorf("Error while loading roofline metrics for job %d", job.ID)
return nil, err
}
flops_, membw_ := jobdata["flops_any"], jobdata["mem_bw"]
if flops_ == nil && membw_ == nil {
cclog.Infof("rooflineHeatmap(): 'flops_any' or 'mem_bw' missing for job %d", job.ID)
log.Infof("rooflineHeatmap(): 'flops_any' or 'mem_bw' missing for job %d", job.ID)
continue
// return nil, fmt.Errorf("GRAPH/UTIL > 'flops_any' or 'mem_bw' missing for job %d", job.ID)
}
@@ -69,7 +70,7 @@ func (r *queryResolver) rooflineHeatmap(
flops, ok1 := flops_["node"]
membw, ok2 := membw_["node"]
if !ok1 || !ok2 {
cclog.Info("rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
log.Info("rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
continue
// TODO/FIXME:
// return nil, errors.New("GRAPH/UTIL > todo: rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
@@ -104,7 +105,7 @@ func (r *queryResolver) rooflineHeatmap(
func (r *queryResolver) jobsFootprints(ctx context.Context, filter []*model.JobFilter, metrics []string) (*model.Footprints, error) {
jobs, err := r.Repo.QueryJobs(ctx, filter, &model.PageRequest{Page: 1, ItemsPerPage: MAX_JOBS_FOR_ANALYSIS + 1}, nil)
if err != nil {
cclog.Error("Error while querying jobs for footprint")
log.Error("Error while querying jobs for footprint")
return nil, err
}
if len(jobs) > MAX_JOBS_FOR_ANALYSIS {
@@ -127,7 +128,7 @@ func (r *queryResolver) jobsFootprints(ctx context.Context, filter []*model.JobF
}
if err := metricDataDispatcher.LoadAverages(job, metrics, avgs, ctx); err != nil {
cclog.Error("Error while loading averages for footprint")
log.Error("Error while loading averages for footprint")
return nil, err
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package importer
@@ -15,8 +15,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
// Import all jobs specified as `<path-to-meta.json>:<path-to-data.json>,...`
@@ -31,7 +31,7 @@ func HandleImportFlag(flag string) error {
raw, err := os.ReadFile(files[0])
if err != nil {
cclog.Warn("Error while reading metadata file for import")
log.Warn("Error while reading metadata file for import")
return err
}
@@ -42,18 +42,15 @@ func HandleImportFlag(flag string) error {
}
dec := json.NewDecoder(bytes.NewReader(raw))
dec.DisallowUnknownFields()
job := schema.Job{
Shared: "none",
MonitoringStatus: schema.MonitoringStatusRunningOrArchiving,
}
job := schema.JobMeta{BaseJob: schema.JobDefaults}
if err = dec.Decode(&job); err != nil {
cclog.Warn("Error while decoding raw json metadata for import")
log.Warn("Error while decoding raw json metadata for import")
return err
}
raw, err = os.ReadFile(files[1])
if err != nil {
cclog.Warn("Error while reading jobdata file for import")
log.Warn("Error while reading jobdata file for import")
return err
}
@@ -66,7 +63,7 @@ func HandleImportFlag(flag string) error {
dec.DisallowUnknownFields()
jobData := schema.JobData{}
if err = dec.Decode(&jobData); err != nil {
cclog.Warn("Error while decoding raw json jobdata for import")
log.Warn("Error while decoding raw json jobdata for import")
return err
}
@@ -74,7 +71,7 @@ func HandleImportFlag(flag string) error {
sc, err := archive.GetSubCluster(job.Cluster, job.SubCluster)
if err != nil {
cclog.Errorf("cannot get subcluster: %s", err.Error())
log.Errorf("cannot get subcluster: %s", err.Error())
return err
}
@@ -94,80 +91,72 @@ func HandleImportFlag(flag string) error {
job.RawFootprint, err = json.Marshal(job.Footprint)
if err != nil {
cclog.Warn("Error while marshaling job footprint")
log.Warn("Error while marshaling job footprint")
return err
}
job.EnergyFootprint = make(map[string]float64)
var totalEnergy float64
var energy float64
// Total Job Energy Outside Loop
totalEnergy := 0.0
for _, fp := range sc.EnergyFootprint {
// Always Init Metric Energy Inside Loop
metricEnergy := 0.0
if i, err := archive.MetricIndex(sc.MetricConfig, fp); err == nil {
// Note: For DB data, calculate and save as kWh
// Energy: Power (in Watts) * Time (in Seconds)
if sc.MetricConfig[i].Energy == "energy" { // this metric has energy as unit (Joules)
cclog.Warnf("Update EnergyFootprint for Job %d and Metric %s on cluster %s: Set to 'energy' in cluster.json: Not implemented, will return 0.0", job.JobID, job.Cluster, fp)
// FIXME: Needs sum as stats type
} else if sc.MetricConfig[i].Energy == "power" { // this metric has power as unit (Watt)
// Energy: Power (in Watts) * Time (in Seconds)
// Unit: (W * (s / 3600)) / 1000 = kWh
// Round 2 Digits: round(Energy * 100) / 100
// Here: (All-Node Metric Average * Number of Nodes) * (Job Duration in Seconds / 3600) / 1000
// Note: Shared Jobs handled correctly since "Node Average" is based on partial resources, while "numNodes" factor is 1
rawEnergy := ((repository.LoadJobStat(&job, fp, "avg") * float64(job.NumNodes)) * (float64(job.Duration) / 3600.0)) / 1000.0
metricEnergy = math.Round(rawEnergy*100.0) / 100.0
// Unit: ( W * s ) / 3600 / 1000 = kWh ; Rounded to 2 nearest digits
energy = math.Round(((repository.LoadJobStat(&job, fp, "avg")*float64(job.Duration))/3600/1000)*100) / 100
}
} else {
cclog.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, job.ID)
log.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, job.ID)
}
job.EnergyFootprint[fp] = metricEnergy
totalEnergy += metricEnergy
job.EnergyFootprint[fp] = energy
totalEnergy += energy
}
job.Energy = (math.Round(totalEnergy*100.0) / 100.0)
job.Energy = (math.Round(totalEnergy*100) / 100)
if job.RawEnergyFootprint, err = json.Marshal(job.EnergyFootprint); err != nil {
cclog.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", job.ID)
log.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", job.ID)
return err
}
job.RawResources, err = json.Marshal(job.Resources)
if err != nil {
cclog.Warn("Error while marshaling job resources")
log.Warn("Error while marshaling job resources")
return err
}
job.RawMetaData, err = json.Marshal(job.MetaData)
if err != nil {
cclog.Warn("Error while marshaling job metadata")
log.Warn("Error while marshaling job metadata")
return err
}
if err = SanityChecks(&job); err != nil {
cclog.Warn("BaseJob SanityChecks failed")
if err = SanityChecks(&job.BaseJob); err != nil {
log.Warn("BaseJob SanityChecks failed")
return err
}
if err = archive.GetHandle().ImportJob(&job, &jobData); err != nil {
cclog.Error("Error while importing job")
log.Error("Error while importing job")
return err
}
id, err := r.InsertJob(&job)
if err != nil {
cclog.Warn("Error while job db insert")
log.Warn("Error while job db insert")
return err
}
for _, tag := range job.Tags {
if err := r.ImportTag(id, tag.Type, tag.Name, tag.Scope); err != nil {
cclog.Error("Error while adding or creating tag on import")
log.Error("Error while adding or creating tag on import")
return err
}
}
cclog.Infof("successfully imported a new job (jobId: %d, cluster: %s, dbid: %d)", job.JobID, job.Cluster, id)
log.Infof("successfully imported a new job (jobId: %d, cluster: %s, dbid: %d)", job.JobID, job.Cluster, id)
}
return nil
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package importer_test
@@ -16,8 +16,7 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/importer"
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/log"
)
func copyFile(s string, d string) error {
@@ -37,16 +36,15 @@ func copyFile(s string, d string) error {
func setup(t *testing.T) *repository.JobRepository {
const testconfig = `{
"main": {
"addr": "0.0.0.0:8080",
"validate": false,
"apiAllowedIPs": [
"*"
]},
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"jwts": {
"max-age": "2m"
},
"clusters": [
{
"name": "testcluster",
@@ -77,7 +75,7 @@ func setup(t *testing.T) *repository.JobRepository {
}
]}`
cclog.Init("info", true)
log.Init("info", true)
tmpdir := t.TempDir()
jobarchive := filepath.Join(tmpdir, "job-archive")
@@ -107,19 +105,7 @@ func setup(t *testing.T) *repository.JobRepository {
t.Fatal(err)
}
ccconf.Init(cfgFilePath)
// Load and check main configuration
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
config.Init(cfg, clustercfg)
} else {
t.Fatal("Cluster configuration must be present")
}
} else {
t.Fatal("Main configuration must be present")
}
config.Init(cfgFilePath)
archiveCfg := fmt.Sprintf("{\"kind\": \"file\",\"path\": \"%s\"}", jobarchive)
if err := archive.Init(json.RawMessage(archiveCfg), config.Keys.DisableArchive); err != nil {
@@ -177,7 +163,7 @@ func TestHandleImportFlag(t *testing.T) {
}
result := readResult(t, testname)
job, err := r.FindCached(&result.JobId, &result.Cluster, &result.StartTime)
job, err := r.Find(&result.JobId, &result.Cluster, &result.StartTime)
if err != nil {
t.Fatal(err)
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package importer
@@ -13,8 +13,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/repository"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
const (
@@ -27,20 +27,20 @@ const (
func InitDB() error {
r := repository.GetJobRepository()
if err := r.Flush(); err != nil {
cclog.Errorf("repository initDB(): %v", err)
log.Errorf("repository initDB(): %v", err)
return err
}
starttime := time.Now()
cclog.Print("Building job table...")
log.Print("Building job table...")
t, err := r.TransactionInit()
if err != nil {
cclog.Warn("Error while initializing SQL transactions")
log.Warn("Error while initializing SQL transactions")
return err
}
tags := make(map[string]int64)
// Not using cclog.Print because we want the line to end with `\r` and
// Not using log.Print because we want the line to end with `\r` and
// this function is only ever called when a special command line flag
// is passed anyways.
fmt.Printf("%d jobs inserted...\r", 0)
@@ -60,14 +60,19 @@ func InitDB() error {
}
jobMeta.MonitoringStatus = schema.MonitoringStatusArchivingSuccessful
job := schema.Job{
BaseJob: jobMeta.BaseJob,
StartTime: time.Unix(jobMeta.StartTime, 0),
StartTimeUnix: jobMeta.StartTime,
}
sc, err := archive.GetSubCluster(jobMeta.Cluster, jobMeta.SubCluster)
if err != nil {
cclog.Errorf("cannot get subcluster: %s", err.Error())
log.Errorf("cannot get subcluster: %s", err.Error())
return err
}
jobMeta.Footprint = make(map[string]float64)
job.Footprint = make(map[string]float64)
for _, fp := range sc.Footprint {
statType := "avg"
@@ -78,79 +83,71 @@ func InitDB() error {
name := fmt.Sprintf("%s_%s", fp, statType)
jobMeta.Footprint[name] = repository.LoadJobStat(jobMeta, fp, statType)
job.Footprint[name] = repository.LoadJobStat(jobMeta, fp, statType)
}
jobMeta.RawFootprint, err = json.Marshal(jobMeta.Footprint)
job.RawFootprint, err = json.Marshal(job.Footprint)
if err != nil {
cclog.Warn("Error while marshaling job footprint")
log.Warn("Error while marshaling job footprint")
return err
}
jobMeta.EnergyFootprint = make(map[string]float64)
job.EnergyFootprint = make(map[string]float64)
var totalEnergy float64
var energy float64
// Total Job Energy Outside Loop
totalEnergy := 0.0
for _, fp := range sc.EnergyFootprint {
// Always Init Metric Energy Inside Loop
metricEnergy := 0.0
if i, err := archive.MetricIndex(sc.MetricConfig, fp); err == nil {
// Note: For DB data, calculate and save as kWh
// Energy: Power (in Watts) * Time (in Seconds)
if sc.MetricConfig[i].Energy == "energy" { // this metric has energy as unit (Joules)
cclog.Warnf("Update EnergyFootprint for Job %d and Metric %s on cluster %s: Set to 'energy' in cluster.json: Not implemented, will return 0.0", jobMeta.JobID, jobMeta.Cluster, fp)
// FIXME: Needs sum as stats type
} else if sc.MetricConfig[i].Energy == "power" { // this metric has power as unit (Watt)
// Energy: Power (in Watts) * Time (in Seconds)
// Unit: (W * (s / 3600)) / 1000 = kWh
// Round 2 Digits: round(Energy * 100) / 100
// Here: (All-Node Metric Average * Number of Nodes) * (Job Duration in Seconds / 3600) / 1000
// Note: Shared Jobs handled correctly since "Node Average" is based on partial resources, while "numNodes" factor is 1
rawEnergy := ((repository.LoadJobStat(jobMeta, fp, "avg") * float64(jobMeta.NumNodes)) * (float64(jobMeta.Duration) / 3600.0)) / 1000.0
metricEnergy = math.Round(rawEnergy*100.0) / 100.0
// Unit: ( W * s ) / 3600 / 1000 = kWh ; Rounded to 2 nearest digits
energy = math.Round(((repository.LoadJobStat(jobMeta, fp, "avg")*float64(jobMeta.Duration))/3600/1000)*100) / 100
}
} else {
cclog.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, jobMeta.ID)
log.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, jobMeta.ID)
}
jobMeta.EnergyFootprint[fp] = metricEnergy
totalEnergy += metricEnergy
job.EnergyFootprint[fp] = energy
totalEnergy += energy
}
jobMeta.Energy = (math.Round(totalEnergy*100.0) / 100.0)
if jobMeta.RawEnergyFootprint, err = json.Marshal(jobMeta.EnergyFootprint); err != nil {
cclog.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
job.Energy = (math.Round(totalEnergy*100) / 100)
if job.RawEnergyFootprint, err = json.Marshal(job.EnergyFootprint); err != nil {
log.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
return err
}
jobMeta.RawResources, err = json.Marshal(jobMeta.Resources)
job.RawResources, err = json.Marshal(job.Resources)
if err != nil {
cclog.Errorf("repository initDB(): %v", err)
log.Errorf("repository initDB(): %v", err)
errorOccured++
continue
}
jobMeta.RawMetaData, err = json.Marshal(jobMeta.MetaData)
job.RawMetaData, err = json.Marshal(job.MetaData)
if err != nil {
cclog.Errorf("repository initDB(): %v", err)
log.Errorf("repository initDB(): %v", err)
errorOccured++
continue
}
if err := SanityChecks(jobMeta); err != nil {
cclog.Errorf("repository initDB(): %v", err)
if err := SanityChecks(&job.BaseJob); err != nil {
log.Errorf("repository initDB(): %v", err)
errorOccured++
continue
}
id, err := r.TransactionAddNamed(t,
repository.NamedJobInsert, jobMeta)
repository.NamedJobInsert, job)
if err != nil {
cclog.Errorf("repository initDB(): %v", err)
log.Errorf("repository initDB(): %v", err)
errorOccured++
continue
}
for _, tag := range jobMeta.Tags {
for _, tag := range job.Tags {
tagstr := tag.Name + ":" + tag.Type
tagId, ok := tags[tagstr]
if !ok {
@@ -158,7 +155,7 @@ func InitDB() error {
addTagQuery,
tag.Name, tag.Type)
if err != nil {
cclog.Errorf("Error adding tag: %v", err)
log.Errorf("Error adding tag: %v", err)
errorOccured++
continue
}
@@ -176,21 +173,21 @@ func InitDB() error {
}
if errorOccured > 0 {
cclog.Warnf("Error in import of %d jobs!", errorOccured)
log.Warnf("Error in import of %d jobs!", errorOccured)
}
r.TransactionEnd(t)
cclog.Printf("A total of %d jobs have been registered in %.3f seconds.\n", i, time.Since(starttime).Seconds())
log.Printf("A total of %d jobs have been registered in %.3f seconds.\n", i, time.Since(starttime).Seconds())
return nil
}
// This function also sets the subcluster if necessary!
func SanityChecks(job *schema.Job) error {
func SanityChecks(job *schema.BaseJob) error {
if c := archive.GetCluster(job.Cluster); c == nil {
return fmt.Errorf("no such cluster: %v", job.Cluster)
}
if err := archive.AssignSubCluster(job); err != nil {
cclog.Warn("Error while assigning subcluster to job")
log.Warn("Error while assigning subcluster to job")
return err
}
if !job.State.Valid() {

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package importer
@@ -7,7 +7,7 @@ package importer
import (
"math"
ccunits "github.com/ClusterCockpit/cc-lib/ccUnits"
ccunits "github.com/ClusterCockpit/cc-units"
)
func getNormalizationFactor(v float64) (float64, int) {

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package importer
@@ -8,7 +8,7 @@ import (
"fmt"
"testing"
ccunits "github.com/ClusterCockpit/cc-lib/ccUnits"
ccunits "github.com/ClusterCockpit/cc-units"
)
func TestNormalizeFactor(t *testing.T) {

View File

@@ -1 +1 @@
{"jobId":398955,"user":"k106eb10","project":"k106eb","cluster":"fritz","subCluster":"main","partition":"singlenode","arrayJobId":0,"numNodes":1,"numHwthreads":72,"numAcc":0,"shared":"none","monitoringStatus":1,"smt":0,"jobState":"completed","duration":260,"walltime":86340,"resources":[{"hostname":"f0720"}],"metaData":{"jobName":"ams_pipeline","jobScript":"#!/bin/bash -l\n#SBATCH --job-name=ams_pipeline\n#SBATCH --time=23:59:00\n#SBATCH --partition=singlenode\n#SBATCH --ntasks=72\n#SBATCH --hint=multithread\n#SBATCH --chdir=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\n#SBATCH --export=NONE\nunset SLURM_EXPORT_ENV\nuss=$(whoami)\nfind /dev/shm/ -user $uss -type f -mmin +30 -delete\ncd \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\"\nams_pipeline pipeline.json \u003e \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh.out\" 2\u003e \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh.err\"\n","slurmInfo":"\nJobId=398955 JobName=ams_pipeline\n UserId=k106eb10(210387) GroupId=80111\n Account=k106eb QOS=normal \n Requeue=False Restarts=0 BatchFlag=True \n TimeLimit=1439\n SubmitTime=2023-02-09T14:11:22\n Partition=singlenode \n NodeList=f0720\n NumNodes=1 NumCPUs=72 NumTasks=72 CPUs/Task=1\n NTasksPerNode:Socket:Core=0:None:None\n TRES_req=cpu=72,mem=250000M,node=1,billing=72\n TRES_alloc=cpu=72,node=1,billing=72\n Command=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh\n WorkDir=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\n StdErr=\n StdOut=ams_pipeline.o%j\n"},"startTime":1675956725,"statistics":{"clock":{"unit":{"base":"Hz","prefix":"M"},"avg":2335.254,"min":800.418,"max":2734.922},"cpu_load":{"unit":{"base":""},"avg":52.72,"min":34.46,"max":71.91},"cpu_power":{"unit":{"base":"W"},"avg":407.767,"min":93.932,"max":497.636},"cpu_user":{"unit":{"base":""},"avg":63.678,"min":19.872,"max":96.633},"flops_any":{"unit":{"base":"F/s","prefix":"G"},"avg":635.672,"min":0,"max":1332.874},"flops_dp":{"unit":{"base":"F/s","prefix":"G"},"avg":261.006,"min":0,"max":382.294},"flops_sp":{"unit":{"base":"F/s","prefix":"G"},"avg":113.659,"min":0,"max":568.286},"ib_recv":{"unit":{"base":"B/s"},"avg":27981.111,"min":69.4,"max":48084.589},"ib_recv_pkts":{"unit":{"base":"packets/s"},"avg":398.939,"min":0.5,"max":693.817},"ib_xmit":{"unit":{"base":"B/s"},"avg":188.513,"min":39.597,"max":724.568},"ib_xmit_pkts":{"unit":{"base":"packets/s"},"avg":0.867,"min":0.2,"max":2.933},"ipc":{"unit":{"base":"IPC"},"avg":0.944,"min":0.564,"max":1.291},"mem_bw":{"unit":{"base":"B/s","prefix":"G"},"avg":79.565,"min":0.021,"max":116.02},"mem_power":{"unit":{"base":"W"},"avg":24.692,"min":7.883,"max":31.318},"mem_used":{"unit":{"base":"B","prefix":"G"},"avg":22.566,"min":8.225,"max":27.613},"nfs4_read":{"unit":{"base":"B/s","prefix":"M"},"avg":647,"min":0,"max":1946},"nfs4_total":{"unit":{"base":"B/s","prefix":"M"},"avg":6181.6,"min":1270,"max":11411},"nfs4_write":{"unit":{"base":"B/s","prefix":"M"},"avg":22.4,"min":11,"max":29},"vectorization_ratio":{"unit":{"base":"%"},"avg":77.351,"min":0,"max":98.837}}}
{"jobId":398955,"user":"k106eb10","project":"k106eb","cluster":"fritz","subCluster":"main","partition":"singlenode","arrayJobId":0,"numNodes":1,"numHwthreads":72,"numAcc":0,"exclusive":1,"monitoringStatus":1,"smt":0,"jobState":"completed","duration":260,"walltime":86340,"resources":[{"hostname":"f0720"}],"metaData":{"jobName":"ams_pipeline","jobScript":"#!/bin/bash -l\n#SBATCH --job-name=ams_pipeline\n#SBATCH --time=23:59:00\n#SBATCH --partition=singlenode\n#SBATCH --ntasks=72\n#SBATCH --hint=multithread\n#SBATCH --chdir=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\n#SBATCH --export=NONE\nunset SLURM_EXPORT_ENV\nuss=$(whoami)\nfind /dev/shm/ -user $uss -type f -mmin +30 -delete\ncd \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\"\nams_pipeline pipeline.json \u003e \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh.out\" 2\u003e \"/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh.err\"\n","slurmInfo":"\nJobId=398955 JobName=ams_pipeline\n UserId=k106eb10(210387) GroupId=80111\n Account=k106eb QOS=normal \n Requeue=False Restarts=0 BatchFlag=True \n TimeLimit=1439\n SubmitTime=2023-02-09T14:11:22\n Partition=singlenode \n NodeList=f0720\n NumNodes=1 NumCPUs=72 NumTasks=72 CPUs/Task=1\n NTasksPerNode:Socket:Core=0:None:None\n TRES_req=cpu=72,mem=250000M,node=1,billing=72\n TRES_alloc=cpu=72,node=1,billing=72\n Command=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11/ams_pipeline_job.sh\n WorkDir=/home/atuin/k106eb/k106eb10/ACE/Ni-Al/DFT/VASP_PBE_500_0.125_0.1_NM/AlNi/binaries/bulk/base-hcp/occ-shaken/hcp16.occ.4.shake.0/cfg/NiAl3NiAl11\n StdErr=\n StdOut=ams_pipeline.o%j\n"},"startTime":1675956725,"statistics":{"clock":{"unit":{"base":"Hz","prefix":"M"},"avg":2335.254,"min":800.418,"max":2734.922},"cpu_load":{"unit":{"base":""},"avg":52.72,"min":34.46,"max":71.91},"cpu_power":{"unit":{"base":"W"},"avg":407.767,"min":93.932,"max":497.636},"cpu_user":{"unit":{"base":""},"avg":63.678,"min":19.872,"max":96.633},"flops_any":{"unit":{"base":"F/s","prefix":"G"},"avg":635.672,"min":0,"max":1332.874},"flops_dp":{"unit":{"base":"F/s","prefix":"G"},"avg":261.006,"min":0,"max":382.294},"flops_sp":{"unit":{"base":"F/s","prefix":"G"},"avg":113.659,"min":0,"max":568.286},"ib_recv":{"unit":{"base":"B/s"},"avg":27981.111,"min":69.4,"max":48084.589},"ib_recv_pkts":{"unit":{"base":"packets/s"},"avg":398.939,"min":0.5,"max":693.817},"ib_xmit":{"unit":{"base":"B/s"},"avg":188.513,"min":39.597,"max":724.568},"ib_xmit_pkts":{"unit":{"base":"packets/s"},"avg":0.867,"min":0.2,"max":2.933},"ipc":{"unit":{"base":"IPC"},"avg":0.944,"min":0.564,"max":1.291},"mem_bw":{"unit":{"base":"B/s","prefix":"G"},"avg":79.565,"min":0.021,"max":116.02},"mem_power":{"unit":{"base":"W"},"avg":24.692,"min":7.883,"max":31.318},"mem_used":{"unit":{"base":"B","prefix":"G"},"avg":22.566,"min":8.225,"max":27.613},"nfs4_read":{"unit":{"base":"B/s","prefix":"M"},"avg":647,"min":0,"max":1946},"nfs4_total":{"unit":{"base":"B/s","prefix":"M"},"avg":6181.6,"min":1270,"max":11411},"nfs4_write":{"unit":{"base":"B/s","prefix":"M"},"avg":22.4,"min":11,"max":29},"vectorization_ratio":{"unit":{"base":"%"},"avg":77.351,"min":0,"max":98.837}}}

View File

@@ -1 +1 @@
{"jobId":398764,"user":"k106eb10","project":"k106eb","cluster":"fritz","subCluster":"main","numNodes":1,"shared":"none","jobState":"completed","duration":177,"resources":[{"hostname":"f0649"}],"startTime":1675954353,"statistics":{"clock":{"unit":{"base":"Hz","prefix":"M"},"avg":1336.519,"min":801.564,"max":2348.215},"cpu_load":{"unit":{"base":""},"avg":31.64,"min":17.36,"max":45.54},"cpu_power":{"unit":{"base":"W"},"avg":150.018,"min":93.672,"max":261.592},"cpu_user":{"unit":{"base":""},"avg":28.518,"min":0.09,"max":57.343},"flops_any":{"unit":{"base":"F/s","prefix":"G"},"avg":45.012,"min":0,"max":135.037},"flops_dp":{"unit":{"base":"F/s","prefix":"G"},"avg":22.496,"min":0,"max":67.488},"flops_sp":{"unit":{"base":"F/s","prefix":"G"},"avg":0.02,"min":0,"max":0.061},"ib_recv":{"unit":{"base":"B/s"},"avg":14442.82,"min":219.998,"max":42581.368},"ib_recv_pkts":{"unit":{"base":"packets/s"},"avg":201.532,"min":1.25,"max":601.345},"ib_xmit":{"unit":{"base":"B/s"},"avg":282.098,"min":56.2,"max":569.363},"ib_xmit_pkts":{"unit":{"base":"packets/s"},"avg":1.228,"min":0.433,"max":2},"ipc":{"unit":{"base":"IPC"},"avg":0.77,"min":0.564,"max":0.906},"mem_bw":{"unit":{"base":"B/s","prefix":"G"},"avg":4.872,"min":0.025,"max":14.552},"mem_power":{"unit":{"base":"W"},"avg":7.725,"min":6.286,"max":10.556},"mem_used":{"unit":{"base":"B","prefix":"G"},"avg":6.162,"min":6.103,"max":6.226},"nfs4_read":{"unit":{"base":"B/s","prefix":"M"},"avg":1045.333,"min":311,"max":1525},"nfs4_total":{"unit":{"base":"B/s","prefix":"M"},"avg":6430,"min":2796,"max":11518},"nfs4_write":{"unit":{"base":"B/s","prefix":"M"},"avg":24.333,"min":0,"max":38},"vectorization_ratio":{"unit":{"base":"%"},"avg":25.528,"min":0,"max":76.585}}}
{"jobId":398764,"user":"k106eb10","project":"k106eb","cluster":"fritz","subCluster":"main","numNodes":1,"exclusive":1,"jobState":"completed","duration":177,"resources":[{"hostname":"f0649"}],"startTime":1675954353,"statistics":{"clock":{"unit":{"base":"Hz","prefix":"M"},"avg":1336.519,"min":801.564,"max":2348.215},"cpu_load":{"unit":{"base":""},"avg":31.64,"min":17.36,"max":45.54},"cpu_power":{"unit":{"base":"W"},"avg":150.018,"min":93.672,"max":261.592},"cpu_user":{"unit":{"base":""},"avg":28.518,"min":0.09,"max":57.343},"flops_any":{"unit":{"base":"F/s","prefix":"G"},"avg":45.012,"min":0,"max":135.037},"flops_dp":{"unit":{"base":"F/s","prefix":"G"},"avg":22.496,"min":0,"max":67.488},"flops_sp":{"unit":{"base":"F/s","prefix":"G"},"avg":0.02,"min":0,"max":0.061},"ib_recv":{"unit":{"base":"B/s"},"avg":14442.82,"min":219.998,"max":42581.368},"ib_recv_pkts":{"unit":{"base":"packets/s"},"avg":201.532,"min":1.25,"max":601.345},"ib_xmit":{"unit":{"base":"B/s"},"avg":282.098,"min":56.2,"max":569.363},"ib_xmit_pkts":{"unit":{"base":"packets/s"},"avg":1.228,"min":0.433,"max":2},"ipc":{"unit":{"base":"IPC"},"avg":0.77,"min":0.564,"max":0.906},"mem_bw":{"unit":{"base":"B/s","prefix":"G"},"avg":4.872,"min":0.025,"max":14.552},"mem_power":{"unit":{"base":"W"},"avg":7.725,"min":6.286,"max":10.556},"mem_used":{"unit":{"base":"B","prefix":"G"},"avg":6.162,"min":6.103,"max":6.226},"nfs4_read":{"unit":{"base":"B/s","prefix":"M"},"avg":1045.333,"min":311,"max":1525},"nfs4_total":{"unit":{"base":"B/s","prefix":"M"},"avg":6430,"min":2796,"max":11518},"nfs4_write":{"unit":{"base":"B/s","prefix":"M"},"avg":24.333,"min":0,"max":38},"vectorization_ratio":{"unit":{"base":"%"},"avg":25.528,"min":0,"max":76.585}}}

View File

@@ -1,217 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"math"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-lib/util"
)
type APIMetricData struct {
Error *string `json:"error,omitempty"`
Data schema.FloatArray `json:"data,omitempty"`
From int64 `json:"from"`
To int64 `json:"to"`
Resolution int64 `json:"resolution"`
Avg schema.Float `json:"avg"`
Min schema.Float `json:"min"`
Max schema.Float `json:"max"`
}
type APIQueryRequest struct {
Cluster string `json:"cluster"`
Queries []APIQuery `json:"queries"`
ForAllNodes []string `json:"for-all-nodes"`
From int64 `json:"from"`
To int64 `json:"to"`
WithStats bool `json:"with-stats"`
WithData bool `json:"with-data"`
WithPadding bool `json:"with-padding"`
}
type APIQueryResponse struct {
Queries []APIQuery `json:"queries,omitempty"`
Results [][]APIMetricData `json:"results"`
}
type APIQuery struct {
Type *string `json:"type,omitempty"`
SubType *string `json:"subtype,omitempty"`
Metric string `json:"metric"`
Hostname string `json:"host"`
Resolution int64 `json:"resolution"`
TypeIds []string `json:"type-ids,omitempty"`
SubTypeIds []string `json:"subtype-ids,omitempty"`
ScaleFactor schema.Float `json:"scale-by,omitempty"`
Aggregate bool `json:"aggreg"`
}
// TODO: Optimize this, just like the stats endpoint!
func (data *APIMetricData) AddStats() {
n := 0
sum, min, max := 0.0, math.MaxFloat64, -math.MaxFloat64
for _, x := range data.Data {
if x.IsNaN() {
continue
}
n += 1
sum += float64(x)
min = math.Min(min, float64(x))
max = math.Max(max, float64(x))
}
if n > 0 {
avg := sum / float64(n)
data.Avg = schema.Float(avg)
data.Min = schema.Float(min)
data.Max = schema.Float(max)
} else {
data.Avg, data.Min, data.Max = schema.NaN, schema.NaN, schema.NaN
}
}
func (data *APIMetricData) ScaleBy(f schema.Float) {
if f == 0 || f == 1 {
return
}
data.Avg *= f
data.Min *= f
data.Max *= f
for i := 0; i < len(data.Data); i++ {
data.Data[i] *= f
}
}
func (data *APIMetricData) PadDataWithNull(ms *MemoryStore, from, to int64, metric string) {
minfo, ok := ms.Metrics[metric]
if !ok {
return
}
if (data.From / minfo.Frequency) > (from / minfo.Frequency) {
padfront := int((data.From / minfo.Frequency) - (from / minfo.Frequency))
ndata := make([]schema.Float, 0, padfront+len(data.Data))
for range padfront {
ndata = append(ndata, schema.NaN)
}
for j := 0; j < len(data.Data); j++ {
ndata = append(ndata, data.Data[j])
}
data.Data = ndata
}
}
func FetchData(req APIQueryRequest) (*APIQueryResponse, error) {
req.WithData = true
req.WithData = true
req.WithData = true
ms := GetMemoryStore()
response := APIQueryResponse{
Results: make([][]APIMetricData, 0, len(req.Queries)),
}
if req.ForAllNodes != nil {
nodes := ms.ListChildren([]string{req.Cluster})
for _, node := range nodes {
for _, metric := range req.ForAllNodes {
q := APIQuery{
Metric: metric,
Hostname: node,
}
req.Queries = append(req.Queries, q)
response.Queries = append(response.Queries, q)
}
}
}
for _, query := range req.Queries {
sels := make([]util.Selector, 0, 1)
if query.Aggregate || query.Type == nil {
sel := util.Selector{{String: req.Cluster}, {String: query.Hostname}}
if query.Type != nil {
if len(query.TypeIds) == 1 {
sel = append(sel, util.SelectorElement{String: *query.Type + query.TypeIds[0]})
} else {
ids := make([]string, len(query.TypeIds))
for i, id := range query.TypeIds {
ids[i] = *query.Type + id
}
sel = append(sel, util.SelectorElement{Group: ids})
}
if query.SubType != nil {
if len(query.SubTypeIds) == 1 {
sel = append(sel, util.SelectorElement{String: *query.SubType + query.SubTypeIds[0]})
} else {
ids := make([]string, len(query.SubTypeIds))
for i, id := range query.SubTypeIds {
ids[i] = *query.SubType + id
}
sel = append(sel, util.SelectorElement{Group: ids})
}
}
}
sels = append(sels, sel)
} else {
for _, typeID := range query.TypeIds {
if query.SubType != nil {
for _, subTypeID := range query.SubTypeIds {
sels = append(sels, util.Selector{
{String: req.Cluster},
{String: query.Hostname},
{String: *query.Type + typeID},
{String: *query.SubType + subTypeID},
})
}
} else {
sels = append(sels, util.Selector{
{String: req.Cluster},
{String: query.Hostname},
{String: *query.Type + typeID},
})
}
}
}
// log.Printf("query: %#v\n", query)
// log.Printf("sels: %#v\n", sels)
var err error
res := make([]APIMetricData, 0, len(sels))
for _, sel := range sels {
data := APIMetricData{}
data.Data, data.From, data.To, data.Resolution, err = ms.Read(sel, query.Metric, req.From, req.To, query.Resolution)
if err != nil {
msg := err.Error()
data.Error = &msg
res = append(res, data)
continue
}
if req.WithStats {
data.AddStats()
}
if query.ScaleFactor != 0 {
data.ScaleBy(query.ScaleFactor)
}
if req.WithPadding {
data.PadDataWithNull(ms, req.From, req.To, query.Metric)
}
if !req.WithData {
data.Data = nil
}
res = append(res, data)
}
response.Results = append(response.Results, res)
}
return &response, nil
}

View File

@@ -1,191 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"archive/zip"
"bufio"
"context"
"errors"
"fmt"
"io"
"os"
"path/filepath"
"sync"
"sync/atomic"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
)
func Archiving(wg *sync.WaitGroup, ctx context.Context) {
go func() {
defer wg.Done()
d, err := time.ParseDuration(Keys.Archive.Interval)
if err != nil {
cclog.Fatalf("[METRICSTORE]> error parsing archive interval duration: %v\n", err)
}
if d <= 0 {
return
}
ticks := func() <-chan time.Time {
if d <= 0 {
return nil
}
return time.NewTicker(d).C
}()
for {
select {
case <-ctx.Done():
return
case <-ticks:
t := time.Now().Add(-d)
cclog.Printf("[METRICSTORE]> start archiving checkpoints (older than %s)...\n", t.Format(time.RFC3339))
n, err := ArchiveCheckpoints(Keys.Checkpoints.RootDir,
Keys.Archive.RootDir, t.Unix(), Keys.Archive.DeleteInstead)
if err != nil {
cclog.Printf("[METRICSTORE]> archiving failed: %s\n", err.Error())
} else {
cclog.Printf("[METRICSTORE]> done: %d files zipped and moved to archive\n", n)
}
}
}
}()
}
var ErrNoNewArchiveData error = errors.New("all data already archived")
// ZIP all checkpoint files older than `from` together and write them to the `archiveDir`,
// deleting them from the `checkpointsDir`.
func ArchiveCheckpoints(checkpointsDir, archiveDir string, from int64, deleteInstead bool) (int, error) {
entries1, err := os.ReadDir(checkpointsDir)
if err != nil {
return 0, err
}
type workItem struct {
cdir, adir string
cluster, host string
}
var wg sync.WaitGroup
n, errs := int32(0), int32(0)
work := make(chan workItem, NumWorkers)
wg.Add(NumWorkers)
for worker := 0; worker < NumWorkers; worker++ {
go func() {
defer wg.Done()
for workItem := range work {
m, err := archiveCheckpoints(workItem.cdir, workItem.adir, from, deleteInstead)
if err != nil {
cclog.Errorf("error while archiving %s/%s: %s", workItem.cluster, workItem.host, err.Error())
atomic.AddInt32(&errs, 1)
}
atomic.AddInt32(&n, int32(m))
}
}()
}
for _, de1 := range entries1 {
entries2, e := os.ReadDir(filepath.Join(checkpointsDir, de1.Name()))
if e != nil {
err = e
}
for _, de2 := range entries2 {
cdir := filepath.Join(checkpointsDir, de1.Name(), de2.Name())
adir := filepath.Join(archiveDir, de1.Name(), de2.Name())
work <- workItem{
adir: adir, cdir: cdir,
cluster: de1.Name(), host: de2.Name(),
}
}
}
close(work)
wg.Wait()
if err != nil {
return int(n), err
}
if errs > 0 {
return int(n), fmt.Errorf("%d errors happend while archiving (%d successes)", errs, n)
}
return int(n), nil
}
// Helper function for `ArchiveCheckpoints`.
func archiveCheckpoints(dir string, archiveDir string, from int64, deleteInstead bool) (int, error) {
entries, err := os.ReadDir(dir)
if err != nil {
return 0, err
}
extension := Keys.Checkpoints.FileFormat
files, err := findFiles(entries, from, extension, false)
if err != nil {
return 0, err
}
if deleteInstead {
n := 0
for _, checkpoint := range files {
filename := filepath.Join(dir, checkpoint)
if err = os.Remove(filename); err != nil {
return n, err
}
n += 1
}
return n, nil
}
filename := filepath.Join(archiveDir, fmt.Sprintf("%d.zip", from))
f, err := os.OpenFile(filename, os.O_CREATE|os.O_WRONLY, 0o644)
if err != nil && os.IsNotExist(err) {
err = os.MkdirAll(archiveDir, 0o755)
if err == nil {
f, err = os.OpenFile(filename, os.O_CREATE|os.O_WRONLY, 0o644)
}
}
if err != nil {
return 0, err
}
defer f.Close()
bw := bufio.NewWriter(f)
defer bw.Flush()
zw := zip.NewWriter(bw)
defer zw.Close()
n := 0
for _, checkpoint := range files {
filename := filepath.Join(dir, checkpoint)
r, err := os.Open(filename)
if err != nil {
return n, err
}
defer r.Close()
w, err := zw.Create(checkpoint)
if err != nil {
return n, err
}
if _, err = io.Copy(w, r); err != nil {
return n, err
}
if err = os.Remove(filename); err != nil {
return n, err
}
n += 1
}
return n, nil
}

View File

@@ -1,482 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"bufio"
"encoding/json"
"errors"
"fmt"
"os"
"path"
"sort"
"strconv"
"strings"
"sync"
"sync/atomic"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/linkedin/goavro/v2"
)
var NumAvroWorkers int = 4
var startUp bool = true
var ErrNoNewData error = errors.New("no data in the pool")
func (as *AvroStore) ToCheckpoint(dir string, dumpAll bool) (int, error) {
levels := make([]*AvroLevel, 0)
selectors := make([][]string, 0)
as.root.lock.RLock()
// Cluster
for sel1, l1 := range as.root.children {
l1.lock.RLock()
// Node
for sel2, l2 := range l1.children {
l2.lock.RLock()
// Frequency
for sel3, l3 := range l2.children {
levels = append(levels, l3)
selectors = append(selectors, []string{sel1, sel2, sel3})
}
l2.lock.RUnlock()
}
l1.lock.RUnlock()
}
as.root.lock.RUnlock()
type workItem struct {
level *AvroLevel
dir string
selector []string
}
n, errs := int32(0), int32(0)
var wg sync.WaitGroup
wg.Add(NumAvroWorkers)
work := make(chan workItem, NumAvroWorkers*2)
for range NumAvroWorkers {
go func() {
defer wg.Done()
for workItem := range work {
from := getTimestamp(workItem.dir)
if err := workItem.level.toCheckpoint(workItem.dir, from, dumpAll); err != nil {
if err == ErrNoNewArchiveData {
continue
}
cclog.Errorf("error while checkpointing %#v: %s", workItem.selector, err.Error())
atomic.AddInt32(&errs, 1)
} else {
atomic.AddInt32(&n, 1)
}
}
}()
}
for i := range len(levels) {
dir := path.Join(dir, path.Join(selectors[i]...))
work <- workItem{
level: levels[i],
dir: dir,
selector: selectors[i],
}
}
close(work)
wg.Wait()
if errs > 0 {
return int(n), fmt.Errorf("%d errors happend while creating avro checkpoints (%d successes)", errs, n)
}
startUp = false
return int(n), nil
}
// getTimestamp returns the timestamp from the directory name
func getTimestamp(dir string) int64 {
// Extract the resolution and timestamp from the directory name
// The existing avro file will be in epoch timestamp format
// iterate over all the files in the directory and find the maximum timestamp
// and return it
resolution := path.Base(dir)
dir = path.Dir(dir)
files, err := os.ReadDir(dir)
if err != nil {
return 0
}
var maxTS int64 = 0
if len(files) == 0 {
return 0
}
for _, file := range files {
if file.IsDir() {
continue
}
name := file.Name()
if len(name) < 5 || !strings.HasSuffix(name, ".avro") || !strings.HasPrefix(name, resolution+"_") {
continue
}
ts, err := strconv.ParseInt(name[strings.Index(name, "_")+1:len(name)-5], 10, 64)
if err != nil {
fmt.Printf("error while parsing timestamp: %s\n", err.Error())
continue
}
if ts > maxTS {
maxTS = ts
}
}
interval, _ := time.ParseDuration(Keys.Checkpoints.Interval)
updateTime := time.Unix(maxTS, 0).Add(interval).Add(time.Duration(CheckpointBufferMinutes-1) * time.Minute).Unix()
if startUp {
return 0
}
if updateTime < time.Now().Unix() {
return 0
}
return maxTS
}
func (l *AvroLevel) toCheckpoint(dir string, from int64, dumpAll bool) error {
l.lock.Lock()
defer l.lock.Unlock()
// fmt.Printf("Checkpointing directory: %s\n", dir)
// filepath contains the resolution
intRes, _ := strconv.Atoi(path.Base(dir))
// find smallest overall timestamp in l.data map and delete it from l.data
minTS := int64(1<<63 - 1)
for ts, dat := range l.data {
if ts < minTS && len(dat) != 0 {
minTS = ts
}
}
if from == 0 && minTS != int64(1<<63-1) {
from = minTS
}
if from == 0 {
return ErrNoNewArchiveData
}
var schema string
var codec *goavro.Codec
recordList := make([]map[string]any, 0)
var f *os.File
filePath := dir + fmt.Sprintf("_%d.avro", from)
var err error
fp_, err_ := os.Stat(filePath)
if errors.Is(err_, os.ErrNotExist) {
err = os.MkdirAll(path.Dir(dir), 0o755)
if err != nil {
return fmt.Errorf("failed to create directory: %v", err)
}
} else if fp_.Size() != 0 {
f, err = os.Open(filePath)
if err != nil {
return fmt.Errorf("failed to open existing avro file: %v", err)
}
br := bufio.NewReader(f)
reader, err := goavro.NewOCFReader(br)
if err != nil {
return fmt.Errorf("failed to create OCF reader: %v", err)
}
codec = reader.Codec()
schema = codec.Schema()
f.Close()
}
timeRef := time.Now().Add(time.Duration(-CheckpointBufferMinutes+1) * time.Minute).Unix()
if dumpAll {
timeRef = time.Now().Unix()
}
// Empty values
if len(l.data) == 0 {
// we checkpoint avro files every 60 seconds
repeat := 60 / intRes
for range repeat {
recordList = append(recordList, make(map[string]any))
}
}
readFlag := true
for ts := range l.data {
flag := false
if ts < timeRef {
data := l.data[ts]
schemaGen, err := generateSchema(data)
if err != nil {
return err
}
flag, schema, err = compareSchema(schema, schemaGen)
if err != nil {
return fmt.Errorf("failed to compare read and generated schema: %v", err)
}
if flag && readFlag && !errors.Is(err_, os.ErrNotExist) {
f.Close()
f, err = os.Open(filePath)
if err != nil {
return fmt.Errorf("failed to open Avro file: %v", err)
}
br := bufio.NewReader(f)
ocfReader, err := goavro.NewOCFReader(br)
if err != nil {
return fmt.Errorf("failed to create OCF reader while changing schema: %v", err)
}
for ocfReader.Scan() {
record, err := ocfReader.Read()
if err != nil {
return fmt.Errorf("failed to read record: %v", err)
}
recordList = append(recordList, record.(map[string]any))
}
f.Close()
err = os.Remove(filePath)
if err != nil {
return fmt.Errorf("failed to delete file: %v", err)
}
readFlag = false
}
codec, err = goavro.NewCodec(schema)
if err != nil {
return fmt.Errorf("failed to create codec after merged schema: %v", err)
}
recordList = append(recordList, generateRecord(data))
delete(l.data, ts)
}
}
if len(recordList) == 0 {
return ErrNoNewArchiveData
}
f, err = os.OpenFile(filePath, os.O_CREATE|os.O_APPEND|os.O_RDWR, 0o644)
if err != nil {
return fmt.Errorf("failed to append new avro file: %v", err)
}
// fmt.Printf("Codec : %#v\n", codec)
writer, err := goavro.NewOCFWriter(goavro.OCFConfig{
W: f,
Codec: codec,
CompressionName: goavro.CompressionDeflateLabel,
})
if err != nil {
return fmt.Errorf("failed to create OCF writer: %v", err)
}
// Append the new record
if err := writer.Append(recordList); err != nil {
return fmt.Errorf("failed to append record: %v", err)
}
f.Close()
return nil
}
func compareSchema(schemaRead, schemaGen string) (bool, string, error) {
var genSchema, readSchema AvroSchema
if schemaRead == "" {
return false, schemaGen, nil
}
// Unmarshal the schema strings into AvroSchema structs
if err := json.Unmarshal([]byte(schemaGen), &genSchema); err != nil {
return false, "", fmt.Errorf("failed to parse generated schema: %v", err)
}
if err := json.Unmarshal([]byte(schemaRead), &readSchema); err != nil {
return false, "", fmt.Errorf("failed to parse read schema: %v", err)
}
sort.Slice(genSchema.Fields, func(i, j int) bool {
return genSchema.Fields[i].Name < genSchema.Fields[j].Name
})
sort.Slice(readSchema.Fields, func(i, j int) bool {
return readSchema.Fields[i].Name < readSchema.Fields[j].Name
})
// Check if schemas are identical
schemasEqual := true
if len(genSchema.Fields) <= len(readSchema.Fields) {
for i := range genSchema.Fields {
if genSchema.Fields[i].Name != readSchema.Fields[i].Name {
schemasEqual = false
break
}
}
// If schemas are identical, return the read schema
if schemasEqual {
return false, schemaRead, nil
}
}
// Create a map to hold unique fields from both schemas
fieldMap := make(map[string]AvroField)
// Add fields from the read schema
for _, field := range readSchema.Fields {
fieldMap[field.Name] = field
}
// Add or update fields from the generated schema
for _, field := range genSchema.Fields {
fieldMap[field.Name] = field
}
// Create a union schema by collecting fields from the map
var mergedFields []AvroField
for _, field := range fieldMap {
mergedFields = append(mergedFields, field)
}
// Sort fields by name for consistency
sort.Slice(mergedFields, func(i, j int) bool {
return mergedFields[i].Name < mergedFields[j].Name
})
// Create the merged schema
mergedSchema := AvroSchema{
Type: "record",
Name: genSchema.Name,
Fields: mergedFields,
}
// Check if schemas are identical
schemasEqual = len(mergedSchema.Fields) == len(readSchema.Fields)
if schemasEqual {
for i := range mergedSchema.Fields {
if mergedSchema.Fields[i].Name != readSchema.Fields[i].Name {
schemasEqual = false
break
}
}
if schemasEqual {
return false, schemaRead, nil
}
}
// Marshal the merged schema back to JSON
mergedSchemaJSON, err := json.Marshal(mergedSchema)
if err != nil {
return false, "", fmt.Errorf("failed to marshal merged schema: %v", err)
}
return true, string(mergedSchemaJSON), nil
}
func generateSchema(data map[string]schema.Float) (string, error) {
// Define the Avro schema structure
schema := map[string]any{
"type": "record",
"name": "DataRecord",
"fields": []map[string]any{},
}
fieldTracker := make(map[string]struct{})
for key := range data {
if _, exists := fieldTracker[key]; !exists {
key = correctKey(key)
field := map[string]any{
"name": key,
"type": "double",
"default": -1.0,
}
schema["fields"] = append(schema["fields"].([]map[string]any), field)
fieldTracker[key] = struct{}{}
}
}
schemaString, err := json.Marshal(schema)
if err != nil {
return "", fmt.Errorf("failed to marshal schema: %v", err)
}
return string(schemaString), nil
}
func generateRecord(data map[string]schema.Float) map[string]any {
record := make(map[string]any)
// Iterate through each map in data
for key, value := range data {
key = correctKey(key)
// Set the value in the record
// avro only accepts basic types
record[key] = value.Double()
}
return record
}
func correctKey(key string) string {
// Replace any invalid characters in the key
// For example, replace spaces with underscores
key = strings.ReplaceAll(key, ":", "___")
key = strings.ReplaceAll(key, ".", "__")
return key
}
func ReplaceKey(key string) string {
// Replace any invalid characters in the key
// For example, replace spaces with underscores
key = strings.ReplaceAll(key, "___", ":")
key = strings.ReplaceAll(key, "__", ".")
return key
}

View File

@@ -1,84 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"context"
"slices"
"strconv"
"sync"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
)
func DataStaging(wg *sync.WaitGroup, ctx context.Context) {
// AvroPool is a pool of Avro writers.
go func() {
if Keys.Checkpoints.FileFormat == "json" {
wg.Done() // Mark this goroutine as done
return // Exit the goroutine
}
defer wg.Done()
var avroLevel *AvroLevel
oldSelector := make([]string, 0)
for {
select {
case <-ctx.Done():
return
case val := <-LineProtocolMessages:
// Fetch the frequency of the metric from the global configuration
freq, err := GetMetricFrequency(val.MetricName)
if err != nil {
cclog.Errorf("Error fetching metric frequency: %s\n", err)
continue
}
metricName := ""
for _, selectorName := range val.Selector {
metricName += selectorName + Delimiter
}
metricName += val.MetricName
// Create a new selector for the Avro level
// The selector is a slice of strings that represents the path to the
// Avro level. It is created by appending the cluster, node, and metric
// name to the selector.
var selector []string
selector = append(selector, val.Cluster, val.Node, strconv.FormatInt(freq, 10))
if !testEq(oldSelector, selector) {
// Get the Avro level for the metric
avroLevel = avroStore.root.findAvroLevelOrCreate(selector)
// If the Avro level is nil, create a new one
if avroLevel == nil {
cclog.Errorf("Error creating or finding the level with cluster : %s, node : %s, metric : %s\n", val.Cluster, val.Node, val.MetricName)
}
oldSelector = slices.Clone(selector)
}
avroLevel.addMetric(metricName, val.Value, val.Timestamp, int(freq))
}
}
}()
}
func testEq(a, b []string) bool {
if len(a) != len(b) {
return false
}
for i := range a {
if a[i] != b[i] {
return false
}
}
return true
}

View File

@@ -1,168 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"sync"
"github.com/ClusterCockpit/cc-lib/schema"
)
var (
LineProtocolMessages = make(chan *AvroStruct)
Delimiter = "ZZZZZ"
)
// CheckpointBufferMinutes should always be in minutes.
// Its controls the amount of data to hold for given amount of time.
var CheckpointBufferMinutes = 3
type AvroStruct struct {
MetricName string
Cluster string
Node string
Selector []string
Value schema.Float
Timestamp int64
}
type AvroStore struct {
root AvroLevel
}
var avroStore AvroStore
type AvroLevel struct {
children map[string]*AvroLevel
data map[int64]map[string]schema.Float
lock sync.RWMutex
}
type AvroField struct {
Name string `json:"name"`
Type any `json:"type"`
Default any `json:"default,omitempty"`
}
type AvroSchema struct {
Type string `json:"type"`
Name string `json:"name"`
Fields []AvroField `json:"fields"`
}
func (l *AvroLevel) findAvroLevelOrCreate(selector []string) *AvroLevel {
if len(selector) == 0 {
return l
}
// Allow concurrent reads:
l.lock.RLock()
var child *AvroLevel
var ok bool
if l.children == nil {
// Children map needs to be created...
l.lock.RUnlock()
} else {
child, ok := l.children[selector[0]]
l.lock.RUnlock()
if ok {
return child.findAvroLevelOrCreate(selector[1:])
}
}
// The level does not exist, take write lock for unqiue access:
l.lock.Lock()
// While this thread waited for the write lock, another thread
// could have created the child node.
if l.children != nil {
child, ok = l.children[selector[0]]
if ok {
l.lock.Unlock()
return child.findAvroLevelOrCreate(selector[1:])
}
}
child = &AvroLevel{
data: make(map[int64]map[string]schema.Float, 0),
children: nil,
}
if l.children != nil {
l.children[selector[0]] = child
} else {
l.children = map[string]*AvroLevel{selector[0]: child}
}
l.lock.Unlock()
return child.findAvroLevelOrCreate(selector[1:])
}
func (l *AvroLevel) addMetric(metricName string, value schema.Float, timestamp int64, Freq int) {
l.lock.Lock()
defer l.lock.Unlock()
KeyCounter := int(CheckpointBufferMinutes * 60 / Freq)
// Create keys in advance for the given amount of time
if len(l.data) != KeyCounter {
if len(l.data) == 0 {
for i := range KeyCounter {
l.data[timestamp+int64(i*Freq)] = make(map[string]schema.Float, 0)
}
} else {
// Get the last timestamp
var lastTS int64
for ts := range l.data {
if ts > lastTS {
lastTS = ts
}
}
// Create keys for the next KeyCounter timestamps
l.data[lastTS+int64(Freq)] = make(map[string]schema.Float, 0)
}
}
closestTS := int64(0)
minDiff := int64(Freq) + 1 // Start with diff just outside the valid range
found := false
// Iterate over timestamps and choose the one which is within range.
// Since its epoch time, we check if the difference is less than 60 seconds.
for ts, dat := range l.data {
// Check if timestamp is within range
diff := timestamp - ts
if diff < -int64(Freq) || diff > int64(Freq) {
continue
}
// Metric already present at this timestamp — skip
if _, ok := dat[metricName]; ok {
continue
}
// Check if this is the closest timestamp so far
if Abs(diff) < minDiff {
minDiff = Abs(diff)
closestTS = ts
found = true
}
}
if found {
l.data[closestTS][metricName] = value
}
}
func GetAvroStore() *AvroStore {
return &avroStore
}
// Abs returns the absolute value of x.
func Abs(x int64) int64 {
if x < 0 {
return -x
}
return x
}

View File

@@ -1,238 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"errors"
"sync"
"github.com/ClusterCockpit/cc-lib/schema"
)
// Default buffer capacity.
// `buffer.data` will only ever grow up to it's capacity and a new link
// in the buffer chain will be created if needed so that no copying
// of data or reallocation needs to happen on writes.
const (
BufferCap int = 512
)
// So that we can reuse allocations
var bufferPool sync.Pool = sync.Pool{
New: func() any {
return &buffer{
data: make([]schema.Float, 0, BufferCap),
}
},
}
var (
ErrNoData error = errors.New("[METRICSTORE]> no data for this metric/level")
ErrDataDoesNotAlign error = errors.New("[METRICSTORE]> data from lower granularities does not align")
)
// Each metric on each level has it's own buffer.
// This is where the actual values go.
// If `cap(data)` is reached, a new buffer is created and
// becomes the new head of a buffer list.
type buffer struct {
prev *buffer
next *buffer
data []schema.Float
frequency int64
start int64
archived bool
closed bool
}
func newBuffer(ts, freq int64) *buffer {
b := bufferPool.Get().(*buffer)
b.frequency = freq
b.start = ts - (freq / 2)
b.prev = nil
b.next = nil
b.archived = false
b.closed = false
b.data = b.data[:0]
return b
}
// If a new buffer was created, the new head is returnd.
// Otherwise, the existing buffer is returnd.
// Normaly, only "newer" data should be written, but if the value would
// end up in the same buffer anyways it is allowed.
func (b *buffer) write(ts int64, value schema.Float) (*buffer, error) {
if ts < b.start {
return nil, errors.New("[METRICSTORE]> cannot write value to buffer from past")
}
// idx := int((ts - b.start + (b.frequency / 3)) / b.frequency)
idx := int((ts - b.start) / b.frequency)
if idx >= cap(b.data) {
newbuf := newBuffer(ts, b.frequency)
newbuf.prev = b
b.next = newbuf
b.close()
b = newbuf
idx = 0
}
// Overwriting value or writing value from past
if idx < len(b.data) {
b.data[idx] = value
return b, nil
}
// Fill up unwritten slots with NaN
for i := len(b.data); i < idx; i++ {
b.data = append(b.data, schema.NaN)
}
b.data = append(b.data, value)
return b, nil
}
func (b *buffer) end() int64 {
return b.firstWrite() + int64(len(b.data))*b.frequency
}
func (b *buffer) firstWrite() int64 {
return b.start + (b.frequency / 2)
}
func (b *buffer) close() {}
/*
func (b *buffer) close() {
if b.closed {
return
}
b.closed = true
n, sum, min, max := 0, 0., math.MaxFloat64, -math.MaxFloat64
for _, x := range b.data {
if x.IsNaN() {
continue
}
n += 1
f := float64(x)
sum += f
min = math.Min(min, f)
max = math.Max(max, f)
}
b.statisticts.samples = n
if n > 0 {
b.statisticts.avg = Float(sum / float64(n))
b.statisticts.min = Float(min)
b.statisticts.max = Float(max)
} else {
b.statisticts.avg = NaN
b.statisticts.min = NaN
b.statisticts.max = NaN
}
}
*/
// func interpolate(idx int, data []Float) Float {
// if idx == 0 || idx+1 == len(data) {
// return NaN
// }
// return (data[idx-1] + data[idx+1]) / 2.0
// }
// Return all known values from `from` to `to`. Gaps of information are represented as NaN.
// Simple linear interpolation is done between the two neighboring cells if possible.
// If values at the start or end are missing, instead of NaN values, the second and thrid
// return values contain the actual `from`/`to`.
// This function goes back the buffer chain if `from` is older than the currents buffer start.
// The loaded values are added to `data` and `data` is returned, possibly with a shorter length.
// If `data` is not long enough to hold all values, this function will panic!
func (b *buffer) read(from, to int64, data []schema.Float) ([]schema.Float, int64, int64, error) {
if from < b.firstWrite() {
if b.prev != nil {
return b.prev.read(from, to, data)
}
from = b.firstWrite()
}
i := 0
t := from
for ; t < to; t += b.frequency {
idx := int((t - b.start) / b.frequency)
if idx >= cap(b.data) {
if b.next == nil {
break
}
b = b.next
idx = 0
}
if idx >= len(b.data) {
if b.next == nil || to <= b.next.start {
break
}
data[i] += schema.NaN
} else if t < b.start {
data[i] += schema.NaN
// } else if b.data[idx].IsNaN() {
// data[i] += interpolate(idx, b.data)
} else {
data[i] += b.data[idx]
}
i++
}
return data[:i], from, t, nil
}
// Returns true if this buffer needs to be freed.
func (b *buffer) free(t int64) (delme bool, n int) {
if b.prev != nil {
delme, m := b.prev.free(t)
n += m
if delme {
b.prev.next = nil
if cap(b.prev.data) == BufferCap {
bufferPool.Put(b.prev)
}
b.prev = nil
}
}
end := b.end()
if end < t {
return true, n + 1
}
return false, n
}
// Call `callback` on every buffer that contains data in the range from `from` to `to`.
func (b *buffer) iterFromTo(from, to int64, callback func(b *buffer) error) error {
if b == nil {
return nil
}
if err := b.prev.iterFromTo(from, to, callback); err != nil {
return err
}
if from <= b.end() && b.start <= to {
return callback(b)
}
return nil
}
func (b *buffer) count() int64 {
res := int64(len(b.data))
if b.prev != nil {
res += b.prev.count()
}
return res
}

View File

@@ -1,768 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"bufio"
"context"
"encoding/json"
"errors"
"fmt"
"io/fs"
"os"
"path"
"path/filepath"
"runtime"
"sort"
"strconv"
"strings"
"sync"
"sync/atomic"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/linkedin/goavro/v2"
)
// Whenever changed, update MarshalJSON as well!
type CheckpointMetrics struct {
Data []schema.Float `json:"data"`
Frequency int64 `json:"frequency"`
Start int64 `json:"start"`
}
type CheckpointFile struct {
Metrics map[string]*CheckpointMetrics `json:"metrics"`
Children map[string]*CheckpointFile `json:"children"`
From int64 `json:"from"`
To int64 `json:"to"`
}
var lastCheckpoint time.Time
func Checkpointing(wg *sync.WaitGroup, ctx context.Context) {
lastCheckpoint = time.Now()
if Keys.Checkpoints.FileFormat == "json" {
ms := GetMemoryStore()
go func() {
defer wg.Done()
d, err := time.ParseDuration(Keys.Checkpoints.Interval)
if err != nil {
cclog.Fatal(err)
}
if d <= 0 {
return
}
ticks := func() <-chan time.Time {
if d <= 0 {
return nil
}
return time.NewTicker(d).C
}()
for {
select {
case <-ctx.Done():
return
case <-ticks:
cclog.Printf("[METRICSTORE]> start checkpointing (starting at %s)...\n", lastCheckpoint.Format(time.RFC3339))
now := time.Now()
n, err := ms.ToCheckpoint(Keys.Checkpoints.RootDir,
lastCheckpoint.Unix(), now.Unix())
if err != nil {
cclog.Printf("[METRICSTORE]> checkpointing failed: %s\n", err.Error())
} else {
cclog.Printf("[METRICSTORE]> done: %d checkpoint files created\n", n)
lastCheckpoint = now
}
}
}
}()
} else {
go func() {
defer wg.Done()
d, _ := time.ParseDuration("1m")
select {
case <-ctx.Done():
return
case <-time.After(time.Duration(CheckpointBufferMinutes) * time.Minute):
// This is the first tick untill we collect the data for given minutes.
GetAvroStore().ToCheckpoint(Keys.Checkpoints.RootDir, false)
// log.Printf("Checkpointing %d avro files", count)
}
ticks := func() <-chan time.Time {
if d <= 0 {
return nil
}
return time.NewTicker(d).C
}()
for {
select {
case <-ctx.Done():
return
case <-ticks:
// Regular ticks of 1 minute to write data.
GetAvroStore().ToCheckpoint(Keys.Checkpoints.RootDir, false)
// log.Printf("Checkpointing %d avro files", count)
}
}
}()
}
}
// As `Float` implements a custom MarshalJSON() function,
// serializing an array of such types has more overhead
// than one would assume (because of extra allocations, interfaces and so on).
func (cm *CheckpointMetrics) MarshalJSON() ([]byte, error) {
buf := make([]byte, 0, 128+len(cm.Data)*8)
buf = append(buf, `{"frequency":`...)
buf = strconv.AppendInt(buf, cm.Frequency, 10)
buf = append(buf, `,"start":`...)
buf = strconv.AppendInt(buf, cm.Start, 10)
buf = append(buf, `,"data":[`...)
for i, x := range cm.Data {
if i != 0 {
buf = append(buf, ',')
}
if x.IsNaN() {
buf = append(buf, `null`...)
} else {
buf = strconv.AppendFloat(buf, float64(x), 'f', 1, 32)
}
}
buf = append(buf, `]}`...)
return buf, nil
}
// Metrics stored at the lowest 2 levels are not stored away (root and cluster)!
// On a per-host basis a new JSON file is created. I have no idea if this will scale.
// The good thing: Only a host at a time is locked, so this function can run
// in parallel to writes/reads.
func (m *MemoryStore) ToCheckpoint(dir string, from, to int64) (int, error) {
levels := make([]*Level, 0)
selectors := make([][]string, 0)
m.root.lock.RLock()
for sel1, l1 := range m.root.children {
l1.lock.RLock()
for sel2, l2 := range l1.children {
levels = append(levels, l2)
selectors = append(selectors, []string{sel1, sel2})
}
l1.lock.RUnlock()
}
m.root.lock.RUnlock()
type workItem struct {
level *Level
dir string
selector []string
}
n, errs := int32(0), int32(0)
var wg sync.WaitGroup
wg.Add(NumWorkers)
work := make(chan workItem, NumWorkers*2)
for worker := 0; worker < NumWorkers; worker++ {
go func() {
defer wg.Done()
for workItem := range work {
if err := workItem.level.toCheckpoint(workItem.dir, from, to, m); err != nil {
if err == ErrNoNewArchiveData {
continue
}
cclog.Printf("[METRICSTORE]> error while checkpointing %#v: %s", workItem.selector, err.Error())
atomic.AddInt32(&errs, 1)
} else {
atomic.AddInt32(&n, 1)
}
}
}()
}
for i := 0; i < len(levels); i++ {
dir := path.Join(dir, path.Join(selectors[i]...))
work <- workItem{
level: levels[i],
dir: dir,
selector: selectors[i],
}
}
close(work)
wg.Wait()
if errs > 0 {
return int(n), fmt.Errorf("[METRICSTORE]> %d errors happend while creating checkpoints (%d successes)", errs, n)
}
return int(n), nil
}
func (l *Level) toCheckpointFile(from, to int64, m *MemoryStore) (*CheckpointFile, error) {
l.lock.RLock()
defer l.lock.RUnlock()
retval := &CheckpointFile{
From: from,
To: to,
Metrics: make(map[string]*CheckpointMetrics),
Children: make(map[string]*CheckpointFile),
}
for metric, minfo := range m.Metrics {
b := l.metrics[minfo.offset]
if b == nil {
continue
}
allArchived := true
b.iterFromTo(from, to, func(b *buffer) error {
if !b.archived {
allArchived = false
}
return nil
})
if allArchived {
continue
}
data := make([]schema.Float, (to-from)/b.frequency+1)
data, start, end, err := b.read(from, to, data)
if err != nil {
return nil, err
}
for i := int((end - start) / b.frequency); i < len(data); i++ {
data[i] = schema.NaN
}
retval.Metrics[metric] = &CheckpointMetrics{
Frequency: b.frequency,
Start: start,
Data: data,
}
}
for name, child := range l.children {
val, err := child.toCheckpointFile(from, to, m)
if err != nil {
return nil, err
}
if val != nil {
retval.Children[name] = val
}
}
if len(retval.Children) == 0 && len(retval.Metrics) == 0 {
return nil, nil
}
return retval, nil
}
func (l *Level) toCheckpoint(dir string, from, to int64, m *MemoryStore) error {
cf, err := l.toCheckpointFile(from, to, m)
if err != nil {
return err
}
if cf == nil {
return ErrNoNewArchiveData
}
filepath := path.Join(dir, fmt.Sprintf("%d.json", from))
f, err := os.OpenFile(filepath, os.O_CREATE|os.O_WRONLY, 0o644)
if err != nil && os.IsNotExist(err) {
err = os.MkdirAll(dir, 0o755)
if err == nil {
f, err = os.OpenFile(filepath, os.O_CREATE|os.O_WRONLY, 0o644)
}
}
if err != nil {
return err
}
defer f.Close()
bw := bufio.NewWriter(f)
if err = json.NewEncoder(bw).Encode(cf); err != nil {
return err
}
return bw.Flush()
}
func (m *MemoryStore) FromCheckpoint(dir string, from int64, extension string) (int, error) {
var wg sync.WaitGroup
work := make(chan [2]string, NumWorkers)
n, errs := int32(0), int32(0)
wg.Add(NumWorkers)
for worker := 0; worker < NumWorkers; worker++ {
go func() {
defer wg.Done()
for host := range work {
lvl := m.root.findLevelOrCreate(host[:], len(m.Metrics))
nn, err := lvl.fromCheckpoint(m, filepath.Join(dir, host[0], host[1]), from, extension)
if err != nil {
cclog.Fatalf("[METRICSTORE]> error while loading checkpoints: %s", err.Error())
atomic.AddInt32(&errs, 1)
}
atomic.AddInt32(&n, int32(nn))
}
}()
}
i := 0
clustersDir, err := os.ReadDir(dir)
for _, clusterDir := range clustersDir {
if !clusterDir.IsDir() {
err = errors.New("[METRICSTORE]> expected only directories at first level of checkpoints/ directory")
goto done
}
hostsDir, e := os.ReadDir(filepath.Join(dir, clusterDir.Name()))
if e != nil {
err = e
goto done
}
for _, hostDir := range hostsDir {
if !hostDir.IsDir() {
err = errors.New("[METRICSTORE]> expected only directories at second level of checkpoints/ directory")
goto done
}
i++
if i%NumWorkers == 0 && i > 100 {
// Forcing garbage collection runs here regulary during the loading of checkpoints
// will decrease the total heap size after loading everything back to memory is done.
// While loading data, the heap will grow fast, so the GC target size will double
// almost always. By forcing GCs here, we can keep it growing more slowly so that
// at the end, less memory is wasted.
runtime.GC()
}
work <- [2]string{clusterDir.Name(), hostDir.Name()}
}
}
done:
close(work)
wg.Wait()
if err != nil {
return int(n), err
}
if errs > 0 {
return int(n), fmt.Errorf("[METRICSTORE]> %d errors happend while creating checkpoints (%d successes)", errs, n)
}
return int(n), nil
}
// Metrics stored at the lowest 2 levels are not loaded (root and cluster)!
// This function can only be called once and before the very first write or read.
// Different host's data is loaded to memory in parallel.
func (m *MemoryStore) FromCheckpointFiles(dir string, from int64) (int, error) {
if _, err := os.Stat(dir); os.IsNotExist(err) {
// The directory does not exist, so create it using os.MkdirAll()
err := os.MkdirAll(dir, 0o755) // 0755 sets the permissions for the directory
if err != nil {
cclog.Fatalf("[METRICSTORE]> Error creating directory: %#v\n", err)
}
cclog.Printf("[METRICSTORE]> %#v Directory created successfully.\n", dir)
}
// Config read (replace with your actual config read)
fileFormat := Keys.Checkpoints.FileFormat
if fileFormat == "" {
fileFormat = "avro"
}
// Map to easily get the fallback format
oppositeFormat := map[string]string{
"json": "avro",
"avro": "json",
}
// First, attempt to load the specified format
if found, err := checkFilesWithExtension(dir, fileFormat); err != nil {
return 0, fmt.Errorf("[METRICSTORE]> error checking files with extension: %v", err)
} else if found {
cclog.Printf("[METRICSTORE]> Loading %s files because fileformat is %s\n", fileFormat, fileFormat)
return m.FromCheckpoint(dir, from, fileFormat)
}
// If not found, attempt the opposite format
altFormat := oppositeFormat[fileFormat]
if found, err := checkFilesWithExtension(dir, altFormat); err != nil {
return 0, fmt.Errorf("[METRICSTORE]> error checking files with extension: %v", err)
} else if found {
cclog.Printf("[METRICSTORE]> Loading %s files but fileformat is %s\n", altFormat, fileFormat)
return m.FromCheckpoint(dir, from, altFormat)
}
cclog.Print("[METRICSTORE]> No valid checkpoint files found in the directory")
return 0, nil
}
func checkFilesWithExtension(dir string, extension string) (bool, error) {
found := false
err := filepath.Walk(dir, func(path string, info os.FileInfo, err error) error {
if err != nil {
return fmt.Errorf("[METRICSTORE]> error accessing path %s: %v", path, err)
}
if !info.IsDir() && filepath.Ext(info.Name()) == "."+extension {
found = true
return nil
}
return nil
})
if err != nil {
return false, fmt.Errorf("[METRICSTORE]> error walking through directories: %s", err)
}
return found, nil
}
func (l *Level) loadAvroFile(m *MemoryStore, f *os.File, from int64) error {
br := bufio.NewReader(f)
fileName := f.Name()[strings.LastIndex(f.Name(), "/")+1:]
resolution, err := strconv.ParseInt(fileName[0:strings.Index(fileName, "_")], 10, 64)
if err != nil {
return fmt.Errorf("[METRICSTORE]> error while reading avro file (resolution parsing) : %s", err)
}
fromTimestamp, err := strconv.ParseInt(fileName[strings.Index(fileName, "_")+1:len(fileName)-5], 10, 64)
// Same logic according to lineprotocol
fromTimestamp -= (resolution / 2)
if err != nil {
return fmt.Errorf("[METRICSTORE]> error converting timestamp from the avro file : %s", err)
}
// fmt.Printf("File : %s with resolution : %d\n", fileName, resolution)
var recordCounter int64 = 0
// Create a new OCF reader from the buffered reader
ocfReader, err := goavro.NewOCFReader(br)
if err != nil {
panic(err)
}
metricsData := make(map[string]schema.FloatArray)
for ocfReader.Scan() {
datum, err := ocfReader.Read()
if err != nil {
return fmt.Errorf("[METRICSTORE]> error while reading avro file : %s", err)
}
record, ok := datum.(map[string]any)
if !ok {
panic("[METRICSTORE]> failed to assert datum as map[string]interface{}")
}
for key, value := range record {
metricsData[key] = append(metricsData[key], schema.ConvertToFloat(value.(float64)))
}
recordCounter += 1
}
to := (fromTimestamp + (recordCounter / (60 / resolution) * 60))
if to < from {
return nil
}
for key, floatArray := range metricsData {
metricName := ReplaceKey(key)
if strings.Contains(metricName, Delimiter) {
subString := strings.Split(metricName, Delimiter)
lvl := l
for i := 0; i < len(subString)-1; i++ {
sel := subString[i]
if lvl.children == nil {
lvl.children = make(map[string]*Level)
}
child, ok := lvl.children[sel]
if !ok {
child = &Level{
metrics: make([]*buffer, len(m.Metrics)),
children: nil,
}
lvl.children[sel] = child
}
lvl = child
}
leafMetricName := subString[len(subString)-1]
err = lvl.createBuffer(m, leafMetricName, floatArray, fromTimestamp, resolution)
if err != nil {
return fmt.Errorf("[METRICSTORE]> error while creating buffers from avroReader : %s", err)
}
} else {
err = l.createBuffer(m, metricName, floatArray, fromTimestamp, resolution)
if err != nil {
return fmt.Errorf("[METRICSTORE]> error while creating buffers from avroReader : %s", err)
}
}
}
return nil
}
func (l *Level) createBuffer(m *MemoryStore, metricName string, floatArray schema.FloatArray, from int64, resolution int64) error {
n := len(floatArray)
b := &buffer{
frequency: resolution,
start: from,
data: floatArray[0:n:n],
prev: nil,
next: nil,
archived: true,
}
b.close()
minfo, ok := m.Metrics[metricName]
if !ok {
return nil
// return errors.New("Unkown metric: " + name)
}
prev := l.metrics[minfo.offset]
if prev == nil {
l.metrics[minfo.offset] = b
} else {
if prev.start > b.start {
return errors.New("wooops")
}
b.prev = prev
prev.next = b
missingCount := ((int(b.start) - int(prev.start)) - len(prev.data)*int(b.frequency))
if missingCount > 0 {
missingCount /= int(b.frequency)
for range missingCount {
prev.data = append(prev.data, schema.NaN)
}
prev.data = prev.data[0:len(prev.data):len(prev.data)]
}
}
l.metrics[minfo.offset] = b
return nil
}
func (l *Level) loadJSONFile(m *MemoryStore, f *os.File, from int64) error {
br := bufio.NewReader(f)
cf := &CheckpointFile{}
if err := json.NewDecoder(br).Decode(cf); err != nil {
return err
}
if cf.To != 0 && cf.To < from {
return nil
}
if err := l.loadFile(cf, m); err != nil {
return err
}
return nil
}
func (l *Level) loadFile(cf *CheckpointFile, m *MemoryStore) error {
for name, metric := range cf.Metrics {
n := len(metric.Data)
b := &buffer{
frequency: metric.Frequency,
start: metric.Start,
data: metric.Data[0:n:n], // Space is wasted here :(
prev: nil,
next: nil,
archived: true,
}
b.close()
minfo, ok := m.Metrics[name]
if !ok {
continue
// return errors.New("Unkown metric: " + name)
}
prev := l.metrics[minfo.offset]
if prev == nil {
l.metrics[minfo.offset] = b
} else {
if prev.start > b.start {
return errors.New("wooops")
}
b.prev = prev
prev.next = b
}
l.metrics[minfo.offset] = b
}
if len(cf.Children) > 0 && l.children == nil {
l.children = make(map[string]*Level)
}
for sel, childCf := range cf.Children {
child, ok := l.children[sel]
if !ok {
child = &Level{
metrics: make([]*buffer, len(m.Metrics)),
children: nil,
}
l.children[sel] = child
}
if err := child.loadFile(childCf, m); err != nil {
return err
}
}
return nil
}
func (l *Level) fromCheckpoint(m *MemoryStore, dir string, from int64, extension string) (int, error) {
direntries, err := os.ReadDir(dir)
if err != nil {
if os.IsNotExist(err) {
return 0, nil
}
return 0, err
}
allFiles := make([]fs.DirEntry, 0)
filesLoaded := 0
for _, e := range direntries {
if e.IsDir() {
child := &Level{
metrics: make([]*buffer, len(m.Metrics)),
children: make(map[string]*Level),
}
files, err := child.fromCheckpoint(m, path.Join(dir, e.Name()), from, extension)
filesLoaded += files
if err != nil {
return filesLoaded, err
}
l.children[e.Name()] = child
} else if strings.HasSuffix(e.Name(), "."+extension) {
allFiles = append(allFiles, e)
} else {
continue
}
}
files, err := findFiles(allFiles, from, extension, true)
if err != nil {
return filesLoaded, err
}
loaders := map[string]func(*MemoryStore, *os.File, int64) error{
"json": l.loadJSONFile,
"avro": l.loadAvroFile,
}
loader := loaders[extension]
for _, filename := range files {
f, err := os.Open(path.Join(dir, filename))
if err != nil {
return filesLoaded, err
}
defer f.Close()
if err = loader(m, f, from); err != nil {
return filesLoaded, err
}
filesLoaded += 1
}
return filesLoaded, nil
}
// This will probably get very slow over time!
// A solution could be some sort of an index file in which all other files
// and the timespan they contain is listed.
func findFiles(direntries []fs.DirEntry, t int64, extension string, findMoreRecentFiles bool) ([]string, error) {
nums := map[string]int64{}
for _, e := range direntries {
if !strings.HasSuffix(e.Name(), "."+extension) {
continue
}
ts, err := strconv.ParseInt(e.Name()[strings.Index(e.Name(), "_")+1:len(e.Name())-5], 10, 64)
if err != nil {
return nil, err
}
nums[e.Name()] = ts
}
sort.Slice(direntries, func(i, j int) bool {
a, b := direntries[i], direntries[j]
return nums[a.Name()] < nums[b.Name()]
})
filenames := make([]string, 0)
for i := range direntries {
e := direntries[i]
ts1 := nums[e.Name()]
if findMoreRecentFiles && t <= ts1 {
filenames = append(filenames, e.Name())
}
if i == len(direntries)-1 {
continue
}
enext := direntries[i+1]
ts2 := nums[enext.Name()]
if findMoreRecentFiles {
if ts1 < t && t < ts2 {
filenames = append(filenames, e.Name())
}
} else {
if ts2 < t {
filenames = append(filenames, e.Name())
}
}
}
return filenames, nil
}

View File

@@ -1,118 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"fmt"
)
var InternalCCMSFlag bool = false
type MetricStoreConfig struct {
Checkpoints struct {
FileFormat string `json:"file-format"`
Interval string `json:"interval"`
RootDir string `json:"directory"`
Restore string `json:"restore"`
} `json:"checkpoints"`
Debug struct {
DumpToFile string `json:"dump-to-file"`
EnableGops bool `json:"gops"`
} `json:"debug"`
RetentionInMemory string `json:"retention-in-memory"`
Archive struct {
Interval string `json:"interval"`
RootDir string `json:"directory"`
DeleteInstead bool `json:"delete-instead"`
} `json:"archive"`
Nats []*NatsConfig `json:"nats"`
}
type NatsConfig struct {
// Address of the nats server
Address string `json:"address"`
// Username/Password, optional
Username string `json:"username"`
Password string `json:"password"`
// Creds file path
Credsfilepath string `json:"creds-file-path"`
Subscriptions []struct {
// Channel name
SubscribeTo string `json:"subscribe-to"`
// Allow lines without a cluster tag, use this as default, optional
ClusterTag string `json:"cluster-tag"`
} `json:"subscriptions"`
}
var Keys MetricStoreConfig
// AggregationStrategy for aggregation over multiple values at different cpus/sockets/..., not time!
type AggregationStrategy int
const (
NoAggregation AggregationStrategy = iota
SumAggregation
AvgAggregation
)
func AssignAggregationStratergy(str string) (AggregationStrategy, error) {
switch str {
case "":
return NoAggregation, nil
case "sum":
return SumAggregation, nil
case "avg":
return AvgAggregation, nil
default:
return NoAggregation, fmt.Errorf("[METRICSTORE]> unknown aggregation strategy: %s", str)
}
}
type MetricConfig struct {
// Interval in seconds at which measurements are stored
Frequency int64
// Can be 'sum', 'avg' or null. Describes how to aggregate metrics from the same timestep over the hierarchy.
Aggregation AggregationStrategy
// Private, used internally...
offset int
}
var Metrics map[string]MetricConfig
func GetMetricFrequency(metricName string) (int64, error) {
if metric, ok := Metrics[metricName]; ok {
return metric.Frequency, nil
}
return 0, fmt.Errorf("[METRICSTORE]> metric %s not found", metricName)
}
// AddMetric adds logic to add metrics. Redundant metrics should be updated with max frequency.
// use metric.Name to check if the metric already exists.
// if not, add it to the Metrics map.
func AddMetric(name string, metric MetricConfig) error {
if Metrics == nil {
Metrics = make(map[string]MetricConfig, 0)
}
if existingMetric, ok := Metrics[name]; ok {
if existingMetric.Frequency != metric.Frequency {
if existingMetric.Frequency < metric.Frequency {
existingMetric.Frequency = metric.Frequency
Metrics[name] = existingMetric
}
}
} else {
Metrics[name] = metric
}
return nil
}

View File

@@ -1,89 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
const configSchema = `{
"type": "object",
"description": "Configuration specific to built-in metric-store.",
"properties": {
"checkpoints": {
"description": "Configuration for checkpointing the metrics within metric-store",
"type": "object",
"properties": {
"file-format": {
"description": "Specify the type of checkpoint file. There are 2 variants: 'avro' and 'json'. If nothing is specified, 'avro' is default.",
"type": "string"
},
"interval": {
"description": "Interval at which the metrics should be checkpointed.",
"type": "string"
},
"directory": {
"description": "Specify the parent directy in which the checkpointed files should be placed.",
"type": "string"
},
"restore": {
"description": "When cc-backend starts up, look for checkpointed files that are less than X hours old and load metrics from these selected checkpoint files.",
"type": "string"
}
}
},
"archive": {
"description": "Configuration for archiving the already checkpointed files.",
"type": "object",
"properties": {
"interval": {
"description": "Interval at which the checkpointed files should be archived.",
"type": "string"
},
"directory": {
"description": "Specify the parent directy in which the archived files should be placed.",
"type": "string"
}
}
},
"retention-in-memory": {
"description": "Keep the metrics within memory for given time interval. Retention for X hours, then the metrics would be freed.",
"type": "string"
},
"nats": {
"description": "Configuration for accepting published data through NATS.",
"type": "object",
"properties": {
"address": {
"description": "Address of the NATS server.",
"type": "string"
},
"username": {
"description": "Optional: If configured with username/password method.",
"type": "string"
},
"password": {
"description": "Optional: If configured with username/password method.",
"type": "string"
},
"creds-file-path": {
"description": "Optional: If configured with Credential File method. Path to your NATS cred file.",
"type": "string"
},
"subscriptions": {
"description": "Array of various subscriptions. Allows to subscibe to different subjects and publishers.",
"type": "object",
"properties": {
"subscribe-to": {
"description": "Channel name",
"type": "string"
},
"cluster-tag": {
"description": "Optional: Allow lines without a cluster tag, use this as default",
"type": "string"
}
}
}
}
}
}
}`

View File

@@ -1,112 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"bufio"
"fmt"
"strconv"
)
func (b *buffer) debugDump(buf []byte) []byte {
if b.prev != nil {
buf = b.prev.debugDump(buf)
}
start, len, end := b.start, len(b.data), b.start+b.frequency*int64(len(b.data))
buf = append(buf, `{"start":`...)
buf = strconv.AppendInt(buf, start, 10)
buf = append(buf, `,"len":`...)
buf = strconv.AppendInt(buf, int64(len), 10)
buf = append(buf, `,"end":`...)
buf = strconv.AppendInt(buf, end, 10)
if b.archived {
buf = append(buf, `,"saved":true`...)
}
if b.next != nil {
buf = append(buf, `},`...)
} else {
buf = append(buf, `}`...)
}
return buf
}
func (l *Level) debugDump(m *MemoryStore, w *bufio.Writer, lvlname string, buf []byte, depth int) ([]byte, error) {
l.lock.RLock()
defer l.lock.RUnlock()
for i := 0; i < depth; i++ {
buf = append(buf, '\t')
}
buf = append(buf, '"')
buf = append(buf, lvlname...)
buf = append(buf, "\":{\n"...)
depth += 1
objitems := 0
for name, mc := range m.Metrics {
if b := l.metrics[mc.offset]; b != nil {
for i := 0; i < depth; i++ {
buf = append(buf, '\t')
}
buf = append(buf, '"')
buf = append(buf, name...)
buf = append(buf, `":[`...)
buf = b.debugDump(buf)
buf = append(buf, "],\n"...)
objitems++
}
}
for name, lvl := range l.children {
_, err := w.Write(buf)
if err != nil {
return nil, err
}
buf = buf[0:0]
buf, err = lvl.debugDump(m, w, name, buf, depth)
if err != nil {
return nil, err
}
buf = append(buf, ',', '\n')
objitems++
}
// remove final `,`:
if objitems > 0 {
buf = append(buf[0:len(buf)-1], '\n')
}
depth -= 1
for i := 0; i < depth; i++ {
buf = append(buf, '\t')
}
buf = append(buf, '}')
return buf, nil
}
func (m *MemoryStore) DebugDump(w *bufio.Writer, selector []string) error {
lvl := m.root.findLevel(selector)
if lvl == nil {
return fmt.Errorf("[METRICSTORE]> not found: %#v", selector)
}
buf := make([]byte, 0, 2048)
buf = append(buf, "{"...)
buf, err := lvl.debugDump(m, w, "data", buf, 0)
if err != nil {
return err
}
buf = append(buf, "}\n"...)
if _, err = w.Write(buf); err != nil {
return err
}
return w.Flush()
}

View File

@@ -1,92 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"bufio"
"fmt"
"time"
)
// MaxMissingDataPoints is a threshold that allows a node to be healthy with certain number of data points missing.
// Suppose a node does not receive last 5 data points, then healthCheck endpoint will still say a
// node is healthy. Anything more than 5 missing points in metrics of the node will deem the node unhealthy.
const MaxMissingDataPoints int64 = 5
// MaxUnhealthyMetrics is a threshold which allows upto certain number of metrics in a node to be unhealthly.
// Works with MaxMissingDataPoints. Say 5 metrics (including submetrics) do not receive the last
// MaxMissingDataPoints data points, then the node will be deemed healthy. Any more metrics that does
// not receive data for MaxMissingDataPoints data points will deem the node unhealthy.
const MaxUnhealthyMetrics int64 = 5
func (b *buffer) healthCheck() int64 {
// Check if the buffer is empty
if b.data == nil {
return 1
}
bufferEnd := b.start + b.frequency*int64(len(b.data))
t := time.Now().Unix()
// Check if the buffer is too old
if t-bufferEnd > MaxMissingDataPoints*b.frequency {
return 1
}
return 0
}
func (l *Level) healthCheck(m *MemoryStore, count int64) (int64, error) {
l.lock.RLock()
defer l.lock.RUnlock()
for _, mc := range m.Metrics {
if b := l.metrics[mc.offset]; b != nil {
count += b.healthCheck()
}
}
for _, lvl := range l.children {
c, err := lvl.healthCheck(m, 0)
if err != nil {
return 0, err
}
count += c
}
return count, nil
}
func (m *MemoryStore) HealthCheck(w *bufio.Writer, selector []string) error {
lvl := m.root.findLevel(selector)
if lvl == nil {
return fmt.Errorf("[METRICSTORE]> not found: %#v", selector)
}
buf := make([]byte, 0, 25)
// buf = append(buf, "{"...)
var count int64 = 0
unhealthyMetricsCount, err := lvl.healthCheck(m, count)
if err != nil {
return err
}
if unhealthyMetricsCount < MaxUnhealthyMetrics {
buf = append(buf, "Healthy"...)
} else {
buf = append(buf, "Unhealthy"...)
}
// buf = append(buf, "}\n"...)
if _, err = w.Write(buf); err != nil {
return err
}
return w.Flush()
}

View File

@@ -1,192 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"sync"
"unsafe"
"github.com/ClusterCockpit/cc-lib/util"
)
// Could also be called "node" as this forms a node in a tree structure.
// Called Level because "node" might be confusing here.
// Can be both a leaf or a inner node. In this tree structue, inner nodes can
// also hold data (in `metrics`).
type Level struct {
children map[string]*Level
metrics []*buffer
lock sync.RWMutex
}
// Find the correct level for the given selector, creating it if
// it does not exist. Example selector in the context of the
// ClusterCockpit could be: []string{ "emmy", "host123", "cpu0" }.
// This function would probably benefit a lot from `level.children` beeing a `sync.Map`?
func (l *Level) findLevelOrCreate(selector []string, nMetrics int) *Level {
if len(selector) == 0 {
return l
}
// Allow concurrent reads:
l.lock.RLock()
var child *Level
var ok bool
if l.children == nil {
// Children map needs to be created...
l.lock.RUnlock()
} else {
child, ok := l.children[selector[0]]
l.lock.RUnlock()
if ok {
return child.findLevelOrCreate(selector[1:], nMetrics)
}
}
// The level does not exist, take write lock for unqiue access:
l.lock.Lock()
// While this thread waited for the write lock, another thread
// could have created the child node.
if l.children != nil {
child, ok = l.children[selector[0]]
if ok {
l.lock.Unlock()
return child.findLevelOrCreate(selector[1:], nMetrics)
}
}
child = &Level{
metrics: make([]*buffer, nMetrics),
children: nil,
}
if l.children != nil {
l.children[selector[0]] = child
} else {
l.children = map[string]*Level{selector[0]: child}
}
l.lock.Unlock()
return child.findLevelOrCreate(selector[1:], nMetrics)
}
func (l *Level) free(t int64) (int, error) {
l.lock.Lock()
defer l.lock.Unlock()
n := 0
for i, b := range l.metrics {
if b != nil {
delme, m := b.free(t)
n += m
if delme {
if cap(b.data) == BufferCap {
bufferPool.Put(b)
}
l.metrics[i] = nil
}
}
}
for _, l := range l.children {
m, err := l.free(t)
n += m
if err != nil {
return n, err
}
}
return n, nil
}
func (l *Level) sizeInBytes() int64 {
l.lock.RLock()
defer l.lock.RUnlock()
size := int64(0)
for _, b := range l.metrics {
if b != nil {
size += b.count() * int64(unsafe.Sizeof(util.Float(0)))
}
}
for _, child := range l.children {
size += child.sizeInBytes()
}
return size
}
func (l *Level) findLevel(selector []string) *Level {
if len(selector) == 0 {
return l
}
l.lock.RLock()
defer l.lock.RUnlock()
lvl := l.children[selector[0]]
if lvl == nil {
return nil
}
return lvl.findLevel(selector[1:])
}
func (l *Level) findBuffers(selector util.Selector, offset int, f func(b *buffer) error) error {
l.lock.RLock()
defer l.lock.RUnlock()
if len(selector) == 0 {
b := l.metrics[offset]
if b != nil {
return f(b)
}
for _, lvl := range l.children {
err := lvl.findBuffers(nil, offset, f)
if err != nil {
return err
}
}
return nil
}
sel := selector[0]
if len(sel.String) != 0 && l.children != nil {
lvl, ok := l.children[sel.String]
if ok {
err := lvl.findBuffers(selector[1:], offset, f)
if err != nil {
return err
}
}
return nil
}
if sel.Group != nil && l.children != nil {
for _, key := range sel.Group {
lvl, ok := l.children[key]
if ok {
err := lvl.findBuffers(selector[1:], offset, f)
if err != nil {
return err
}
}
}
return nil
}
if sel.Any && l.children != nil {
for _, lvl := range l.children {
if err := lvl.findBuffers(selector[1:], offset, f); err != nil {
return err
}
}
return nil
}
return nil
}

View File

@@ -1,351 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"context"
"fmt"
"sync"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/influxdata/line-protocol/v2/lineprotocol"
"github.com/nats-io/nats.go"
)
// Each connection is handled in it's own goroutine. This is a blocking function.
// func ReceiveRaw(ctx context.Context,
// listener net.Listener,
// handleLine func(*lineprotocol.Decoder, string) error,
// ) error {
// var wg sync.WaitGroup
// wg.Add(1)
// go func() {
// defer wg.Done()
// <-ctx.Done()
// if err := listener.Close(); err != nil {
// log.Printf("listener.Close(): %s", err.Error())
// }
// }()
// for {
// conn, err := listener.Accept()
// if err != nil {
// if errors.Is(err, net.ErrClosed) {
// break
// }
// log.Printf("listener.Accept(): %s", err.Error())
// }
// wg.Add(2)
// go func() {
// defer wg.Done()
// defer conn.Close()
// dec := lineprotocol.NewDecoder(conn)
// connctx, cancel := context.WithCancel(context.Background())
// defer cancel()
// go func() {
// defer wg.Done()
// select {
// case <-connctx.Done():
// conn.Close()
// case <-ctx.Done():
// conn.Close()
// }
// }()
// if err := handleLine(dec, "default"); err != nil {
// if errors.Is(err, net.ErrClosed) {
// return
// }
// log.Printf("%s: %s", conn.RemoteAddr().String(), err.Error())
// errmsg := make([]byte, 128)
// errmsg = append(errmsg, `error: `...)
// errmsg = append(errmsg, err.Error()...)
// errmsg = append(errmsg, '\n')
// conn.Write(errmsg)
// }
// }()
// }
// wg.Wait()
// return nil
// }
// ReceiveNats connects to a nats server and subscribes to "updates". This is a
// blocking function. handleLine will be called for each line recieved via
// nats. Send `true` through the done channel for gracefull termination.
func ReceiveNats(conf *(NatsConfig),
ms *MemoryStore,
workers int,
ctx context.Context,
) error {
var opts []nats.Option
if conf.Username != "" && conf.Password != "" {
opts = append(opts, nats.UserInfo(conf.Username, conf.Password))
}
if conf.Credsfilepath != "" {
opts = append(opts, nats.UserCredentials(conf.Credsfilepath))
}
nc, err := nats.Connect(conf.Address, opts...)
if err != nil {
return err
}
defer nc.Close()
var wg sync.WaitGroup
var subs []*nats.Subscription
msgs := make(chan *nats.Msg, workers*2)
for _, sc := range conf.Subscriptions {
clusterTag := sc.ClusterTag
var sub *nats.Subscription
if workers > 1 {
wg.Add(workers)
for range workers {
go func() {
for m := range msgs {
dec := lineprotocol.NewDecoderWithBytes(m.Data)
if err := DecodeLine(dec, ms, clusterTag); err != nil {
cclog.Printf("error: %s\n", err.Error())
}
}
wg.Done()
}()
}
sub, err = nc.Subscribe(sc.SubscribeTo, func(m *nats.Msg) {
msgs <- m
})
} else {
sub, err = nc.Subscribe(sc.SubscribeTo, func(m *nats.Msg) {
dec := lineprotocol.NewDecoderWithBytes(m.Data)
if err := DecodeLine(dec, ms, clusterTag); err != nil {
cclog.Printf("error: %s\n", err.Error())
}
})
}
if err != nil {
return err
}
cclog.Printf("NATS subscription to '%s' on '%s' established\n", sc.SubscribeTo, conf.Address)
subs = append(subs, sub)
}
<-ctx.Done()
for _, sub := range subs {
err = sub.Unsubscribe()
if err != nil {
cclog.Printf("NATS unsubscribe failed: %s", err.Error())
}
}
close(msgs)
wg.Wait()
nc.Close()
cclog.Print("NATS connection closed")
return nil
}
// Place `prefix` in front of `buf` but if possible,
// do that inplace in `buf`.
func reorder(buf, prefix []byte) []byte {
n := len(prefix)
m := len(buf)
if cap(buf) < m+n {
return append(prefix[:n:n], buf...)
} else {
buf = buf[:n+m]
for i := m - 1; i >= 0; i-- {
buf[i+n] = buf[i]
}
for i := range n {
buf[i] = prefix[i]
}
return buf
}
}
// Decode lines using dec and make write calls to the MemoryStore.
// If a line is missing its cluster tag, use clusterDefault as default.
func DecodeLine(dec *lineprotocol.Decoder,
ms *MemoryStore,
clusterDefault string,
) error {
// Reduce allocations in loop:
t := time.Now()
metric, metricBuf := Metric{}, make([]byte, 0, 16)
selector := make([]string, 0, 4)
typeBuf, subTypeBuf := make([]byte, 0, 16), make([]byte, 0)
// Optimize for the case where all lines in a "batch" are about the same
// cluster and host. By using `WriteToLevel` (level = host), we do not need
// to take the root- and cluster-level lock as often.
var lvl *Level = nil
prevCluster, prevHost := "", ""
var ok bool
for dec.Next() {
rawmeasurement, err := dec.Measurement()
if err != nil {
return err
}
// Needs to be copied because another call to dec.* would
// invalidate the returned slice.
metricBuf = append(metricBuf[:0], rawmeasurement...)
// The go compiler optimizes map[string(byteslice)] lookups:
metric.MetricConfig, ok = ms.Metrics[string(rawmeasurement)]
if !ok {
continue
}
typeBuf, subTypeBuf := typeBuf[:0], subTypeBuf[:0]
cluster, host := clusterDefault, ""
for {
key, val, err := dec.NextTag()
if err != nil {
return err
}
if key == nil {
break
}
// The go compiler optimizes string([]byte{...}) == "...":
switch string(key) {
case "cluster":
if string(val) == prevCluster {
cluster = prevCluster
} else {
cluster = string(val)
lvl = nil
}
case "hostname", "host":
if string(val) == prevHost {
host = prevHost
} else {
host = string(val)
lvl = nil
}
case "type":
if string(val) == "node" {
break
}
// We cannot be sure that the "type" tag comes before the "type-id" tag:
if len(typeBuf) == 0 {
typeBuf = append(typeBuf, val...)
} else {
typeBuf = reorder(typeBuf, val)
}
case "type-id":
typeBuf = append(typeBuf, val...)
case "subtype":
// We cannot be sure that the "subtype" tag comes before the "stype-id" tag:
if len(subTypeBuf) == 0 {
subTypeBuf = append(subTypeBuf, val...)
} else {
subTypeBuf = reorder(subTypeBuf, val)
// subTypeBuf = reorder(typeBuf, val)
}
case "stype-id":
subTypeBuf = append(subTypeBuf, val...)
default:
// Ignore unkown tags (cc-metric-collector might send us a unit for example that we do not need)
// return fmt.Errorf("unkown tag: '%s' (value: '%s')", string(key), string(val))
}
}
// If the cluster or host changed, the lvl was set to nil
if lvl == nil {
selector = selector[:2]
selector[0], selector[1] = cluster, host
lvl = ms.GetLevel(selector)
prevCluster, prevHost = cluster, host
}
// subtypes:
selector = selector[:0]
if len(typeBuf) > 0 {
selector = append(selector, string(typeBuf)) // <- Allocation :(
if len(subTypeBuf) > 0 {
selector = append(selector, string(subTypeBuf))
}
}
for {
key, val, err := dec.NextField()
if err != nil {
return err
}
if key == nil {
break
}
if string(key) != "value" {
return fmt.Errorf("host %s: unknown field: '%s' (value: %#v)", host, string(key), val)
}
if val.Kind() == lineprotocol.Float {
metric.Value = schema.Float(val.FloatV())
} else if val.Kind() == lineprotocol.Int {
metric.Value = schema.Float(val.IntV())
} else if val.Kind() == lineprotocol.Uint {
metric.Value = schema.Float(val.UintV())
} else {
return fmt.Errorf("host %s: unsupported value type in message: %s", host, val.Kind().String())
}
}
if t, err = dec.Time(lineprotocol.Second, t); err != nil {
t = time.Now()
if t, err = dec.Time(lineprotocol.Millisecond, t); err != nil {
t = time.Now()
if t, err = dec.Time(lineprotocol.Microsecond, t); err != nil {
t = time.Now()
if t, err = dec.Time(lineprotocol.Nanosecond, t); err != nil {
return fmt.Errorf("host %s: timestamp : %#v with error : %#v", host, t, err.Error())
}
}
}
}
if err != nil {
return fmt.Errorf("host %s: timestamp : %#v with error : %#v", host, t, err.Error())
}
time := t.Unix()
if Keys.Checkpoints.FileFormat != "json" {
LineProtocolMessages <- &AvroStruct{
MetricName: string(metricBuf),
Cluster: cluster,
Node: host,
Selector: append([]string{}, selector...),
Value: metric.Value,
Timestamp: time,
}
}
if err := ms.WriteToLevel(lvl, selector, time, []Metric{metric}); err != nil {
return err
}
}
return nil
}

View File

@@ -1,486 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"bytes"
"context"
"encoding/json"
"errors"
"os"
"os/signal"
"runtime"
"sync"
"syscall"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/resampler"
"github.com/ClusterCockpit/cc-lib/runtimeEnv"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-lib/util"
)
var (
singleton sync.Once
msInstance *MemoryStore
)
var NumWorkers int = 4
func init() {
maxWorkers := 10
NumWorkers = min(runtime.NumCPU()/2+1, maxWorkers)
}
type Metric struct {
Name string
Value schema.Float
MetricConfig MetricConfig
}
type MemoryStore struct {
Metrics map[string]MetricConfig
root Level
}
func Init(rawConfig json.RawMessage, wg *sync.WaitGroup) {
startupTime := time.Now()
if rawConfig != nil {
config.Validate(configSchema, rawConfig)
dec := json.NewDecoder(bytes.NewReader(rawConfig))
// dec.DisallowUnknownFields()
if err := dec.Decode(&Keys); err != nil {
cclog.Abortf("[METRICSTORE]> Metric Store Config Init: Could not decode config file '%s'.\nError: %s\n", rawConfig, err.Error())
}
}
for _, c := range archive.Clusters {
for _, mc := range c.MetricConfig {
agg, err := AssignAggregationStratergy(mc.Aggregation)
if err != nil {
cclog.Warnf("Could not find aggregation stratergy for metric config '%s': %s", mc.Name, err.Error())
}
AddMetric(mc.Name, MetricConfig{
Frequency: int64(mc.Timestep),
Aggregation: agg,
})
}
for _, sc := range c.SubClusters {
for _, mc := range sc.MetricConfig {
agg, err := AssignAggregationStratergy(mc.Aggregation)
if err != nil {
cclog.Warnf("Could not find aggregation stratergy for metric config '%s': %s", mc.Name, err.Error())
}
AddMetric(mc.Name, MetricConfig{
Frequency: int64(mc.Timestep),
Aggregation: agg,
})
}
}
}
// Pass the config.MetricStoreKeys
InitMetrics(Metrics)
ms := GetMemoryStore()
d, err := time.ParseDuration(Keys.Checkpoints.Restore)
if err != nil {
cclog.Fatal(err)
}
restoreFrom := startupTime.Add(-d)
cclog.Infof("[METRICSTORE]> Loading checkpoints newer than %s\n", restoreFrom.Format(time.RFC3339))
files, err := ms.FromCheckpointFiles(Keys.Checkpoints.RootDir, restoreFrom.Unix())
loadedData := ms.SizeInBytes() / 1024 / 1024 // In MB
if err != nil {
cclog.Fatalf("[METRICSTORE]> Loading checkpoints failed: %s\n", err.Error())
} else {
cclog.Infof("[METRICSTORE]> Checkpoints loaded (%d files, %d MB, that took %fs)\n", files, loadedData, time.Since(startupTime).Seconds())
}
// Try to use less memory by forcing a GC run here and then
// lowering the target percentage. The default of 100 means
// that only once the ratio of new allocations execeds the
// previously active heap, a GC is triggered.
// Forcing a GC here will set the "previously active heap"
// to a minumum.
runtime.GC()
ctx, shutdown := context.WithCancel(context.Background())
wg.Add(4)
Retention(wg, ctx)
Checkpointing(wg, ctx)
Archiving(wg, ctx)
DataStaging(wg, ctx)
wg.Add(1)
sigs := make(chan os.Signal, 1)
signal.Notify(sigs, syscall.SIGINT, syscall.SIGTERM)
go func() {
defer wg.Done()
<-sigs
runtimeEnv.SystemdNotifiy(false, "[METRICSTORE]> Shutting down ...")
shutdown()
}()
if Keys.Nats != nil {
for _, natsConf := range Keys.Nats {
// TODO: When multiple nats configs share a URL, do a single connect.
wg.Add(1)
nc := natsConf
go func() {
// err := ReceiveNats(conf.Nats, decodeLine, runtime.NumCPU()-1, ctx)
err := ReceiveNats(nc, ms, 1, ctx)
if err != nil {
cclog.Fatal(err)
}
wg.Done()
}()
}
}
}
// InitMetrics creates a new, initialized instance of a MemoryStore.
// Will panic if values in the metric configurations are invalid.
func InitMetrics(metrics map[string]MetricConfig) {
singleton.Do(func() {
offset := 0
for key, cfg := range metrics {
if cfg.Frequency == 0 {
panic("[METRICSTORE]> invalid frequency")
}
metrics[key] = MetricConfig{
Frequency: cfg.Frequency,
Aggregation: cfg.Aggregation,
offset: offset,
}
offset += 1
}
msInstance = &MemoryStore{
root: Level{
metrics: make([]*buffer, len(metrics)),
children: make(map[string]*Level),
},
Metrics: metrics,
}
})
}
func GetMemoryStore() *MemoryStore {
if msInstance == nil {
cclog.Fatalf("[METRICSTORE]> MemoryStore not initialized!")
}
return msInstance
}
func Shutdown() {
cclog.Infof("[METRICSTORE]> Writing to '%s'...\n", Keys.Checkpoints.RootDir)
var files int
var err error
ms := GetMemoryStore()
if Keys.Checkpoints.FileFormat == "json" {
files, err = ms.ToCheckpoint(Keys.Checkpoints.RootDir, lastCheckpoint.Unix(), time.Now().Unix())
} else {
files, err = GetAvroStore().ToCheckpoint(Keys.Checkpoints.RootDir, true)
close(LineProtocolMessages)
}
if err != nil {
cclog.Errorf("[METRICSTORE]> Writing checkpoint failed: %s\n", err.Error())
}
cclog.Infof("[METRICSTORE]> Done! (%d files written)\n", files)
// ms.PrintHeirarchy()
}
// func (m *MemoryStore) PrintHeirarchy() {
// m.root.lock.Lock()
// defer m.root.lock.Unlock()
// fmt.Printf("Root : \n")
// for lvl1, sel1 := range m.root.children {
// fmt.Printf("\t%s\n", lvl1)
// for lvl2, sel2 := range sel1.children {
// fmt.Printf("\t\t%s\n", lvl2)
// if lvl1 == "fritz" && lvl2 == "f0201" {
// for name, met := range m.Metrics {
// mt := sel2.metrics[met.Offset]
// fmt.Printf("\t\t\t\t%s\n", name)
// fmt.Printf("\t\t\t\t")
// for mt != nil {
// // if name == "cpu_load" {
// fmt.Printf("%d(%d) -> %#v", mt.start, len(mt.data), mt.data)
// // }
// mt = mt.prev
// }
// fmt.Printf("\n")
// }
// }
// for lvl3, sel3 := range sel2.children {
// if lvl1 == "fritz" && lvl2 == "f0201" && lvl3 == "hwthread70" {
// fmt.Printf("\t\t\t\t\t%s\n", lvl3)
// for name, met := range m.Metrics {
// mt := sel3.metrics[met.Offset]
// fmt.Printf("\t\t\t\t\t\t%s\n", name)
// fmt.Printf("\t\t\t\t\t\t")
// for mt != nil {
// // if name == "clock" {
// fmt.Printf("%d(%d) -> %#v", mt.start, len(mt.data), mt.data)
// mt = mt.prev
// }
// fmt.Printf("\n")
// }
// // for i, _ := range sel3.metrics {
// // fmt.Printf("\t\t\t\t\t%s\n", getName(configmetrics, i))
// // }
// }
// }
// }
// }
// }
func getName(m *MemoryStore, i int) string {
for key, val := range m.Metrics {
if val.offset == i {
return key
}
}
return ""
}
func Retention(wg *sync.WaitGroup, ctx context.Context) {
ms := GetMemoryStore()
go func() {
defer wg.Done()
d, err := time.ParseDuration(Keys.RetentionInMemory)
if err != nil {
cclog.Fatal(err)
}
if d <= 0 {
return
}
ticks := func() <-chan time.Time {
d := d / 2
if d <= 0 {
return nil
}
return time.NewTicker(d).C
}()
for {
select {
case <-ctx.Done():
return
case <-ticks:
t := time.Now().Add(-d)
cclog.Infof("[METRICSTORE]> start freeing buffers (older than %s)...\n", t.Format(time.RFC3339))
freed, err := ms.Free(nil, t.Unix())
if err != nil {
cclog.Errorf("[METRICSTORE]> freeing up buffers failed: %s\n", err.Error())
} else {
cclog.Infof("[METRICSTORE]> done: %d buffers freed\n", freed)
}
}
}
}()
}
// Write all values in `metrics` to the level specified by `selector` for time `ts`.
// Look at `findLevelOrCreate` for how selectors work.
func (m *MemoryStore) Write(selector []string, ts int64, metrics []Metric) error {
var ok bool
for i, metric := range metrics {
if metric.MetricConfig.Frequency == 0 {
metric.MetricConfig, ok = m.Metrics[metric.Name]
if !ok {
metric.MetricConfig.Frequency = 0
}
metrics[i] = metric
}
}
return m.WriteToLevel(&m.root, selector, ts, metrics)
}
func (m *MemoryStore) GetLevel(selector []string) *Level {
return m.root.findLevelOrCreate(selector, len(m.Metrics))
}
// WriteToLevel assumes that `minfo` in `metrics` is filled in
func (m *MemoryStore) WriteToLevel(l *Level, selector []string, ts int64, metrics []Metric) error {
l = l.findLevelOrCreate(selector, len(m.Metrics))
l.lock.Lock()
defer l.lock.Unlock()
for _, metric := range metrics {
if metric.MetricConfig.Frequency == 0 {
continue
}
b := l.metrics[metric.MetricConfig.offset]
if b == nil {
// First write to this metric and level
b = newBuffer(ts, metric.MetricConfig.Frequency)
l.metrics[metric.MetricConfig.offset] = b
}
nb, err := b.write(ts, metric.Value)
if err != nil {
return err
}
// Last write created a new buffer...
if b != nb {
l.metrics[metric.MetricConfig.offset] = nb
}
}
return nil
}
// Read returns all values for metric `metric` from `from` to `to` for the selected level(s).
// If the level does not hold the metric itself, the data will be aggregated recursively from the children.
// The second and third return value are the actual from/to for the data. Those can be different from
// the range asked for if no data was available.
func (m *MemoryStore) Read(selector util.Selector, metric string, from, to, resolution int64) ([]schema.Float, int64, int64, int64, error) {
if from > to {
return nil, 0, 0, 0, errors.New("[METRICSTORE]> invalid time range")
}
minfo, ok := m.Metrics[metric]
if !ok {
return nil, 0, 0, 0, errors.New("[METRICSTORE]> unkown metric: " + metric)
}
n, data := 0, make([]schema.Float, (to-from)/minfo.Frequency+1)
err := m.root.findBuffers(selector, minfo.offset, func(b *buffer) error {
cdata, cfrom, cto, err := b.read(from, to, data)
if err != nil {
return err
}
if n == 0 {
from, to = cfrom, cto
} else if from != cfrom || to != cto || len(data) != len(cdata) {
missingfront, missingback := int((from-cfrom)/minfo.Frequency), int((to-cto)/minfo.Frequency)
if missingfront != 0 {
return ErrDataDoesNotAlign
}
newlen := len(cdata) - missingback
if newlen < 1 {
return ErrDataDoesNotAlign
}
cdata = cdata[0:newlen]
if len(cdata) != len(data) {
return ErrDataDoesNotAlign
}
from, to = cfrom, cto
}
data = cdata
n += 1
return nil
})
if err != nil {
return nil, 0, 0, 0, err
} else if n == 0 {
return nil, 0, 0, 0, errors.New("[METRICSTORE]> metric or host not found")
} else if n > 1 {
if minfo.Aggregation == AvgAggregation {
normalize := 1. / schema.Float(n)
for i := 0; i < len(data); i++ {
data[i] *= normalize
}
} else if minfo.Aggregation != SumAggregation {
return nil, 0, 0, 0, errors.New("[METRICSTORE]> invalid aggregation")
}
}
data, resolution, err = resampler.LargestTriangleThreeBucket(data, minfo.Frequency, resolution)
if err != nil {
return nil, 0, 0, 0, err
}
return data, from, to, resolution, nil
}
// Free releases all buffers for the selected level and all its children that
// contain only values older than `t`.
func (m *MemoryStore) Free(selector []string, t int64) (int, error) {
return m.GetLevel(selector).free(t)
}
func (m *MemoryStore) FreeAll() error {
for k := range m.root.children {
delete(m.root.children, k)
}
return nil
}
func (m *MemoryStore) SizeInBytes() int64 {
return m.root.sizeInBytes()
}
// ListChildren , given a selector, returns a list of all children of the level
// selected.
func (m *MemoryStore) ListChildren(selector []string) []string {
lvl := &m.root
for lvl != nil && len(selector) != 0 {
lvl.lock.RLock()
next := lvl.children[selector[0]]
lvl.lock.RUnlock()
lvl = next
selector = selector[1:]
}
if lvl == nil {
return nil
}
lvl.lock.RLock()
defer lvl.lock.RUnlock()
children := make([]string, 0, len(lvl.children))
for child := range lvl.children {
children = append(children, child)
}
return children
}

View File

@@ -1,124 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package memorystore
import (
"errors"
"math"
"github.com/ClusterCockpit/cc-lib/util"
)
type Stats struct {
Samples int
Avg util.Float
Min util.Float
Max util.Float
}
func (b *buffer) stats(from, to int64) (Stats, int64, int64, error) {
if from < b.start {
if b.prev != nil {
return b.prev.stats(from, to)
}
from = b.start
}
// TODO: Check if b.closed and if so and the full buffer is queried,
// use b.statistics instead of iterating over the buffer.
samples := 0
sum, min, max := 0.0, math.MaxFloat32, -math.MaxFloat32
var t int64
for t = from; t < to; t += b.frequency {
idx := int((t - b.start) / b.frequency)
if idx >= cap(b.data) {
b = b.next
if b == nil {
break
}
idx = 0
}
if t < b.start || idx >= len(b.data) {
continue
}
xf := float64(b.data[idx])
if math.IsNaN(xf) {
continue
}
samples += 1
sum += xf
min = math.Min(min, xf)
max = math.Max(max, xf)
}
return Stats{
Samples: samples,
Avg: util.Float(sum) / util.Float(samples),
Min: util.Float(min),
Max: util.Float(max),
}, from, t, nil
}
// Returns statistics for the requested metric on the selected node/level.
// Data is aggregated to the selected level the same way as in `MemoryStore.Read`.
// If `Stats.Samples` is zero, the statistics should not be considered as valid.
func (m *MemoryStore) Stats(selector util.Selector, metric string, from, to int64) (*Stats, int64, int64, error) {
if from > to {
return nil, 0, 0, errors.New("invalid time range")
}
minfo, ok := m.Metrics[metric]
if !ok {
return nil, 0, 0, errors.New("unkown metric: " + metric)
}
n, samples := 0, 0
avg, min, max := util.Float(0), math.MaxFloat32, -math.MaxFloat32
err := m.root.findBuffers(selector, minfo.offset, func(b *buffer) error {
stats, cfrom, cto, err := b.stats(from, to)
if err != nil {
return err
}
if n == 0 {
from, to = cfrom, cto
} else if from != cfrom || to != cto {
return ErrDataDoesNotAlign
}
samples += stats.Samples
avg += stats.Avg
min = math.Min(min, float64(stats.Min))
max = math.Max(max, float64(stats.Max))
n += 1
return nil
})
if err != nil {
return nil, 0, 0, err
}
if n == 0 {
return nil, 0, 0, ErrNoData
}
if minfo.Aggregation == AvgAggregation {
avg /= util.Float(n)
} else if n > 1 && minfo.Aggregation != SumAggregation {
return nil, 0, 0, errors.New("invalid aggregation")
}
return &Stats{
Samples: samples,
Avg: avg,
Min: util.Float(min),
Max: util.Float(max),
}, from, to, nil
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package metricDataDispatcher
@@ -7,17 +7,16 @@ package metricDataDispatcher
import (
"context"
"fmt"
"math"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/internal/metricdata"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/lrucache"
"github.com/ClusterCockpit/cc-lib/resampler"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/lrucache"
"github.com/ClusterCockpit/cc-backend/pkg/resampler"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
var cache *lrucache.Cache = lrucache.New(128 * 1024 * 1024)
@@ -41,7 +40,7 @@ func LoadData(job *schema.Job,
ctx context.Context,
resolution int,
) (schema.JobData, error) {
data := cache.Get(cacheKey(job, metrics, scopes, resolution), func() (_ any, ttl time.Duration, size int) {
data := cache.Get(cacheKey(job, metrics, scopes, resolution), func() (_ interface{}, ttl time.Duration, size int) {
var jd schema.JobData
var err error
@@ -68,10 +67,10 @@ func LoadData(job *schema.Job,
jd, err = repo.LoadData(job, metrics, scopes, ctx, resolution)
if err != nil {
if len(jd) != 0 {
cclog.Warnf("partial error: %s", err.Error())
log.Warnf("partial error: %s", err.Error())
// return err, 0, 0 // Reactivating will block archiving on one partial error
} else {
cclog.Error("Error while loading job data from metric repository")
log.Error("Error while loading job data from metric repository")
return err, 0, 0
}
}
@@ -80,25 +79,25 @@ func LoadData(job *schema.Job,
var jd_temp schema.JobData
jd_temp, err = archive.GetHandle().LoadJobData(job)
if err != nil {
cclog.Error("Error while loading job data from archive")
log.Error("Error while loading job data from archive")
return err, 0, 0
}
// Deep copy the cached archive hashmap
//Deep copy the cached archive hashmap
jd = metricdata.DeepCopy(jd_temp)
// Resampling for archived data.
// Pass the resolution from frontend here.
//Resampling for archived data.
//Pass the resolution from frontend here.
for _, v := range jd {
for _, v_ := range v {
timestep := int64(0)
timestep := 0
for i := 0; i < len(v_.Series); i += 1 {
v_.Series[i].Data, timestep, err = resampler.LargestTriangleThreeBucket(v_.Series[i].Data, int64(v_.Timestep), int64(resolution))
v_.Series[i].Data, timestep, err = resampler.LargestTriangleThreeBucket(v_.Series[i].Data, v_.Timestep, resolution)
if err != nil {
return err, 0, 0
}
}
v_.Timestep = int(timestep)
v_.Timestep = timestep
}
}
@@ -171,14 +170,11 @@ func LoadData(job *schema.Job,
jd.AddNodeScope("mem_bw")
}
// Round Resulting Stat Values
jd.RoundMetricStats()
return jd, ttl, size
})
if err, ok := data.(error); ok {
cclog.Error("Error in returned dataset")
log.Error("Error in returned dataset")
return nil, err
}
@@ -203,7 +199,7 @@ func LoadAverages(
stats, err := repo.LoadStats(job, metrics, ctx) // #166 how to handle stats for acc normalizazion?
if err != nil {
cclog.Errorf("Error while loading statistics for job %v (User %v, Project %v)", job.JobID, job.User, job.Project)
log.Errorf("Error while loading statistics for job %v (User %v, Project %v)", job.JobID, job.User, job.Project)
return err
}
@@ -224,77 +220,6 @@ func LoadAverages(
return nil
}
// Used for statsTable in frontend: Return scoped statistics by metric.
func LoadScopedJobStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
) (schema.ScopedJobStats, error) {
if job.State != schema.JobStateRunning && !config.Keys.DisableArchive {
return archive.LoadScopedStatsFromArchive(job, metrics, scopes)
}
repo, err := metricdata.GetMetricDataRepo(job.Cluster)
if err != nil {
return nil, fmt.Errorf("job %d: no metric data repository configured for '%s'", job.JobID, job.Cluster)
}
scopedStats, err := repo.LoadScopedStats(job, metrics, scopes, ctx)
if err != nil {
cclog.Errorf("error while loading scoped statistics for job %d (User %s, Project %s)", job.JobID, job.User, job.Project)
return nil, err
}
return scopedStats, nil
}
// Used for polar plots in frontend: Aggregates statistics for all nodes to single values for job per metric.
func LoadJobStats(
job *schema.Job,
metrics []string,
ctx context.Context,
) (map[string]schema.MetricStatistics, error) {
if job.State != schema.JobStateRunning && !config.Keys.DisableArchive {
return archive.LoadStatsFromArchive(job, metrics)
}
data := make(map[string]schema.MetricStatistics, len(metrics))
repo, err := metricdata.GetMetricDataRepo(job.Cluster)
if err != nil {
return data, fmt.Errorf("job %d: no metric data repository configured for '%s'", job.JobID, job.Cluster)
}
stats, err := repo.LoadStats(job, metrics, ctx)
if err != nil {
cclog.Errorf("error while loading statistics for job %d (User %s, Project %s)", job.JobID, job.User, job.Project)
return data, err
}
for _, m := range metrics {
sum, avg, min, max := 0.0, 0.0, 0.0, 0.0
nodes, ok := stats[m]
if !ok {
data[m] = schema.MetricStatistics{Min: min, Avg: avg, Max: max}
continue
}
for _, node := range nodes {
sum += node.Avg
min = math.Min(min, node.Min)
max = math.Max(max, node.Max)
}
data[m] = schema.MetricStatistics{
Avg: (math.Round((sum/float64(job.NumNodes))*100) / 100),
Min: (math.Round(min*100) / 100),
Max: (math.Round(max*100) / 100),
}
}
return data, nil
}
// Used for the classic node/system view. Returns a map of nodes to a map of metrics.
func LoadNodeData(
cluster string,
@@ -317,9 +242,9 @@ func LoadNodeData(
data, err := repo.LoadNodeData(cluster, metrics, nodes, scopes, from, to, ctx)
if err != nil {
if len(data) != 0 {
cclog.Warnf("partial error: %s", err.Error())
log.Warnf("partial error: %s", err.Error())
} else {
cclog.Error("Error while loading node data from metric repository")
log.Error("Error while loading node data from metric repository")
return nil, err
}
}
@@ -354,9 +279,9 @@ func LoadNodeListData(
data, totalNodes, hasNextPage, err := repo.LoadNodeListData(cluster, subCluster, nodeFilter, metrics, scopes, resolution, from, to, page, ctx)
if err != nil {
if len(data) != 0 {
cclog.Warnf("partial error: %s", err.Error())
log.Warnf("partial error: %s", err.Error())
} else {
cclog.Error("Error while loading node data from metric repository")
log.Error("Error while loading node data from metric repository")
return nil, totalNodes, hasNextPage, err
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -12,13 +12,14 @@ import (
"fmt"
"net/http"
"sort"
"strconv"
"strings"
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
type CCMetricStoreConfig struct {
@@ -81,7 +82,7 @@ type ApiMetricData struct {
func (ccms *CCMetricStore) Init(rawConfig json.RawMessage) error {
var config CCMetricStoreConfig
if err := json.Unmarshal(rawConfig, &config); err != nil {
cclog.Warn("Error while unmarshaling raw json config")
log.Warn("Error while unmarshaling raw json config")
return err
}
@@ -128,13 +129,13 @@ func (ccms *CCMetricStore) doRequest(
) (*ApiQueryResponse, error) {
buf := &bytes.Buffer{}
if err := json.NewEncoder(buf).Encode(body); err != nil {
cclog.Errorf("Error while encoding request body: %s", err.Error())
log.Warn("Error while encoding request body")
return nil, err
}
req, err := http.NewRequestWithContext(ctx, http.MethodGet, ccms.queryEndpoint, buf)
if err != nil {
cclog.Errorf("Error while building request body: %s", err.Error())
log.Warn("Error while building request body")
return nil, err
}
if ccms.jwt != "" {
@@ -150,7 +151,7 @@ func (ccms *CCMetricStore) doRequest(
res, err := ccms.client.Do(req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
log.Error("Error while performing request")
return nil, err
}
@@ -160,7 +161,7 @@ func (ccms *CCMetricStore) doRequest(
var resBody ApiQueryResponse
if err := json.NewDecoder(bufio.NewReader(res.Body)).Decode(&resBody); err != nil {
cclog.Errorf("Error while decoding result body: %s", err.Error())
log.Warn("Error while decoding result body")
return nil, err
}
@@ -176,14 +177,14 @@ func (ccms *CCMetricStore) LoadData(
) (schema.JobData, error) {
queries, assignedScope, err := ccms.buildQueries(job, metrics, scopes, resolution)
if err != nil {
cclog.Errorf("Error while building queries for jobId %d, Metrics %v, Scopes %v: %s", job.JobID, metrics, scopes, err.Error())
log.Warn("Error while building queries")
return nil, err
}
req := ApiQueryRequest{
Cluster: job.Cluster,
From: job.StartTime,
To: job.StartTime + int64(job.Duration),
From: job.StartTime.Unix(),
To: job.StartTime.Add(time.Duration(job.Duration) * time.Second).Unix(),
Queries: queries,
WithStats: true,
WithData: true,
@@ -191,7 +192,7 @@ func (ccms *CCMetricStore) LoadData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
log.Error("Error while performing request")
return nil, err
}
@@ -269,6 +270,14 @@ func (ccms *CCMetricStore) LoadData(
return jobData, nil
}
var (
hwthreadString = string(schema.MetricScopeHWThread)
coreString = string(schema.MetricScopeCore)
memoryDomainString = string(schema.MetricScopeMemoryDomain)
socketString = string(schema.MetricScopeSocket)
acceleratorString = string(schema.MetricScopeAccelerator)
)
func (ccms *CCMetricStore) buildQueries(
job *schema.Job,
metrics []string,
@@ -289,24 +298,10 @@ func (ccms *CCMetricStore) buildQueries(
mc := archive.GetMetricConfig(job.Cluster, metric)
if mc == nil {
// return nil, fmt.Errorf("METRICDATA/CCMS > metric '%s' is not specified for cluster '%s'", metric, job.Cluster)
cclog.Infof("metric '%s' is not specified for cluster '%s'", metric, job.Cluster)
log.Infof("metric '%s' is not specified for cluster '%s'", metric, job.Cluster)
continue
}
// Skip if metric is removed for subcluster
if len(mc.SubClusters) != 0 {
isRemoved := false
for _, scConfig := range mc.SubClusters {
if scConfig.Name == job.SubCluster && scConfig.Remove {
isRemoved = true
break
}
}
if isRemoved {
continue
}
}
// Avoid duplicates...
handledScopes := make([]schema.MetricScope, 0, 3)
@@ -445,23 +440,6 @@ func (ccms *CCMetricStore) buildQueries(
continue
}
// Core -> Socket
if nativeScope == schema.MetricScopeCore && scope == schema.MetricScopeSocket {
sockets, _ := topology.GetSocketsFromCores(hwthreads)
for _, socket := range sockets {
queries = append(queries, ApiQuery{
Metric: remoteName,
Hostname: host.Hostname,
Aggregate: true,
Type: &coreString,
TypeIds: intToStringSlice(topology.Socket[socket]),
Resolution: resolution,
})
assignedScope = append(assignedScope, scope)
}
continue
}
// Core -> Node
if nativeScope == schema.MetricScopeCore && scope == schema.MetricScopeNode {
cores, _ := topology.GetCoresFromHWThreads(hwthreads)
@@ -562,16 +540,23 @@ func (ccms *CCMetricStore) LoadStats(
ctx context.Context,
) (map[string]map[string]schema.MetricStatistics, error) {
// metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
// resolution := 9000
// for _, mc := range metricConfigs {
// resolution = min(resolution, mc.Timestep)
// }
queries, _, err := ccms.buildQueries(job, metrics, []schema.MetricScope{schema.MetricScopeNode}, 0) // #166 Add scope shere for analysis view accelerator normalization?
if err != nil {
cclog.Errorf("Error while building queries for jobId %d, Metrics %v: %s", job.JobID, metrics, err.Error())
log.Warn("Error while building query")
return nil, err
}
req := ApiQueryRequest{
Cluster: job.Cluster,
From: job.StartTime,
To: job.StartTime + int64(job.Duration),
From: job.StartTime.Unix(),
To: job.StartTime.Add(time.Duration(job.Duration) * time.Second).Unix(),
Queries: queries,
WithStats: true,
WithData: false,
@@ -579,7 +564,7 @@ func (ccms *CCMetricStore) LoadStats(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
log.Error("Error while performing request")
return nil, err
}
@@ -589,8 +574,9 @@ func (ccms *CCMetricStore) LoadStats(
metric := ccms.toLocalName(query.Metric)
data := res[0]
if data.Error != nil {
cclog.Errorf("fetching %s for node %s failed: %s", metric, query.Hostname, *data.Error)
log.Infof("fetching %s for node %s failed: %s", metric, query.Hostname, *data.Error)
continue
// return nil, fmt.Errorf("METRICDATA/CCMS > fetching %s for node %s failed: %s", metric, query.Hostname, *data.Error)
}
metricdata, ok := stats[metric]
@@ -600,8 +586,9 @@ func (ccms *CCMetricStore) LoadStats(
}
if data.Avg.IsNaN() || data.Min.IsNaN() || data.Max.IsNaN() {
cclog.Warnf("fetching %s for node %s failed: one of avg/min/max is NaN", metric, query.Hostname)
log.Infof("fetching %s for node %s failed: one of avg/min/max is NaN", metric, query.Hostname)
continue
// return nil, fmt.Errorf("METRICDATA/CCMS > fetching %s for node %s failed: %s", metric, query.Hostname, "avg/min/max is NaN")
}
metricdata[query.Hostname] = schema.MetricStatistics{
@@ -614,98 +601,7 @@ func (ccms *CCMetricStore) LoadStats(
return stats, nil
}
// Used for Job-View Statistics Table
func (ccms *CCMetricStore) LoadScopedStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
) (schema.ScopedJobStats, error) {
queries, assignedScope, err := ccms.buildQueries(job, metrics, scopes, 0)
if err != nil {
cclog.Errorf("Error while building queries for jobId %d, Metrics %v, Scopes %v: %s", job.JobID, metrics, scopes, err.Error())
return nil, err
}
req := ApiQueryRequest{
Cluster: job.Cluster,
From: job.StartTime,
To: job.StartTime + int64(job.Duration),
Queries: queries,
WithStats: true,
WithData: false,
}
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
return nil, err
}
var errors []string
scopedJobStats := make(schema.ScopedJobStats)
for i, row := range resBody.Results {
query := req.Queries[i]
metric := ccms.toLocalName(query.Metric)
scope := assignedScope[i]
if _, ok := scopedJobStats[metric]; !ok {
scopedJobStats[metric] = make(map[schema.MetricScope][]*schema.ScopedStats)
}
if _, ok := scopedJobStats[metric][scope]; !ok {
scopedJobStats[metric][scope] = make([]*schema.ScopedStats, 0)
}
for ndx, res := range row {
if res.Error != nil {
/* Build list for "partial errors", if any */
errors = append(errors, fmt.Sprintf("failed to fetch '%s' from host '%s': %s", query.Metric, query.Hostname, *res.Error))
continue
}
id := (*string)(nil)
if query.Type != nil {
id = new(string)
*id = query.TypeIds[ndx]
}
if res.Avg.IsNaN() || res.Min.IsNaN() || res.Max.IsNaN() {
// "schema.Float()" because regular float64 can not be JSONed when NaN.
res.Avg = schema.Float(0)
res.Min = schema.Float(0)
res.Max = schema.Float(0)
}
scopedJobStats[metric][scope] = append(scopedJobStats[metric][scope], &schema.ScopedStats{
Hostname: query.Hostname,
Id: id,
Data: &schema.MetricStatistics{
Avg: float64(res.Avg),
Min: float64(res.Min),
Max: float64(res.Max),
},
})
}
// So that one can later check len(scopedJobStats[metric][scope]): Remove from map if empty
if len(scopedJobStats[metric][scope]) == 0 {
delete(scopedJobStats[metric], scope)
if len(scopedJobStats[metric]) == 0 {
delete(scopedJobStats, metric)
}
}
}
if len(errors) != 0 {
/* Returns list for "partial errors" */
return scopedJobStats, fmt.Errorf("METRICDATA/CCMS > Errors: %s", strings.Join(errors, ", "))
}
return scopedJobStats, nil
}
// Used for Systems-View Node-Overview
// TODO: Support sub-node-scope metrics! For this, the partition of a node needs to be known!
func (ccms *CCMetricStore) LoadNodeData(
cluster string,
metrics, nodes []string,
@@ -731,7 +627,7 @@ func (ccms *CCMetricStore) LoadNodeData(
req.Queries = append(req.Queries, ApiQuery{
Hostname: node,
Metric: ccms.toRemoteName(metric),
Resolution: 0, // Default for Node Queries: Will return metric $Timestep Resolution
Resolution: 60, // Default for Node Queries
})
}
}
@@ -739,7 +635,7 @@ func (ccms *CCMetricStore) LoadNodeData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
log.Error(fmt.Sprintf("Error while performing request %#v\n", err))
return nil, err
}
@@ -797,7 +693,6 @@ func (ccms *CCMetricStore) LoadNodeData(
return data, nil
}
// Used for Systems-View Node-List
func (ccms *CCMetricStore) LoadNodeListData(
cluster, subCluster, nodeFilter string,
metrics []string,
@@ -856,7 +751,7 @@ func (ccms *CCMetricStore) LoadNodeListData(
queries, assignedScope, err := ccms.buildNodeQueries(cluster, subCluster, nodes, metrics, scopes, resolution)
if err != nil {
cclog.Errorf("Error while building node queries for Cluster %s, SubCLuster %s, Metrics %v, Scopes %v: %s", cluster, subCluster, metrics, scopes, err.Error())
log.Warn("Error while building queries")
return nil, totalNodes, hasNextPage, err
}
@@ -871,7 +766,7 @@ func (ccms *CCMetricStore) LoadNodeListData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
cclog.Errorf("Error while performing request: %s", err.Error())
log.Error(fmt.Sprintf("Error while performing request %#v\n", err))
return nil, totalNodes, hasNextPage, err
}
@@ -976,7 +871,7 @@ func (ccms *CCMetricStore) buildNodeQueries(
if subCluster != "" {
subClusterTopol, scterr = archive.GetSubCluster(cluster, subCluster)
if scterr != nil {
cclog.Errorf("could not load cluster %s subCluster %s topology: %s", cluster, subCluster, scterr.Error())
// TODO: Log
return nil, nil, scterr
}
}
@@ -986,24 +881,10 @@ func (ccms *CCMetricStore) buildNodeQueries(
mc := archive.GetMetricConfig(cluster, metric)
if mc == nil {
// return nil, fmt.Errorf("METRICDATA/CCMS > metric '%s' is not specified for cluster '%s'", metric, cluster)
cclog.Warnf("metric '%s' is not specified for cluster '%s'", metric, cluster)
log.Infof("metric '%s' is not specified for cluster '%s'", metric, cluster)
continue
}
// Skip if metric is removed for subcluster
if mc.SubClusters != nil {
isRemoved := false
for _, scConfig := range mc.SubClusters {
if scConfig.Name == subCluster && scConfig.Remove {
isRemoved = true
break
}
}
if isRemoved {
continue
}
}
// Avoid duplicates...
handledScopes := make([]schema.MetricScope, 0, 3)
@@ -1157,23 +1038,6 @@ func (ccms *CCMetricStore) buildNodeQueries(
continue
}
// Core -> Socket
if nativeScope == schema.MetricScopeCore && scope == schema.MetricScopeSocket {
sockets, _ := topology.GetSocketsFromCores(topology.Node)
for _, socket := range sockets {
queries = append(queries, ApiQuery{
Metric: remoteName,
Hostname: hostname,
Aggregate: true,
Type: &coreString,
TypeIds: intToStringSlice(topology.Socket[socket]),
Resolution: resolution,
})
assignedScope = append(assignedScope, scope)
}
continue
}
// Core -> Node
if nativeScope == schema.MetricScopeCore && scope == schema.MetricScopeNode {
cores, _ := topology.GetCoresFromHWThreads(topology.Node)
@@ -1267,3 +1131,11 @@ func (ccms *CCMetricStore) buildNodeQueries(
return queries, assignedScope, nil
}
func intToStringSlice(is []int) []string {
ss := make([]string, len(is))
for i, x := range is {
ss[i] = strconv.Itoa(x)
}
return ss
}

View File

@@ -0,0 +1,333 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package metricdata
import (
"context"
"crypto/tls"
"encoding/json"
"errors"
"fmt"
"strings"
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
influxdb2 "github.com/influxdata/influxdb-client-go/v2"
influxdb2Api "github.com/influxdata/influxdb-client-go/v2/api"
)
type InfluxDBv2DataRepositoryConfig struct {
Url string `json:"url"`
Token string `json:"token"`
Bucket string `json:"bucket"`
Org string `json:"org"`
SkipTls bool `json:"skiptls"`
}
type InfluxDBv2DataRepository struct {
client influxdb2.Client
queryClient influxdb2Api.QueryAPI
bucket, measurement string
}
func (idb *InfluxDBv2DataRepository) Init(rawConfig json.RawMessage) error {
var config InfluxDBv2DataRepositoryConfig
if err := json.Unmarshal(rawConfig, &config); err != nil {
log.Warn("Error while unmarshaling raw json config")
return err
}
idb.client = influxdb2.NewClientWithOptions(config.Url, config.Token, influxdb2.DefaultOptions().SetTLSConfig(&tls.Config{InsecureSkipVerify: config.SkipTls}))
idb.queryClient = idb.client.QueryAPI(config.Org)
idb.bucket = config.Bucket
return nil
}
func (idb *InfluxDBv2DataRepository) formatTime(t time.Time) string {
return t.Format(time.RFC3339) // Like “2006-01-02T15:04:05Z07:00”
}
func (idb *InfluxDBv2DataRepository) epochToTime(epoch int64) time.Time {
return time.Unix(epoch, 0)
}
func (idb *InfluxDBv2DataRepository) LoadData(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
resolution int) (schema.JobData, error) {
measurementsConds := make([]string, 0, len(metrics))
for _, m := range metrics {
measurementsConds = append(measurementsConds, fmt.Sprintf(`r["_measurement"] == "%s"`, m))
}
measurementsCond := strings.Join(measurementsConds, " or ")
hostsConds := make([]string, 0, len(job.Resources))
for _, h := range job.Resources {
if h.HWThreads != nil || h.Accelerators != nil {
// TODO
return nil, errors.New("METRICDATA/INFLUXV2 > the InfluxDB metric data repository does not yet support HWThreads or Accelerators")
}
hostsConds = append(hostsConds, fmt.Sprintf(`r["hostname"] == "%s"`, h.Hostname))
}
hostsCond := strings.Join(hostsConds, " or ")
jobData := make(schema.JobData) // Empty Schema: map[<string>FIELD]map[<MetricScope>SCOPE]<*JobMetric>METRIC
// Requested Scopes
for _, scope := range scopes {
query := ""
switch scope {
case "node":
// Get Finest Granularity, Groupy By Measurement and Hostname (== Metric / Node), Calculate Mean for 60s windows
// log.Info("Scope 'node' requested. ")
query = fmt.Sprintf(`
from(bucket: "%s")
|> range(start: %s, stop: %s)
|> filter(fn: (r) => (%s) and (%s) )
|> drop(columns: ["_start", "_stop"])
|> group(columns: ["hostname", "_measurement"])
|> aggregateWindow(every: 60s, fn: mean)
|> drop(columns: ["_time"])`,
idb.bucket,
idb.formatTime(job.StartTime), idb.formatTime(idb.epochToTime(job.StartTimeUnix+int64(job.Duration)+int64(1))),
measurementsCond, hostsCond)
case "socket":
log.Info("Scope 'socket' requested, but not yet supported: Will return 'node' scope only. ")
continue
case "core":
log.Info(" Scope 'core' requested, but not yet supported: Will return 'node' scope only. ")
continue
// Get Finest Granularity only, Set NULL to 0.0
// query = fmt.Sprintf(`
// from(bucket: "%s")
// |> range(start: %s, stop: %s)
// |> filter(fn: (r) => %s )
// |> filter(fn: (r) => %s )
// |> drop(columns: ["_start", "_stop", "cluster"])
// |> map(fn: (r) => (if exists r._value then {r with _value: r._value} else {r with _value: 0.0}))`,
// idb.bucket,
// idb.formatTime(job.StartTime), idb.formatTime(idb.epochToTime(job.StartTimeUnix + int64(job.Duration) + int64(1) )),
// measurementsCond, hostsCond)
default:
log.Infof("Unknown scope '%s' requested: Will return 'node' scope.", scope)
continue
// return nil, errors.New("METRICDATA/INFLUXV2 > the InfluxDB metric data repository does not yet support other scopes than 'node'")
}
rows, err := idb.queryClient.Query(ctx, query)
if err != nil {
log.Error("Error while performing query")
return nil, err
}
// Init Metrics: Only Node level now -> TODO: Matching /check on scope level ...
for _, metric := range metrics {
jobMetric, ok := jobData[metric]
if !ok {
mc := archive.GetMetricConfig(job.Cluster, metric)
jobMetric = map[schema.MetricScope]*schema.JobMetric{
scope: { // uses scope var from above!
Unit: mc.Unit,
Timestep: mc.Timestep,
Series: make([]schema.Series, 0, len(job.Resources)),
StatisticsSeries: nil, // Should be: &schema.StatsSeries{},
},
}
}
jobData[metric] = jobMetric
}
// Process Result: Time-Data
field, host, hostSeries := "", "", schema.Series{}
// typeId := 0
switch scope {
case "node":
for rows.Next() {
row := rows.Record()
if host == "" || host != row.ValueByKey("hostname").(string) || rows.TableChanged() {
if host != "" {
// Append Series before reset
jobData[field][scope].Series = append(jobData[field][scope].Series, hostSeries)
}
field, host = row.Measurement(), row.ValueByKey("hostname").(string)
hostSeries = schema.Series{
Hostname: host,
Statistics: schema.MetricStatistics{}, //TODO Add Statistics
Data: make([]schema.Float, 0),
}
}
val, ok := row.Value().(float64)
if ok {
hostSeries.Data = append(hostSeries.Data, schema.Float(val))
} else {
hostSeries.Data = append(hostSeries.Data, schema.Float(0))
}
}
case "socket":
continue
case "core":
continue
// Include Series.Id in hostSeries
// for rows.Next() {
// row := rows.Record()
// if ( host == "" || host != row.ValueByKey("hostname").(string) || typeId != row.ValueByKey("type-id").(int) || rows.TableChanged() ) {
// if ( host != "" ) {
// // Append Series before reset
// jobData[field][scope].Series = append(jobData[field][scope].Series, hostSeries)
// }
// field, host, typeId = row.Measurement(), row.ValueByKey("hostname").(string), row.ValueByKey("type-id").(int)
// hostSeries = schema.Series{
// Hostname: host,
// Id: &typeId,
// Statistics: nil,
// Data: make([]schema.Float, 0),
// }
// }
// val := row.Value().(float64)
// hostSeries.Data = append(hostSeries.Data, schema.Float(val))
// }
default:
log.Infof("Unknown scope '%s' requested: Will return 'node' scope.", scope)
continue
// return nil, errors.New("the InfluxDB metric data repository does not yet support other scopes than 'node, core'")
}
// Append last Series
jobData[field][scope].Series = append(jobData[field][scope].Series, hostSeries)
}
// Get Stats
stats, err := idb.LoadStats(job, metrics, ctx)
if err != nil {
log.Warn("Error while loading statistics")
return nil, err
}
for _, scope := range scopes {
if scope == "node" { // No 'socket/core' support yet
for metric, nodes := range stats {
for node, stats := range nodes {
for index, _ := range jobData[metric][scope].Series {
if jobData[metric][scope].Series[index].Hostname == node {
jobData[metric][scope].Series[index].Statistics = schema.MetricStatistics{Avg: stats.Avg, Min: stats.Min, Max: stats.Max}
}
}
}
}
}
}
return jobData, nil
}
func (idb *InfluxDBv2DataRepository) LoadStats(
job *schema.Job,
metrics []string,
ctx context.Context) (map[string]map[string]schema.MetricStatistics, error) {
stats := map[string]map[string]schema.MetricStatistics{}
hostsConds := make([]string, 0, len(job.Resources))
for _, h := range job.Resources {
if h.HWThreads != nil || h.Accelerators != nil {
// TODO
return nil, errors.New("METRICDATA/INFLUXV2 > the InfluxDB metric data repository does not yet support HWThreads or Accelerators")
}
hostsConds = append(hostsConds, fmt.Sprintf(`r["hostname"] == "%s"`, h.Hostname))
}
hostsCond := strings.Join(hostsConds, " or ")
// lenMet := len(metrics)
for _, metric := range metrics {
// log.Debugf("<< You are here: %s (Index %d of %d metrics)", metric, index, lenMet)
query := fmt.Sprintf(`
data = from(bucket: "%s")
|> range(start: %s, stop: %s)
|> filter(fn: (r) => r._measurement == "%s" and r._field == "value" and (%s))
union(tables: [data |> mean(column: "_value") |> set(key: "_field", value: "avg"),
data |> min(column: "_value") |> set(key: "_field", value: "min"),
data |> max(column: "_value") |> set(key: "_field", value: "max")])
|> pivot(rowKey: ["hostname"], columnKey: ["_field"], valueColumn: "_value")
|> group()`,
idb.bucket,
idb.formatTime(job.StartTime), idb.formatTime(idb.epochToTime(job.StartTimeUnix+int64(job.Duration)+int64(1))),
metric, hostsCond)
rows, err := idb.queryClient.Query(ctx, query)
if err != nil {
log.Error("Error while performing query")
return nil, err
}
nodes := map[string]schema.MetricStatistics{}
for rows.Next() {
row := rows.Record()
host := row.ValueByKey("hostname").(string)
avg, avgok := row.ValueByKey("avg").(float64)
if !avgok {
// log.Debugf(">> Assertion error for metric %s, statistic AVG. Expected 'float64', got %v", metric, avg)
avg = 0.0
}
min, minok := row.ValueByKey("min").(float64)
if !minok {
// log.Debugf(">> Assertion error for metric %s, statistic MIN. Expected 'float64', got %v", metric, min)
min = 0.0
}
max, maxok := row.ValueByKey("max").(float64)
if !maxok {
// log.Debugf(">> Assertion error for metric %s, statistic MAX. Expected 'float64', got %v", metric, max)
max = 0.0
}
nodes[host] = schema.MetricStatistics{
Avg: avg,
Min: min,
Max: max,
}
}
stats[metric] = nodes
}
return stats, nil
}
func (idb *InfluxDBv2DataRepository) LoadNodeData(
cluster string,
metrics, nodes []string,
scopes []schema.MetricScope,
from, to time.Time,
ctx context.Context) (map[string]map[string][]*schema.JobMetric, error) {
// TODO : Implement to be used in Analysis- und System/Node-View
log.Infof("LoadNodeData unimplemented for InfluxDBv2DataRepository, Args: cluster %s, metrics %v, nodes %v, scopes %v", cluster, metrics, nodes, scopes)
return nil, errors.New("METRICDATA/INFLUXV2 > unimplemented for InfluxDBv2DataRepository")
}
func (idb *InfluxDBv2DataRepository) LoadNodeListData(
cluster, subCluster, nodeFilter string,
metrics []string,
scopes []schema.MetricScope,
resolution int,
from, to time.Time,
page *model.PageRequest,
ctx context.Context,
) (map[string]schema.JobData, int, bool, error) {
var totalNodes int = 0
var hasNextPage bool = false
// TODO : Implement to be used in NodeList-View
log.Infof("LoadNodeListData unimplemented for InfluxDBv2DataRepository, Args: cluster %s, metrics %v, nodeFilter %v, scopes %v", cluster, metrics, nodeFilter, scopes)
return nil, totalNodes, hasNextPage, errors.New("METRICDATA/INFLUXV2 > unimplemented for InfluxDBv2DataRepository")
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package metricdata
import (
@@ -13,9 +12,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/internal/memorystore"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
type MetricDataRepository interface {
@@ -26,12 +24,9 @@ type MetricDataRepository interface {
// Return the JobData for the given job, only with the requested metrics.
LoadData(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context, resolution int) (schema.JobData, error)
// Return a map of metrics to a map of nodes to the metric statistics of the job. node scope only.
// Return a map of metrics to a map of nodes to the metric statistics of the job. node scope assumed for now.
LoadStats(job *schema.Job, metrics []string, ctx context.Context) (map[string]map[string]schema.MetricStatistics, error)
// Return a map of metrics to a map of scopes to the scoped metric statistics of the job.
LoadScopedStats(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context) (schema.ScopedJobStats, error)
// Return a map of hosts to a map of metrics at the requested scopes (currently only node) for that node.
LoadNodeData(cluster string, metrics, nodes []string, scopes []schema.MetricScope, from, to time.Time, ctx context.Context) (map[string]map[string][]*schema.JobMetric, error)
@@ -42,13 +37,13 @@ type MetricDataRepository interface {
var metricDataRepos map[string]MetricDataRepository = map[string]MetricDataRepository{}
func Init() error {
for _, cluster := range config.Clusters {
for _, cluster := range config.Keys.Clusters {
if cluster.MetricDataRepository != nil {
var kind struct {
Kind string `json:"kind"`
}
if err := json.Unmarshal(cluster.MetricDataRepository, &kind); err != nil {
cclog.Warn("Error while unmarshaling raw json MetricDataRepository")
log.Warn("Error while unmarshaling raw json MetricDataRepository")
return err
}
@@ -56,9 +51,8 @@ func Init() error {
switch kind.Kind {
case "cc-metric-store":
mdr = &CCMetricStore{}
case "cc-metric-store-internal":
mdr = &CCMetricStoreInternal{}
memorystore.InternalCCMSFlag = true
case "influxdb":
mdr = &InfluxDBv2DataRepository{}
case "prometheus":
mdr = &PrometheusDataRepository{}
case "test":
@@ -68,7 +62,7 @@ func Init() error {
}
if err := mdr.Init(cluster.MetricDataRepository); err != nil {
cclog.Errorf("Error initializing MetricDataRepository %v for cluster %v", kind.Kind, cluster.Name)
log.Errorf("Error initializing MetricDataRepository %v for cluster %v", kind.Kind, cluster.Name)
return err
}
metricDataRepos[cluster.Name] = mdr

View File

@@ -1,5 +1,5 @@
// Copyright (C) 2022 DKRZ
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package metricdata
@@ -22,8 +22,8 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
promapi "github.com/prometheus/client_golang/api"
promv1 "github.com/prometheus/client_golang/api/prometheus/v1"
promcfg "github.com/prometheus/common/config"
@@ -160,7 +160,7 @@ func (pdb *PrometheusDataRepository) Init(rawConfig json.RawMessage) error {
var config PrometheusDataRepositoryConfig
// parse config
if err := json.Unmarshal(rawConfig, &config); err != nil {
cclog.Warn("Error while unmarshaling raw json config")
log.Warn("Error while unmarshaling raw json config")
return err
}
// support basic authentication
@@ -179,7 +179,7 @@ func (pdb *PrometheusDataRepository) Init(rawConfig json.RawMessage) error {
RoundTripper: rt,
})
if err != nil {
cclog.Error("Error while initializing new prometheus client")
log.Error("Error while initializing new prometheus client")
return err
}
// init query client
@@ -192,9 +192,9 @@ func (pdb *PrometheusDataRepository) Init(rawConfig json.RawMessage) error {
for metric, templ := range config.Templates {
pdb.templates[metric], err = template.New(metric).Parse(templ)
if err == nil {
cclog.Debugf("Added PromQL template for %s: %s", metric, templ)
log.Debugf("Added PromQL template for %s: %s", metric, templ)
} else {
cclog.Warnf("Failed to parse PromQL template %s for metric %s", templ, metric)
log.Warnf("Failed to parse PromQL template %s for metric %s", templ, metric)
}
}
return nil
@@ -221,7 +221,7 @@ func (pdb *PrometheusDataRepository) FormatQuery(
return "", errors.New(fmt.Sprintf("METRICDATA/PROMETHEUS > Error compiling template %v", templ))
} else {
query := buf.String()
cclog.Debugf("PromQL: %s", query)
log.Debugf("PromQL: %s", query)
return query, nil
}
} else {
@@ -279,13 +279,13 @@ func (pdb *PrometheusDataRepository) LoadData(
for i, resource := range job.Resources {
nodes[i] = resource.Hostname
}
from := time.Unix(job.StartTime, 0)
to := time.Unix(job.StartTime+int64(job.Duration), 0)
from := job.StartTime
to := job.StartTime.Add(time.Duration(job.Duration) * time.Second)
for _, scope := range scopes {
if scope != schema.MetricScopeNode {
logOnce.Do(func() {
cclog.Infof("Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
log.Infof("Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
})
continue
}
@@ -293,12 +293,12 @@ func (pdb *PrometheusDataRepository) LoadData(
for _, metric := range metrics {
metricConfig := archive.GetMetricConfig(job.Cluster, metric)
if metricConfig == nil {
cclog.Warnf("Error in LoadData: Metric %s for cluster %s not configured", metric, job.Cluster)
log.Warnf("Error in LoadData: Metric %s for cluster %s not configured", metric, job.Cluster)
return nil, errors.New("Prometheus config error")
}
query, err := pdb.FormatQuery(metric, scope, nodes, job.Cluster)
if err != nil {
cclog.Warn("Error while formatting prometheus query")
log.Warn("Error while formatting prometheus query")
return nil, err
}
@@ -310,11 +310,11 @@ func (pdb *PrometheusDataRepository) LoadData(
}
result, warnings, err := pdb.queryClient.QueryRange(ctx, query, r)
if err != nil {
cclog.Errorf("Prometheus query error in LoadData: %v\nQuery: %s", err, query)
log.Errorf("Prometheus query error in LoadData: %v\nQuery: %s", err, query)
return nil, errors.New("Prometheus query error")
}
if len(warnings) > 0 {
cclog.Warnf("Warnings: %v\n", warnings)
log.Warnf("Warnings: %v\n", warnings)
}
// init data structures
@@ -360,7 +360,7 @@ func (pdb *PrometheusDataRepository) LoadStats(
data, err := pdb.LoadData(job, metrics, []schema.MetricScope{schema.MetricScopeNode}, ctx, 0 /*resolution here*/)
if err != nil {
cclog.Warn("Error while loading job for stats")
log.Warn("Error while loading job for stats")
return nil, err
}
for metric, metricData := range data {
@@ -391,19 +391,19 @@ func (pdb *PrometheusDataRepository) LoadNodeData(
for _, scope := range scopes {
if scope != schema.MetricScopeNode {
logOnce.Do(func() {
cclog.Infof("Note: Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
log.Infof("Note: Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
})
continue
}
for _, metric := range metrics {
metricConfig := archive.GetMetricConfig(cluster, metric)
if metricConfig == nil {
cclog.Warnf("Error in LoadNodeData: Metric %s for cluster %s not configured", metric, cluster)
log.Warnf("Error in LoadNodeData: Metric %s for cluster %s not configured", metric, cluster)
return nil, errors.New("Prometheus config error")
}
query, err := pdb.FormatQuery(metric, scope, nodes, cluster)
if err != nil {
cclog.Warn("Error while formatting prometheus query")
log.Warn("Error while formatting prometheus query")
return nil, err
}
@@ -415,11 +415,11 @@ func (pdb *PrometheusDataRepository) LoadNodeData(
}
result, warnings, err := pdb.queryClient.QueryRange(ctx, query, r)
if err != nil {
cclog.Errorf("Prometheus query error in LoadNodeData: %v\n", err)
log.Errorf("Prometheus query error in LoadNodeData: %v\n", err)
return nil, errors.New("Prometheus query error")
}
if len(warnings) > 0 {
cclog.Warnf("Warnings: %v\n", warnings)
log.Warnf("Warnings: %v\n", warnings)
}
step := int64(metricConfig.Timestep)
@@ -444,55 +444,10 @@ func (pdb *PrometheusDataRepository) LoadNodeData(
}
}
t1 := time.Since(t0)
cclog.Debugf("LoadNodeData of %v nodes took %s", len(data), t1)
log.Debugf("LoadNodeData of %v nodes took %s", len(data), t1)
return data, nil
}
// Implemented by NHR@FAU; Used in Job-View StatsTable
func (pdb *PrometheusDataRepository) LoadScopedStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
) (schema.ScopedJobStats, error) {
// Assumption: pdb.loadData() only returns series node-scope - use node scope for statsTable
scopedJobStats := make(schema.ScopedJobStats)
data, err := pdb.LoadData(job, metrics, []schema.MetricScope{schema.MetricScopeNode}, ctx, 0 /*resolution here*/)
if err != nil {
cclog.Warn("Error while loading job for scopedJobStats")
return nil, err
}
for metric, metricData := range data {
for _, scope := range scopes {
if scope != schema.MetricScopeNode {
logOnce.Do(func() {
cclog.Infof("Note: Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
})
continue
}
if _, ok := scopedJobStats[metric]; !ok {
scopedJobStats[metric] = make(map[schema.MetricScope][]*schema.ScopedStats)
}
if _, ok := scopedJobStats[metric][scope]; !ok {
scopedJobStats[metric][scope] = make([]*schema.ScopedStats, 0)
}
for _, series := range metricData[scope].Series {
scopedJobStats[metric][scope] = append(scopedJobStats[metric][scope], &schema.ScopedStats{
Hostname: series.Hostname,
Data: &series.Statistics,
})
}
}
}
return scopedJobStats, nil
}
// Implemented by NHR@FAU; Used in NodeList-View
func (pdb *PrometheusDataRepository) LoadNodeListData(
cluster, subCluster, nodeFilter string,
metrics []string,
@@ -502,132 +457,11 @@ func (pdb *PrometheusDataRepository) LoadNodeListData(
page *model.PageRequest,
ctx context.Context,
) (map[string]schema.JobData, int, bool, error) {
// Assumption: pdb.loadData() only returns series node-scope - use node scope for NodeList
// 0) Init additional vars
var totalNodes int = 0
var hasNextPage bool = false
// TODO : Implement to be used in NodeList-View
log.Infof("LoadNodeListData unimplemented for PrometheusDataRepository, Args: cluster %s, metrics %v, nodeFilter %v, scopes %v", cluster, metrics, nodeFilter, scopes)
// 1) Get list of all nodes
var nodes []string
if subCluster != "" {
scNodes := archive.NodeLists[cluster][subCluster]
nodes = scNodes.PrintList()
} else {
subClusterNodeLists := archive.NodeLists[cluster]
for _, nodeList := range subClusterNodeLists {
nodes = append(nodes, nodeList.PrintList()...)
}
}
// 2) Filter nodes
if nodeFilter != "" {
filteredNodes := []string{}
for _, node := range nodes {
if strings.Contains(node, nodeFilter) {
filteredNodes = append(filteredNodes, node)
}
}
nodes = filteredNodes
}
// 2.1) Count total nodes && Sort nodes -> Sorting invalidated after return ...
totalNodes = len(nodes)
sort.Strings(nodes)
// 3) Apply paging
if len(nodes) > page.ItemsPerPage {
start := (page.Page - 1) * page.ItemsPerPage
end := start + page.ItemsPerPage
if end >= len(nodes) {
end = len(nodes)
hasNextPage = false
} else {
hasNextPage = true
}
nodes = nodes[start:end]
}
// 4) Fetch Data, based on pdb.LoadNodeData()
t0 := time.Now()
// Map of hosts of jobData
data := make(map[string]schema.JobData)
// query db for each metric
// TODO: scopes seems to be always empty
if len(scopes) == 0 || !contains(scopes, schema.MetricScopeNode) {
scopes = append(scopes, schema.MetricScopeNode)
}
for _, scope := range scopes {
if scope != schema.MetricScopeNode {
logOnce.Do(func() {
cclog.Infof("Note: Scope '%s' requested, but not yet supported: Will return 'node' scope only.", scope)
})
continue
}
for _, metric := range metrics {
metricConfig := archive.GetMetricConfig(cluster, metric)
if metricConfig == nil {
cclog.Warnf("Error in LoadNodeListData: Metric %s for cluster %s not configured", metric, cluster)
return nil, totalNodes, hasNextPage, errors.New("Prometheus config error")
}
query, err := pdb.FormatQuery(metric, scope, nodes, cluster)
if err != nil {
cclog.Warn("Error while formatting prometheus query")
return nil, totalNodes, hasNextPage, err
}
// ranged query over all nodes
r := promv1.Range{
Start: from,
End: to,
Step: time.Duration(metricConfig.Timestep * 1e9),
}
result, warnings, err := pdb.queryClient.QueryRange(ctx, query, r)
if err != nil {
cclog.Errorf("Prometheus query error in LoadNodeData: %v\n", err)
return nil, totalNodes, hasNextPage, errors.New("Prometheus query error")
}
if len(warnings) > 0 {
cclog.Warnf("Warnings: %v\n", warnings)
}
step := int64(metricConfig.Timestep)
steps := int64(to.Sub(from).Seconds()) / step
// iter rows of host, metric, values
for _, row := range result.(promm.Matrix) {
hostname := strings.TrimSuffix(string(row.Metric["exported_instance"]), pdb.suffix)
hostdata, ok := data[hostname]
if !ok {
hostdata = make(schema.JobData)
data[hostname] = hostdata
}
metricdata, ok := hostdata[metric]
if !ok {
metricdata = make(map[schema.MetricScope]*schema.JobMetric)
data[hostname][metric] = metricdata
}
// output per host, metric and scope
scopeData, ok := metricdata[scope]
if !ok {
scopeData = &schema.JobMetric{
Unit: metricConfig.Unit,
Timestep: metricConfig.Timestep,
Series: []schema.Series{pdb.RowToSeries(from, step, steps, row)},
}
data[hostname][metric][scope] = scopeData
}
}
}
}
t1 := time.Since(t0)
cclog.Debugf("LoadNodeListData of %v nodes took %s", len(data), t1)
return data, totalNodes, hasNextPage, nil
return nil, totalNodes, hasNextPage, errors.New("METRICDATA/INFLUXV2 > unimplemented for PrometheusDataRepository")
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package metricdata
import (
@@ -11,14 +10,14 @@ import (
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
var TestLoadDataCallback func(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context, resolution int) (schema.JobData, error) = func(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context, resolution int) (schema.JobData, error) {
panic("TODO")
}
// TestMetricDataRepository is only a mock for unit-testing.
// Only a mock for unit-testing.
type TestMetricDataRepository struct{}
func (tmdr *TestMetricDataRepository) Init(_ json.RawMessage) error {
@@ -30,25 +29,15 @@ func (tmdr *TestMetricDataRepository) LoadData(
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
resolution int,
) (schema.JobData, error) {
resolution int) (schema.JobData, error) {
return TestLoadDataCallback(job, metrics, scopes, ctx, resolution)
}
func (tmdr *TestMetricDataRepository) LoadStats(
job *schema.Job,
metrics []string,
ctx context.Context,
) (map[string]map[string]schema.MetricStatistics, error) {
panic("TODO")
}
metrics []string, ctx context.Context) (map[string]map[string]schema.MetricStatistics, error) {
func (tmdr *TestMetricDataRepository) LoadScopedStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context,
) (schema.ScopedJobStats, error) {
panic("TODO")
}
@@ -57,8 +46,8 @@ func (tmdr *TestMetricDataRepository) LoadNodeData(
metrics, nodes []string,
scopes []schema.MetricScope,
from, to time.Time,
ctx context.Context,
) (map[string]map[string][]*schema.JobMetric, error) {
ctx context.Context) (map[string]map[string][]*schema.JobMetric, error) {
panic("TODO")
}
@@ -71,13 +60,16 @@ func (tmdr *TestMetricDataRepository) LoadNodeListData(
page *model.PageRequest,
ctx context.Context,
) (map[string]schema.JobData, int, bool, error) {
panic("TODO")
}
func DeepCopy(jdTemp schema.JobData) schema.JobData {
jd := make(schema.JobData, len(jdTemp))
for k, v := range jdTemp {
jd[k] = make(map[schema.MetricScope]*schema.JobMetric, len(jdTemp[k]))
func DeepCopy(jd_temp schema.JobData) schema.JobData {
var jd schema.JobData
jd = make(schema.JobData, len(jd_temp))
for k, v := range jd_temp {
jd[k] = make(map[schema.MetricScope]*schema.JobMetric, len(jd_temp[k]))
for k_, v_ := range v {
jd[k][k_] = new(schema.JobMetric)
jd[k][k_].Series = make([]schema.Series, len(v_.Series))

View File

@@ -1,17 +1,15 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
"database/sql"
"fmt"
"net/url"
"sync"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/jmoiron/sqlx"
"github.com/mattn/go-sqlite3"
"github.com/qustavo/sqlhooks/v2"
@@ -35,27 +33,6 @@ type DatabaseOptions struct {
ConnectionMaxIdleTime time.Duration
}
func setupSqlite(db *sql.DB) (err error) {
pragmas := []string{
// "journal_mode = WAL",
// "busy_timeout = 5000",
// "synchronous = NORMAL",
// "cache_size = 1000000000", // 1GB
// "foreign_keys = true",
"temp_store = memory",
// "mmap_size = 3000000000",
}
for _, pragma := range pragmas {
_, err = db.Exec("PRAGMA " + pragma)
if err != nil {
return
}
}
return nil
}
func Connect(driver string, db string) {
var err error
var dbHandle *sqlx.DB
@@ -71,34 +48,28 @@ func Connect(driver string, db string) {
switch driver {
case "sqlite3":
// TODO: Have separate DB handles for Writes and Reads
// Optimize SQLite connection: https://kerkour.com/sqlite-for-servers
connectionUrlParams := make(url.Values)
connectionUrlParams.Add("_txlock", "immediate")
connectionUrlParams.Add("_journal_mode", "WAL")
connectionUrlParams.Add("_busy_timeout", "5000")
connectionUrlParams.Add("_synchronous", "NORMAL")
connectionUrlParams.Add("_cache_size", "1000000000")
connectionUrlParams.Add("_foreign_keys", "true")
opts.URL = fmt.Sprintf("file:%s?%s", opts.URL, connectionUrlParams.Encode())
// - Set WAL mode (not strictly necessary each time because it's persisted in the database, but good for first run)
// - Set busy timeout, so concurrent writers wait on each other instead of erroring immediately
// - Enable foreign key checks
opts.URL += "?_journal=WAL&_timeout=5000&_fk=true"
if cclog.Loglevel() == "debug" {
if log.Loglevel() == "debug" {
sql.Register("sqlite3WithHooks", sqlhooks.Wrap(&sqlite3.SQLiteDriver{}, &Hooks{}))
dbHandle, err = sqlx.Open("sqlite3WithHooks", opts.URL)
} else {
dbHandle, err = sqlx.Open("sqlite3", opts.URL)
}
setupSqlite(dbHandle.DB)
if err != nil {
log.Fatal(err)
}
case "mysql":
opts.URL += "?multiStatements=true"
dbHandle, err = sqlx.Open("mysql", opts.URL)
if err != nil {
log.Fatalf("sqlx.Open() error: %v", err)
}
default:
cclog.Abortf("DB Connection: Unsupported database driver '%s'.\n", driver)
}
if err != nil {
cclog.Abortf("DB Connection: Could not connect to '%s' database with sqlx.Open().\nError: %s\n", driver, err.Error())
log.Fatalf("unsupported database driver: %s", driver)
}
dbHandle.SetMaxOpenConns(opts.MaxOpenConnections)
@@ -109,14 +80,14 @@ func Connect(driver string, db string) {
dbConnInstance = &DBConnection{DB: dbHandle, Driver: driver}
err = checkDBVersion(driver, dbHandle.DB)
if err != nil {
cclog.Abortf("DB Connection: Failed DB version check.\nError: %s\n", err.Error())
log.Fatal(err)
}
})
}
func GetConnection() *DBConnection {
if dbConnInstance == nil {
cclog.Fatalf("Database connection not initialized!")
log.Fatalf("Database connection not initialized!")
}
return dbConnInstance

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -8,21 +8,21 @@ import (
"context"
"time"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/log"
)
// Hooks satisfies the sqlhook.Hooks interface
type Hooks struct{}
// Before hook will print the query with it's args and return the context with the timestamp
func (h *Hooks) Before(ctx context.Context, query string, args ...any) (context.Context, error) {
cclog.Debugf("SQL query %s %q", query, args)
func (h *Hooks) Before(ctx context.Context, query string, args ...interface{}) (context.Context, error) {
log.Debugf("SQL query %s %q", query, args)
return context.WithValue(ctx, "begin", time.Now()), nil
}
// After hook will get the timestamp registered on the Before hook and print the elapsed time
func (h *Hooks) After(ctx context.Context, query string, args ...any) (context.Context, error) {
func (h *Hooks) After(ctx context.Context, query string, args ...interface{}) (context.Context, error) {
begin := ctx.Value("begin").(time.Time)
cclog.Debugf("Took: %s\n", time.Since(begin))
log.Debugf("Took: %s\n", time.Since(begin))
return ctx, nil
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -9,16 +9,16 @@ import (
"encoding/json"
"errors"
"fmt"
"maps"
"math"
"strconv"
"sync"
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/lrucache"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/lrucache"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
"github.com/jmoiron/sqlx"
)
@@ -33,7 +33,6 @@ type JobRepository struct {
stmtCache *sq.StmtCache
cache *lrucache.Cache
driver string
Mutex sync.Mutex
}
func GetJobRepository() *JobRepository {
@@ -52,49 +51,38 @@ func GetJobRepository() *JobRepository {
}
var jobColumns []string = []string{
"job.id", "job.job_id", "job.hpc_user", "job.project", "job.cluster", "job.subcluster",
"job.start_time", "job.cluster_partition", "job.array_job_id", "job.num_nodes",
"job.num_hwthreads", "job.num_acc", "job.shared", "job.monitoring_status",
"job.smt", "job.job_state", "job.duration", "job.walltime", "job.resources",
"job.footprint", "job.energy",
"job.id", "job.job_id", "job.hpc_user", "job.project", "job.cluster", "job.subcluster", "job.start_time", "job.cluster_partition", "job.array_job_id",
"job.num_nodes", "job.num_hwthreads", "job.num_acc", "job.exclusive", "job.monitoring_status", "job.smt", "job.job_state",
"job.duration", "job.walltime", "job.resources", "job.footprint", "job.energy",
}
var jobCacheColumns []string = []string{
"job_cache.id", "job_cache.job_id", "job_cache.hpc_user", "job_cache.project", "job_cache.cluster",
"job_cache.subcluster", "job_cache.start_time", "job_cache.cluster_partition",
"job_cache.array_job_id", "job_cache.num_nodes", "job_cache.num_hwthreads",
"job_cache.num_acc", "job_cache.shared", "job_cache.monitoring_status", "job_cache.smt",
"job_cache.job_state", "job_cache.duration", "job_cache.walltime", "job_cache.resources",
"job_cache.footprint", "job_cache.energy",
}
func scanJob(row interface{ Scan(...any) error }) (*schema.Job, error) {
func scanJob(row interface{ Scan(...interface{}) error }) (*schema.Job, error) {
job := &schema.Job{}
if err := row.Scan(
&job.ID, &job.JobID, &job.User, &job.Project, &job.Cluster, &job.SubCluster,
&job.StartTime, &job.Partition, &job.ArrayJobID, &job.NumNodes, &job.NumHWThreads,
&job.NumAcc, &job.Shared, &job.MonitoringStatus, &job.SMT, &job.State,
&job.ID, &job.JobID, &job.User, &job.Project, &job.Cluster, &job.SubCluster, &job.StartTimeUnix, &job.Partition, &job.ArrayJobId,
&job.NumNodes, &job.NumHWThreads, &job.NumAcc, &job.Exclusive, &job.MonitoringStatus, &job.SMT, &job.State,
&job.Duration, &job.Walltime, &job.RawResources, &job.RawFootprint, &job.Energy); err != nil {
cclog.Warnf("Error while scanning rows (Job): %v", err)
log.Warnf("Error while scanning rows (Job): %v", err)
return nil, err
}
if err := json.Unmarshal(job.RawResources, &job.Resources); err != nil {
cclog.Warn("Error while unmarshaling raw resources json")
log.Warn("Error while unmarshaling raw resources json")
return nil, err
}
job.RawResources = nil
if err := json.Unmarshal(job.RawFootprint, &job.Footprint); err != nil {
cclog.Warnf("Error while unmarshaling raw footprint json: %v", err)
log.Warnf("Error while unmarshaling raw footprint json: %v", err)
return nil, err
}
job.RawFootprint = nil
job.StartTime = time.Unix(job.StartTimeUnix, 0)
// Always ensure accurate duration for running jobs
if job.State == schema.JobStateRunning {
job.Duration = int32(time.Now().Unix() - job.StartTime)
job.Duration = int32(time.Since(job.StartTime).Seconds())
}
return job, nil
@@ -109,7 +97,7 @@ func (r *JobRepository) Optimize() error {
return err
}
case "mysql":
cclog.Info("Optimize currently not supported for mysql driver")
log.Info("Optimize currently not supported for mysql driver")
}
return nil
@@ -150,6 +138,17 @@ func (r *JobRepository) Flush() error {
return nil
}
func scanJobLink(row interface{ Scan(...interface{}) error }) (*model.JobLink, error) {
jobLink := &model.JobLink{}
if err := row.Scan(
&jobLink.ID, &jobLink.JobID); err != nil {
log.Warn("Error while scanning rows (jobLink)")
return nil, err
}
return jobLink, nil
}
func (r *JobRepository) FetchMetadata(job *schema.Job) (map[string]string, error) {
start := time.Now()
cachekey := fmt.Sprintf("metadata:%d", job.ID)
@@ -160,7 +159,7 @@ func (r *JobRepository) FetchMetadata(job *schema.Job) (map[string]string, error
if err := sq.Select("job.meta_data").From("job").Where("job.id = ?", job.ID).
RunWith(r.stmtCache).QueryRow().Scan(&job.RawMetaData); err != nil {
cclog.Warn("Error while scanning for job metadata")
log.Warn("Error while scanning for job metadata")
return nil, err
}
@@ -169,12 +168,12 @@ func (r *JobRepository) FetchMetadata(job *schema.Job) (map[string]string, error
}
if err := json.Unmarshal(job.RawMetaData, &job.MetaData); err != nil {
cclog.Warn("Error while unmarshaling raw metadata json")
log.Warn("Error while unmarshaling raw metadata json")
return nil, err
}
r.cache.Put(cachekey, job.MetaData, len(job.RawMetaData), 24*time.Hour)
cclog.Debugf("Timer FetchMetadata %s", time.Since(start))
log.Debugf("Timer FetchMetadata %s", time.Since(start))
return job.MetaData, nil
}
@@ -183,14 +182,16 @@ func (r *JobRepository) UpdateMetadata(job *schema.Job, key, val string) (err er
r.cache.Del(cachekey)
if job.MetaData == nil {
if _, err = r.FetchMetadata(job); err != nil {
cclog.Warnf("Error while fetching metadata for job, DB ID '%v'", job.ID)
log.Warnf("Error while fetching metadata for job, DB ID '%v'", job.ID)
return err
}
}
if job.MetaData != nil {
cpy := make(map[string]string, len(job.MetaData)+1)
maps.Copy(cpy, job.MetaData)
for k, v := range job.MetaData {
cpy[k] = v
}
cpy[key] = val
job.MetaData = cpy
} else {
@@ -198,7 +199,7 @@ func (r *JobRepository) UpdateMetadata(job *schema.Job, key, val string) (err er
}
if job.RawMetaData, err = json.Marshal(job.MetaData); err != nil {
cclog.Warnf("Error while marshaling metadata for job, DB ID '%v'", job.ID)
log.Warnf("Error while marshaling metadata for job, DB ID '%v'", job.ID)
return err
}
@@ -206,7 +207,7 @@ func (r *JobRepository) UpdateMetadata(job *schema.Job, key, val string) (err er
Set("meta_data", job.RawMetaData).
Where("job.id = ?", job.ID).
RunWith(r.stmtCache).Exec(); err != nil {
cclog.Warnf("Error while updating metadata for job, DB ID '%v'", job.ID)
log.Warnf("Error while updating metadata for job, DB ID '%v'", job.ID)
return err
}
@@ -216,10 +217,15 @@ func (r *JobRepository) UpdateMetadata(job *schema.Job, key, val string) (err er
func (r *JobRepository) FetchFootprint(job *schema.Job) (map[string]float64, error) {
start := time.Now()
cachekey := fmt.Sprintf("footprint:%d", job.ID)
if cached := r.cache.Get(cachekey, nil); cached != nil {
job.Footprint = cached.(map[string]float64)
return job.Footprint, nil
}
if err := sq.Select("job.footprint").From("job").Where("job.id = ?", job.ID).
RunWith(r.stmtCache).QueryRow().Scan(&job.RawFootprint); err != nil {
cclog.Warn("Error while scanning for job footprint")
log.Warn("Error while scanning for job footprint")
return nil, err
}
@@ -228,11 +234,12 @@ func (r *JobRepository) FetchFootprint(job *schema.Job) (map[string]float64, err
}
if err := json.Unmarshal(job.RawFootprint, &job.Footprint); err != nil {
cclog.Warn("Error while unmarshaling raw footprint json")
log.Warn("Error while unmarshaling raw footprint json")
return nil, err
}
cclog.Debugf("Timer FetchFootprint %s", time.Since(start))
r.cache.Put(cachekey, job.Footprint, len(job.Footprint), 24*time.Hour)
log.Debugf("Timer FetchFootprint %s", time.Since(start))
return job.Footprint, nil
}
@@ -246,7 +253,7 @@ func (r *JobRepository) FetchEnergyFootprint(job *schema.Job) (map[string]float6
if err := sq.Select("job.energy_footprint").From("job").Where("job.id = ?", job.ID).
RunWith(r.stmtCache).QueryRow().Scan(&job.RawEnergyFootprint); err != nil {
cclog.Warn("Error while scanning for job energy_footprint")
log.Warn("Error while scanning for job energy_footprint")
return nil, err
}
@@ -255,12 +262,12 @@ func (r *JobRepository) FetchEnergyFootprint(job *schema.Job) (map[string]float6
}
if err := json.Unmarshal(job.RawEnergyFootprint, &job.EnergyFootprint); err != nil {
cclog.Warn("Error while unmarshaling raw energy footprint json")
log.Warn("Error while unmarshaling raw energy footprint json")
return nil, err
}
r.cache.Put(cachekey, job.EnergyFootprint, len(job.EnergyFootprint), 24*time.Hour)
cclog.Debugf("Timer FetchEnergyFootprint %s", time.Since(start))
log.Debugf("Timer FetchEnergyFootprint %s", time.Since(start))
return job.EnergyFootprint, nil
}
@@ -273,9 +280,9 @@ func (r *JobRepository) DeleteJobsBefore(startTime int64) (int, error) {
if err != nil {
s, _, _ := qd.ToSql()
cclog.Errorf(" DeleteJobsBefore(%d) with %s: error %#v", startTime, s, err)
log.Errorf(" DeleteJobsBefore(%d) with %s: error %#v", startTime, s, err)
} else {
cclog.Debugf("DeleteJobsBefore(%d): Deleted %d jobs", startTime, cnt)
log.Debugf("DeleteJobsBefore(%d): Deleted %d jobs", startTime, cnt)
}
return cnt, err
}
@@ -286,9 +293,9 @@ func (r *JobRepository) DeleteJobById(id int64) error {
if err != nil {
s, _, _ := qd.ToSql()
cclog.Errorf("DeleteJobById(%d) with %s : error %#v", id, s, err)
log.Errorf("DeleteJobById(%d) with %s : error %#v", id, s, err)
} else {
cclog.Debugf("DeleteJobById(%d): Success", id)
log.Debugf("DeleteJobById(%d): Success", id)
}
return err
}
@@ -337,10 +344,10 @@ func (r *JobRepository) FindColumnValue(user *schema.User, searchterm string, ta
// theSql, args, theErr := theQuery.ToSql()
// if theErr != nil {
// cclog.Warn("Error while converting query to sql")
// log.Warn("Error while converting query to sql")
// return "", err
// }
// cclog.Debugf("SQL query (FindColumnValue): `%s`, args: %#v", theSql, args)
// log.Debugf("SQL query (FindColumnValue): `%s`, args: %#v", theSql, args)
err := theQuery.RunWith(r.stmtCache).QueryRow().Scan(&result)
@@ -351,7 +358,7 @@ func (r *JobRepository) FindColumnValue(user *schema.User, searchterm string, ta
}
return "", ErrNotFound
} else {
cclog.Infof("Non-Admin User %s : Requested Query '%s' on table '%s' : Forbidden", user.Name, query, table)
log.Infof("Non-Admin User %s : Requested Query '%s' on table '%s' : Forbidden", user.Name, query, table)
return "", ErrForbidden
}
}
@@ -370,7 +377,7 @@ func (r *JobRepository) FindColumnValues(user *schema.User, query string, table
err := rows.Scan(&result)
if err != nil {
rows.Close()
cclog.Warnf("Error while scanning rows: %v", err)
log.Warnf("Error while scanning rows: %v", err)
return emptyResult, err
}
results = append(results, result)
@@ -380,7 +387,7 @@ func (r *JobRepository) FindColumnValues(user *schema.User, query string, table
return emptyResult, ErrNotFound
} else {
cclog.Infof("Non-Admin User %s : Requested Query '%s' on table '%s' : Forbidden", user.Name, query, table)
log.Infof("Non-Admin User %s : Requested Query '%s' on table '%s' : Forbidden", user.Name, query, table)
return emptyResult, ErrForbidden
}
}
@@ -388,7 +395,7 @@ func (r *JobRepository) FindColumnValues(user *schema.User, query string, table
func (r *JobRepository) Partitions(cluster string) ([]string, error) {
var err error
start := time.Now()
partitions := r.cache.Get("partitions:"+cluster, func() (any, time.Duration, int) {
partitions := r.cache.Get("partitions:"+cluster, func() (interface{}, time.Duration, int) {
parts := []string{}
if err = r.DB.Select(&parts, `SELECT DISTINCT job.cluster_partition FROM job WHERE job.cluster = ?;`, cluster); err != nil {
return nil, 0, 1000
@@ -399,7 +406,7 @@ func (r *JobRepository) Partitions(cluster string) ([]string, error) {
if err != nil {
return nil, err
}
cclog.Debugf("Timer Partitions %s", time.Since(start))
log.Debugf("Timer Partitions %s", time.Since(start))
return partitions.([]string), nil
}
@@ -413,7 +420,7 @@ func (r *JobRepository) AllocatedNodes(cluster string) (map[string]map[string]in
Where("job.cluster = ?", cluster).
RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -424,11 +431,11 @@ func (r *JobRepository) AllocatedNodes(cluster string) (map[string]map[string]in
var resources []*schema.Resource
var subcluster string
if err := rows.Scan(&raw, &subcluster); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
if err := json.Unmarshal(raw, &resources); err != nil {
cclog.Warn("Error while unmarshaling raw resources json")
log.Warn("Error while unmarshaling raw resources json")
return nil, err
}
@@ -443,7 +450,7 @@ func (r *JobRepository) AllocatedNodes(cluster string) (map[string]map[string]in
}
}
cclog.Debugf("Timer AllocatedNodes %s", time.Since(start))
log.Debugf("Timer AllocatedNodes %s", time.Since(start))
return subclusters, nil
}
@@ -459,50 +466,23 @@ func (r *JobRepository) StopJobsExceedingWalltimeBy(seconds int) error {
Where(fmt.Sprintf("(%d - job.start_time) > (job.walltime + %d)", time.Now().Unix(), seconds)).
RunWith(r.DB).Exec()
if err != nil {
cclog.Warn("Error while stopping jobs exceeding walltime")
log.Warn("Error while stopping jobs exceeding walltime")
return err
}
rowsAffected, err := res.RowsAffected()
if err != nil {
cclog.Warn("Error while fetching affected rows after stopping due to exceeded walltime")
log.Warn("Error while fetching affected rows after stopping due to exceeded walltime")
return err
}
if rowsAffected > 0 {
cclog.Infof("%d jobs have been marked as failed due to running too long", rowsAffected)
log.Infof("%d jobs have been marked as failed due to running too long", rowsAffected)
}
cclog.Debugf("Timer StopJobsExceedingWalltimeBy %s", time.Since(start))
log.Debugf("Timer StopJobsExceedingWalltimeBy %s", time.Since(start))
return nil
}
func (r *JobRepository) FindJobIdsByTag(tagId int64) ([]int64, error) {
query := sq.Select("job.id").From("job").
Join("jobtag ON jobtag.job_id = job.id").
Where(sq.Eq{"jobtag.tag_id": tagId}).Distinct()
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
return nil, err
}
jobIds := make([]int64, 0, 100)
for rows.Next() {
var jobId int64
if err := rows.Scan(&jobId); err != nil {
rows.Close()
cclog.Warn("Error while scanning rows")
return nil, err
}
jobIds = append(jobIds, jobId)
}
return jobIds, nil
}
// FIXME: Reconsider filtering short jobs with harcoded threshold
func (r *JobRepository) FindRunningJobs(cluster string) ([]*schema.Job, error) {
query := sq.Select(jobColumns...).From("job").
Where(fmt.Sprintf("job.cluster = '%s'", cluster)).
@@ -511,7 +491,7 @@ func (r *JobRepository) FindRunningJobs(cluster string) ([]*schema.Job, error) {
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -520,13 +500,13 @@ func (r *JobRepository) FindRunningJobs(cluster string) ([]*schema.Job, error) {
job, err := scanJob(rows)
if err != nil {
rows.Close()
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
jobs = append(jobs, job)
}
cclog.Infof("Return job count %d", len(jobs))
log.Infof("Return job count %d", len(jobs))
return jobs, nil
}
@@ -551,18 +531,18 @@ func (r *JobRepository) FindJobsBetween(startTimeBegin int64, startTimeEnd int64
}
if startTimeBegin == 0 {
cclog.Infof("Find jobs before %d", startTimeEnd)
log.Infof("Find jobs before %d", startTimeEnd)
query = sq.Select(jobColumns...).From("job").Where(fmt.Sprintf(
"job.start_time < %d", startTimeEnd))
} else {
cclog.Infof("Find jobs between %d and %d", startTimeBegin, startTimeEnd)
log.Infof("Find jobs between %d and %d", startTimeBegin, startTimeEnd)
query = sq.Select(jobColumns...).From("job").Where(fmt.Sprintf(
"job.start_time BETWEEN %d AND %d", startTimeBegin, startTimeEnd))
}
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -571,13 +551,13 @@ func (r *JobRepository) FindJobsBetween(startTimeBegin int64, startTimeEnd int64
job, err := scanJob(rows)
if err != nil {
rows.Close()
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
jobs = append(jobs, job)
}
cclog.Infof("Return job count %d", len(jobs))
log.Infof("Return job count %d", len(jobs))
return jobs, nil
}
@@ -587,7 +567,7 @@ func (r *JobRepository) UpdateMonitoringStatus(job int64, monitoringStatus int32
Where("job.id = ?", job)
_, err = stmt.RunWith(r.stmtCache).Exec()
return err
return
}
func (r *JobRepository) Execute(stmt sq.UpdateBuilder) error {
@@ -607,62 +587,53 @@ func (r *JobRepository) MarkArchived(
func (r *JobRepository) UpdateEnergy(
stmt sq.UpdateBuilder,
jobMeta *schema.Job,
jobMeta *schema.JobMeta,
) (sq.UpdateBuilder, error) {
/* Note: Only Called for Running Jobs during Intermediate Update or on Archiving */
sc, err := archive.GetSubCluster(jobMeta.Cluster, jobMeta.SubCluster)
if err != nil {
cclog.Errorf("cannot get subcluster: %s", err.Error())
log.Errorf("cannot get subcluster: %s", err.Error())
return stmt, err
}
energyFootprint := make(map[string]float64)
var totalEnergy float64
var energy float64
// Total Job Energy Outside Loop
totalEnergy := 0.0
for _, fp := range sc.EnergyFootprint {
// Always Init Metric Energy Inside Loop
metricEnergy := 0.0
if i, err := archive.MetricIndex(sc.MetricConfig, fp); err == nil {
// Note: For DB data, calculate and save as kWh
if sc.MetricConfig[i].Energy == "energy" { // this metric has energy as unit (Joules or Wh)
cclog.Warnf("Update EnergyFootprint for Job %d and Metric %s on cluster %s: Set to 'energy' in cluster.json: Not implemented, will return 0.0", jobMeta.JobID, jobMeta.Cluster, fp)
// FIXME: Needs sum as stats type
} else if sc.MetricConfig[i].Energy == "power" { // this metric has power as unit (Watt)
// Energy: Power (in Watts) * Time (in Seconds)
// Unit: (W * (s / 3600)) / 1000 = kWh
// Round 2 Digits: round(Energy * 100) / 100
// Here: (All-Node Metric Average * Number of Nodes) * (Job Duration in Seconds / 3600) / 1000
// Note: Shared Jobs handled correctly since "Node Average" is based on partial resources, while "numNodes" factor is 1
rawEnergy := ((LoadJobStat(jobMeta, fp, "avg") * float64(jobMeta.NumNodes)) * (float64(jobMeta.Duration) / 3600.0)) / 1000.0
metricEnergy = math.Round(rawEnergy*100.0) / 100.0
// Unit: ( W * s ) / 3600 / 1000 = kWh ; Rounded to 2 nearest digits
energy = math.Round(((LoadJobStat(jobMeta, fp, "avg")*float64(jobMeta.Duration))/3600/1000)*100) / 100
}
} else {
cclog.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, jobMeta.ID)
log.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, jobMeta.ID)
}
energyFootprint[fp] = metricEnergy
totalEnergy += metricEnergy
// cclog.Infof("Metric %s Average %f -> %f kWh | Job %d Total -> %f kWh", fp, LoadJobStat(jobMeta, fp, "avg"), energy, jobMeta.JobID, totalEnergy)
energyFootprint[fp] = energy
totalEnergy += energy
}
var rawFootprint []byte
if rawFootprint, err = json.Marshal(energyFootprint); err != nil {
cclog.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
log.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
return stmt, err
}
return stmt.Set("energy_footprint", string(rawFootprint)).Set("energy", (math.Round(totalEnergy*100.0) / 100.0)), nil
return stmt.Set("energy_footprint", string(rawFootprint)).Set("energy", (math.Round(totalEnergy*100) / 100)), nil
}
func (r *JobRepository) UpdateFootprint(
stmt sq.UpdateBuilder,
jobMeta *schema.Job,
jobMeta *schema.JobMeta,
) (sq.UpdateBuilder, error) {
/* Note: Only Called for Running Jobs during Intermediate Update or on Archiving */
sc, err := archive.GetSubCluster(jobMeta.Cluster, jobMeta.SubCluster)
if err != nil {
cclog.Errorf("cannot get subcluster: %s", err.Error())
log.Errorf("cannot get subcluster: %s", err.Error())
return stmt, err
}
footprint := make(map[string]float64)
@@ -676,7 +647,7 @@ func (r *JobRepository) UpdateFootprint(
}
if statType != "avg" && statType != "min" && statType != "max" {
cclog.Warnf("unknown statType for footprint update: %s", statType)
log.Warnf("unknown statType for footprint update: %s", statType)
return stmt, fmt.Errorf("unknown statType for footprint update: %s", statType)
}
@@ -690,7 +661,7 @@ func (r *JobRepository) UpdateFootprint(
var rawFootprint []byte
if rawFootprint, err = json.Marshal(footprint); err != nil {
cclog.Warnf("Error while marshaling footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
log.Warnf("Error while marshaling footprint for job INTO BYTES, DB ID '%v'", jobMeta.ID)
return stmt, err
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -8,86 +8,37 @@ import (
"encoding/json"
"fmt"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
const NamedJobCacheInsert string = `INSERT INTO job_cache (
job_id, hpc_user, project, cluster, subcluster, cluster_partition, array_job_id, num_nodes, num_hwthreads, num_acc,
shared, monitoring_status, smt, job_state, start_time, duration, walltime, footprint, energy, energy_footprint, resources, meta_data
) VALUES (
:job_id, :hpc_user, :project, :cluster, :subcluster, :cluster_partition, :array_job_id, :num_nodes, :num_hwthreads, :num_acc,
:shared, :monitoring_status, :smt, :job_state, :start_time, :duration, :walltime, :footprint, :energy, :energy_footprint, :resources, :meta_data
);`
const NamedJobInsert string = `INSERT INTO job (
job_id, hpc_user, project, cluster, subcluster, cluster_partition, array_job_id, num_nodes, num_hwthreads, num_acc,
shared, monitoring_status, smt, job_state, start_time, duration, walltime, footprint, energy, energy_footprint, resources, meta_data
exclusive, monitoring_status, smt, job_state, start_time, duration, walltime, footprint, energy, energy_footprint, resources, meta_data
) VALUES (
:job_id, :hpc_user, :project, :cluster, :subcluster, :cluster_partition, :array_job_id, :num_nodes, :num_hwthreads, :num_acc,
:shared, :monitoring_status, :smt, :job_state, :start_time, :duration, :walltime, :footprint, :energy, :energy_footprint, :resources, :meta_data
:exclusive, :monitoring_status, :smt, :job_state, :start_time, :duration, :walltime, :footprint, :energy, :energy_footprint, :resources, :meta_data
);`
func (r *JobRepository) InsertJob(job *schema.Job) (int64, error) {
r.Mutex.Lock()
res, err := r.DB.NamedExec(NamedJobCacheInsert, job)
r.Mutex.Unlock()
func (r *JobRepository) InsertJob(job *schema.JobMeta) (int64, error) {
res, err := r.DB.NamedExec(NamedJobInsert, job)
if err != nil {
cclog.Warn("Error while NamedJobInsert")
log.Warn("Error while NamedJobInsert")
return 0, err
}
id, err := res.LastInsertId()
if err != nil {
cclog.Warn("Error while getting last insert ID")
log.Warn("Error while getting last insert ID")
return 0, err
}
return id, nil
}
func (r *JobRepository) SyncJobs() ([]*schema.Job, error) {
r.Mutex.Lock()
defer r.Mutex.Unlock()
query := sq.Select(jobCacheColumns...).From("job_cache")
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Errorf("Error while running query %v", err)
return nil, err
}
jobs := make([]*schema.Job, 0, 50)
for rows.Next() {
job, err := scanJob(rows)
if err != nil {
rows.Close()
cclog.Warn("Error while scanning rows")
return nil, err
}
jobs = append(jobs, job)
}
_, err = r.DB.Exec(
"INSERT INTO job (job_id, cluster, subcluster, start_time, hpc_user, project, cluster_partition, array_job_id, num_nodes, num_hwthreads, num_acc, shared, monitoring_status, smt, job_state, duration, walltime, footprint, energy, energy_footprint, resources, meta_data) SELECT job_id, cluster, subcluster, start_time, hpc_user, project, cluster_partition, array_job_id, num_nodes, num_hwthreads, num_acc, shared, monitoring_status, smt, job_state, duration, walltime, footprint, energy, energy_footprint, resources, meta_data FROM job_cache")
if err != nil {
cclog.Warnf("Error while Job sync: %v", err)
return nil, err
}
_, err = r.DB.Exec("DELETE FROM job_cache")
if err != nil {
cclog.Warnf("Error while Job cache clean: %v", err)
return nil, err
}
return jobs, nil
}
// Start inserts a new job in the table, returning the unique job ID.
// Statistics are not transfered!
func (r *JobRepository) Start(job *schema.Job) (id int64, err error) {
func (r *JobRepository) Start(job *schema.JobMeta) (id int64, err error) {
job.RawFootprint, err = json.Marshal(job.Footprint)
if err != nil {
return -1, fmt.Errorf("REPOSITORY/JOB > encoding footprint field failed: %w", err)
@@ -120,21 +71,5 @@ func (r *JobRepository) Stop(
Where("job.id = ?", jobId)
_, err = stmt.RunWith(r.stmtCache).Exec()
return err
}
func (r *JobRepository) StopCached(
jobId int64,
duration int32,
state schema.JobState,
monitoringStatus int32,
) (err error) {
stmt := sq.Update("job_cache").
Set("job_state", state).
Set("duration", duration).
Set("monitoring_status", monitoringStatus).
Where("job.id = ?", jobId)
_, err = stmt.RunWith(r.stmtCache).Exec()
return err
return
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -11,8 +11,8 @@ import (
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
@@ -39,27 +39,7 @@ func (r *JobRepository) Find(
q = q.OrderBy("job.id DESC") // always use newest matching job by db id if more than one match
cclog.Debugf("Timer Find %s", time.Since(start))
return scanJob(q.RunWith(r.stmtCache).QueryRow())
}
func (r *JobRepository) FindCached(
jobId *int64,
cluster *string,
startTime *int64,
) (*schema.Job, error) {
q := sq.Select(jobCacheColumns...).From("job_cache").
Where("job_cache.job_id = ?", *jobId)
if cluster != nil {
q = q.Where("job_cache.cluster = ?", *cluster)
}
if startTime != nil {
q = q.Where("job_cache.start_time = ?", *startTime)
}
q = q.OrderBy("job_cache.id DESC") // always use newest matching job by db id if more than one match
log.Debugf("Timer Find %s", time.Since(start))
return scanJob(q.RunWith(r.stmtCache).QueryRow())
}
@@ -86,7 +66,7 @@ func (r *JobRepository) FindAll(
rows, err := q.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -94,44 +74,15 @@ func (r *JobRepository) FindAll(
for rows.Next() {
job, err := scanJob(rows)
if err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
jobs = append(jobs, job)
}
cclog.Debugf("Timer FindAll %s", time.Since(start))
log.Debugf("Timer FindAll %s", time.Since(start))
return jobs, nil
}
// Get complete joblist only consisting of db ids.
// This is useful to process large job counts and intended to be used
// together with FindById to process jobs one by one
func (r *JobRepository) GetJobList() ([]int64, error) {
query := sq.Select("id").From("job").
Where("job.job_state != 'running'")
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Error("Error while running query")
return nil, err
}
jl := make([]int64, 0, 1000)
for rows.Next() {
var id int64
err := rows.Scan(&id)
if err != nil {
rows.Close()
cclog.Warn("Error while scanning rows")
return nil, err
}
jl = append(jl, id)
}
cclog.Infof("Return job count %d", len(jl))
return jl, nil
}
// FindById executes a SQL query to find a specific batch job.
// The job is queried using the database id.
// It returns a pointer to a schema.Job data structure and an error variable.
@@ -227,7 +178,7 @@ func (r *JobRepository) FindConcurrentJobs(
var startTime int64
var stopTime int64
startTime = job.StartTime
startTime = job.StartTimeUnix
hostname := job.Resources[0].Hostname
if job.State == schema.JobStateRunning {
@@ -243,17 +194,15 @@ func (r *JobRepository) FindConcurrentJobs(
queryRunning := query.Where("job.job_state = ?").Where("(job.start_time BETWEEN ? AND ? OR job.start_time < ?)",
"running", startTimeTail, stopTimeTail, startTime)
// Get At Least One Exact Hostname Match from JSON Resources Array in Database
queryRunning = queryRunning.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, '$.hostname') = ?)", hostname)
queryRunning = queryRunning.Where("job.resources LIKE ?", fmt.Sprint("%", hostname, "%"))
query = query.Where("job.job_state != ?").Where("((job.start_time BETWEEN ? AND ?) OR (job.start_time + job.duration) BETWEEN ? AND ? OR (job.start_time < ?) AND (job.start_time + job.duration) > ?)",
"running", startTimeTail, stopTimeTail, startTimeFront, stopTimeTail, startTime, stopTime)
// Get At Least One Exact Hostname Match from JSON Resources Array in Database
query = query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, '$.hostname') = ?)", hostname)
query = query.Where("job.resources LIKE ?", fmt.Sprint("%", hostname, "%"))
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Errorf("Error while running query: %v", err)
log.Errorf("Error while running query: %v", err)
return nil, err
}
@@ -264,7 +213,7 @@ func (r *JobRepository) FindConcurrentJobs(
var id, jobId, startTime sql.NullInt64
if err = rows.Scan(&id, &jobId, &startTime); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
@@ -280,7 +229,7 @@ func (r *JobRepository) FindConcurrentJobs(
rows, err = queryRunning.RunWith(r.stmtCache).Query()
if err != nil {
cclog.Errorf("Error while running query: %v", err)
log.Errorf("Error while running query: %v", err)
return nil, err
}
@@ -288,7 +237,7 @@ func (r *JobRepository) FindConcurrentJobs(
var id, jobId, startTime sql.NullInt64
if err := rows.Scan(&id, &jobId, &startTime); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}

View File

@@ -1,57 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
"sync"
"github.com/ClusterCockpit/cc-lib/schema"
)
type JobHook interface {
JobStartCallback(job *schema.Job)
JobStopCallback(job *schema.Job)
}
var (
initOnce sync.Once
hooks []JobHook
)
func RegisterJobJook(hook JobHook) {
initOnce.Do(func() {
hooks = make([]JobHook, 0)
})
if hook != nil {
hooks = append(hooks, hook)
}
}
func CallJobStartHooks(jobs []*schema.Job) {
if hooks == nil {
return
}
for _, hook := range hooks {
if hook != nil {
for _, job := range jobs {
hook.JobStartCallback(job)
}
}
}
}
func CallJobStopHooks(job *schema.Job) {
if hooks == nil {
return
}
for _, hook := range hooks {
if hook != nil {
hook.JobStopCallback(job)
}
}
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -12,10 +12,9 @@ import (
"strings"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
@@ -68,8 +67,7 @@ func (r *JobRepository) QueryJobs(
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
queryString, queryVars, _ := query.ToSql()
cclog.Errorf("Error while running query '%s' %v: %v", queryString, queryVars, err)
log.Errorf("Error while running query: %v", err)
return nil, err
}
@@ -78,7 +76,7 @@ func (r *JobRepository) QueryJobs(
job, err := scanJob(rows)
if err != nil {
rows.Close()
cclog.Warn("Error while scanning rows (Jobs)")
log.Warn("Error while scanning rows (Jobs)")
return nil, err
}
jobs = append(jobs, job)
@@ -124,7 +122,7 @@ func SecurityCheckWithUser(user *schema.User, query sq.SelectBuilder) (sq.Select
if len(user.Projects) != 0 {
return query.Where(sq.Or{sq.Eq{"job.project": user.Projects}, sq.Eq{"job.hpc_user": user.Username}}), nil
} else {
cclog.Debugf("Manager-User '%s' has no defined projects to lookup! Query only personal jobs ...", user.Username)
log.Debugf("Manager-User '%s' has no defined projects to lookup! Query only personal jobs ...", user.Username)
return query.Where("job.hpc_user = ?", user.Username), nil
}
case user.HasRole(schema.RoleUser): // User : Only personal jobs
@@ -147,11 +145,6 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
// This is an OR-Logic query: Returns all distinct jobs with at least one of the requested tags; TODO: AND-Logic query?
query = query.Join("jobtag ON jobtag.job_id = job.id").Where(sq.Eq{"jobtag.tag_id": filter.Tags}).Distinct()
}
if filter.DbID != nil {
dbIDs := make([]string, len(filter.DbID))
copy(dbIDs, filter.DbID)
query = query.Where(sq.Eq{"job.id": dbIDs})
}
if filter.JobID != nil {
query = buildStringCondition("job.job_id", filter.JobID, query)
}
@@ -183,9 +176,6 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
now := time.Now().Unix() // There does not seam to be a portable way to get the current unix timestamp accross different DBs.
query = query.Where("(job.job_state != 'running' OR (? - job.start_time) > ?)", now, *filter.MinRunningFor)
}
if filter.Shared != nil {
query = query.Where("job.shared = ?", *filter.Shared)
}
if filter.State != nil {
states := make([]string, len(filter.State))
for i, val := range filter.State {
@@ -204,7 +194,7 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
query = buildIntCondition("job.num_hwthreads", filter.NumHWThreads, query)
}
if filter.Node != nil {
query = buildResourceJsonCondition("hostname", filter.Node, query)
query = buildStringCondition("job.resources", filter.Node, query)
}
if filter.Energy != nil {
query = buildFloatCondition("job.energy", filter.Energy, query)
@@ -217,7 +207,7 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
return query
}
func buildIntCondition(field string, cond *config.IntRange, query sq.SelectBuilder) sq.SelectBuilder {
func buildIntCondition(field string, cond *schema.IntRange, query sq.SelectBuilder) sq.SelectBuilder {
return query.Where(field+" BETWEEN ? AND ?", cond.From, cond.To)
}
@@ -225,7 +215,7 @@ func buildFloatCondition(field string, cond *model.FloatRange, query sq.SelectBu
return query.Where(field+" BETWEEN ? AND ?", cond.From, cond.To)
}
func buildTimeCondition(field string, cond *config.TimeRange, query sq.SelectBuilder) sq.SelectBuilder {
func buildTimeCondition(field string, cond *schema.TimeRange, query sq.SelectBuilder) sq.SelectBuilder {
if cond.From != nil && cond.To != nil {
return query.Where(field+" BETWEEN ? AND ?", cond.From.Unix(), cond.To.Unix())
} else if cond.From != nil {
@@ -245,7 +235,7 @@ func buildTimeCondition(field string, cond *config.TimeRange, query sq.SelectBui
case "last30d":
then = now - (60 * 60 * 24 * 30)
default:
cclog.Debugf("No known named timeRange: startTime.range = %s", cond.Range)
log.Debugf("No known named timeRange: startTime.range = %s", cond.Range)
return query
}
return query.Where(field+" BETWEEN ? AND ?", then, now)
@@ -306,28 +296,6 @@ func buildMetaJsonCondition(jsonField string, cond *model.StringInput, query sq.
return query
}
func buildResourceJsonCondition(jsonField string, cond *model.StringInput, query sq.SelectBuilder) sq.SelectBuilder {
// Verify and Search Only in Valid Jsons
query = query.Where("JSON_VALID(resources)")
// add "AND" Sql query Block for field match
if cond.Eq != nil {
return query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, \"$."+jsonField+"\") = ?)", *cond.Eq)
}
if cond.Neq != nil { // Currently Unused
return query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, \"$."+jsonField+"\") != ?)", *cond.Neq)
}
if cond.StartsWith != nil { // Currently Unused
return query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, \"$."+jsonField+"\")) LIKE ?)", fmt.Sprint(*cond.StartsWith, "%"))
}
if cond.EndsWith != nil { // Currently Unused
return query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, \"$."+jsonField+"\") LIKE ?)", fmt.Sprint("%", *cond.EndsWith))
}
if cond.Contains != nil {
return query.Where("EXISTS (SELECT 1 FROM json_each(job.resources) WHERE json_extract(value, \"$."+jsonField+"\") LIKE ?)", fmt.Sprint("%", *cond.Contains, "%"))
}
return query
}
var (
matchFirstCap = regexp.MustCompile("(.)([A-Z][a-z]+)")
matchAllCap = regexp.MustCompile("([a-z0-9])([A-Z])")
@@ -336,7 +304,7 @@ var (
func toSnakeCase(str string) string {
for _, c := range str {
if c == '\'' || c == '\\' {
cclog.Panic("toSnakeCase() attack vector!")
log.Panic("toSnakeCase() attack vector!")
}
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -9,38 +9,38 @@ import (
"fmt"
"testing"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
_ "github.com/mattn/go-sqlite3"
)
func TestFind(t *testing.T) {
r := setup(t)
jobID, cluster, startTime := int64(398800), "fritz", int64(1675954712)
job, err := r.Find(&jobID, &cluster, &startTime)
jobId, cluster, startTime := int64(398998), "fritz", int64(1675957496)
job, err := r.Find(&jobId, &cluster, &startTime)
if err != nil {
t.Fatal(err)
}
// fmt.Printf("%+v", job)
if *job.ID != 345 {
t.Errorf("wrong summary for diagnostic \ngot: %d \nwant: 345", job.JobID)
if job.ID != 5 {
t.Errorf("wrong summary for diagnostic 3\ngot: %d \nwant: 1366", job.JobID)
}
}
func TestFindById(t *testing.T) {
r := setup(t)
job, err := r.FindById(getContext(t), 338)
job, err := r.FindById(getContext(t), 5)
if err != nil {
t.Fatal(err)
}
// fmt.Printf("%+v", job)
if job.JobID != 398793 {
t.Errorf("wrong summary for diagnostic \ngot: %d \nwant: 1404396", job.JobID)
if job.JobID != 398998 {
t.Errorf("wrong summary for diagnostic 3\ngot: %d \nwant: 1404396", job.JobID)
}
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
@@ -10,14 +9,14 @@ import (
"embed"
"fmt"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/golang-migrate/migrate/v4"
"github.com/golang-migrate/migrate/v4/database/mysql"
"github.com/golang-migrate/migrate/v4/database/sqlite3"
"github.com/golang-migrate/migrate/v4/source/iofs"
)
const Version uint = 10
const Version uint = 8
//go:embed migrations/*
var migrationFiles embed.FS
@@ -55,13 +54,13 @@ func checkDBVersion(backend string, db *sql.DB) error {
return err
}
default:
cclog.Abortf("Migration: Unsupported database backend '%s'.\n", backend)
log.Fatalf("unsupported database backend: %s", backend)
}
v, dirty, err := m.Version()
if err != nil {
if err == migrate.ErrNilVersion {
cclog.Warn("Legacy database without version or missing database file!")
log.Warn("Legacy database without version or missing database file!")
} else {
return err
}
@@ -85,7 +84,7 @@ func getMigrateInstance(backend string, db string) (m *migrate.Migrate, err erro
case "sqlite3":
d, err := iofs.New(migrationFiles, "migrations/sqlite3")
if err != nil {
cclog.Fatal(err)
log.Fatal(err)
}
m, err = migrate.NewWithSourceInstance("iofs", d, fmt.Sprintf("sqlite3://%s?_foreign_keys=on", db))
@@ -103,7 +102,7 @@ func getMigrateInstance(backend string, db string) (m *migrate.Migrate, err erro
return m, err
}
default:
cclog.Abortf("Migration: Unsupported database backend '%s'.\n", backend)
log.Fatalf("unsupported database backend: %s", backend)
}
return m, nil
@@ -116,17 +115,8 @@ func MigrateDB(backend string, db string) error {
}
v, dirty, err := m.Version()
if err != nil {
if err == migrate.ErrNilVersion {
cclog.Warn("Legacy database without version or missing database file!")
} else {
return err
}
}
if v < Version {
cclog.Infof("unsupported database version %d, need %d.\nPlease backup your database file and run cc-backend -migrate-db", v, Version)
}
log.Infof("unsupported database version %d, need %d.\nPlease backup your database file and run cc-backend -migrate-db", v, Version)
if dirty {
return fmt.Errorf("last migration to version %d has failed, please fix the db manually and force version with -force-db flag", Version)
@@ -134,7 +124,7 @@ func MigrateDB(backend string, db string) error {
if err := m.Up(); err != nil {
if err == migrate.ErrNoChange {
cclog.Info("DB already up to date!")
log.Info("DB already up to date!")
} else {
return err
}
@@ -152,7 +142,7 @@ func RevertDB(backend string, db string) error {
if err := m.Migrate(Version - 1); err != nil {
if err == migrate.ErrNoChange {
cclog.Info("DB already up to date!")
log.Info("DB already up to date!")
} else {
return err
}

View File

@@ -1 +0,0 @@
DROP TABLE IF EXISTS job_cache;

View File

@@ -1,119 +0,0 @@
CREATE TABLE "job_cache" (
id INTEGER PRIMARY KEY,
job_id BIGINT NOT NULL,
cluster VARCHAR(255) NOT NULL,
subcluster VARCHAR(255) NOT NULL,
submit_time BIGINT NOT NULL DEFAULT 0, -- Unix timestamp
start_time BIGINT NOT NULL DEFAULT 0, -- Unix timestamp
hpc_user VARCHAR(255) NOT NULL,
project VARCHAR(255) NOT NULL,
cluster_partition VARCHAR(255),
array_job_id BIGINT,
duration INT NOT NULL,
walltime INT NOT NULL,
job_state VARCHAR(255) NOT NULL
CHECK (job_state IN (
'boot_fail', 'cancelled', 'completed', 'deadline',
'failed', 'node_fail', 'out-of-memory', 'pending',
'preempted', 'running', 'suspended', 'timeout'
)),
meta_data TEXT, -- JSON
resources TEXT NOT NULL, -- JSON
num_nodes INT NOT NULL,
num_hwthreads INT,
num_acc INT,
smt TINYINT NOT NULL DEFAULT 1 CHECK (smt IN (0, 1)),
shared TEXT NOT NULL
CHECK (shared IN ("none", "single_user", "multi_user")),
monitoring_status TINYINT NOT NULL DEFAULT 1
CHECK (monitoring_status IN (0, 1, 2, 3)),
energy REAL NOT NULL DEFAULT 0.0,
energy_footprint TEXT DEFAULT NULL,
footprint TEXT DEFAULT NULL,
UNIQUE (job_id, cluster, start_time)
);
CREATE TABLE "job_new" (
id INTEGER PRIMARY KEY,
job_id BIGINT NOT NULL,
cluster TEXT NOT NULL,
subcluster TEXT NOT NULL,
submit_time BIGINT NOT NULL DEFAULT 0, -- Unix timestamp
start_time BIGINT NOT NULL DEFAULT 0, -- Unix timestamp
hpc_user TEXT NOT NULL,
project TEXT NOT NULL,
cluster_partition TEXT,
array_job_id BIGINT,
duration INT NOT NULL,
walltime INT NOT NULL,
job_state TEXT NOT NULL
CHECK (job_state IN (
'boot_fail', 'cancelled', 'completed', 'deadline',
'failed', 'node_fail', 'out-of-memory', 'pending',
'preempted', 'running', 'suspended', 'timeout'
)),
meta_data TEXT, -- JSON
resources TEXT NOT NULL, -- JSON
num_nodes INT NOT NULL,
num_hwthreads INT,
num_acc INT,
smt INT NOT NULL DEFAULT 1,
shared TEXT NOT NULL
CHECK (shared IN ("none", "single_user", "multi_user")),
monitoring_status TINYINT NOT NULL DEFAULT 1
CHECK (monitoring_status IN (0, 1, 2, 3)),
energy REAL NOT NULL DEFAULT 0.0,
energy_footprint TEXT DEFAULT NULL,
footprint TEXT DEFAULT NULL,
UNIQUE (job_id, cluster, start_time)
);
CREATE TABLE IF NOT EXISTS lookup_exclusive (
id INTEGER PRIMARY KEY,
name TEXT NOT NULL UNIQUE
);
INSERT INTO lookup_exclusive (id, name) VALUES
(0, 'multi_user'),
(1, 'none'),
(2, 'single_user');
INSERT INTO job_new (
id, job_id, cluster, subcluster, submit_time, start_time, hpc_user, project,
cluster_partition, array_job_id, duration, walltime, job_state, meta_data, resources,
num_nodes, num_hwthreads, num_acc, smt, shared, monitoring_status, energy,
energy_footprint, footprint
) SELECT
id,
job_id,
cluster,
subcluster,
0,
start_time,
hpc_user,
project,
cluster_partition,
array_job_id,
duration,
walltime,
job_state,
meta_data,
resources,
num_nodes,
num_hwthreads,
num_acc,
smt,
(
SELECT name FROM lookup_exclusive
WHERE id = job.exclusive
),
monitoring_status,
energy,
energy_footprint,
footprint
FROM job;
DROP TABLE lookup_exclusive;
DROP TABLE job;
ALTER TABLE job_new RENAME TO job;

View File

@@ -1 +0,0 @@
DROP TABLE IF EXISTS node;

View File

@@ -1,50 +0,0 @@
CREATE TABLE "node" (
id INTEGER PRIMARY KEY,
hostname VARCHAR(255) NOT NULL,
cluster VARCHAR(255) NOT NULL,
subcluster VARCHAR(255) NOT NULL,
meta_data TEXT, -- JSON
UNIQUE (hostname, cluster)
);
CREATE TABLE "node_state" (
id INTEGER PRIMARY KEY,
time_stamp INTEGER NOT NULL,
jobs_running INTEGER DEFAULT 0 NOT NULL,
cpus_allocated INTEGER DEFAULT 0 NOT NULL,
memory_allocated INTEGER DEFAULT 0 NOT NULL,
gpus_allocated INTEGER DEFAULT 0 NOT NULL,
node_state VARCHAR(255) NOT NULL
CHECK (node_state IN (
'allocated', 'reserved', 'idle', 'mixed',
'down', 'unknown'
)),
health_state VARCHAR(255) NOT NULL
CHECK (health_state IN (
'full', 'partial', 'failed'
)),
node_id INTEGER,
FOREIGN KEY (node_id) REFERENCES node (id)
);
-- Add NEW Indices For New Job Table Columns
CREATE INDEX IF NOT EXISTS jobs_cluster_submittime ON job (cluster, submit_time);
CREATE INDEX IF NOT EXISTS jobs_cluster_partition_submittime ON job (cluster, cluster_partition, submit_time);
CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_submittime ON job (
cluster, cluster_partition, job_state, submit_time
);
CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_submittime ON job (cluster, job_state, submit_time);
-- Add NEW Indices For New Node Table VARCHAR Fields
CREATE INDEX IF NOT EXISTS nodes_cluster ON node (cluster);
CREATE INDEX IF NOT EXISTS nodes_cluster_subcluster ON node (cluster, subcluster);
-- Add NEW Indices For New Node_State Table Fields
CREATE INDEX IF NOT EXISTS nodestates_state ON node_state (node_state);
CREATE INDEX IF NOT EXISTS nodestates_health ON node_state (health_state);
CREATE INDEX IF NOT EXISTS nodestates_nodeid_state ON node_state (node_id, node_state);
CREATE INDEX IF NOT EXISTS nodestates_nodeid_health ON node_state (node_id, health_state);
-- Add NEW Indices For Increased Amounts of Tags
CREATE INDEX IF NOT EXISTS tags_jobid ON jobtag (job_id);
CREATE INDEX IF NOT EXISTS tags_tagid ON jobtag (tag_id);

View File

@@ -1,498 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
"context"
"database/sql"
"encoding/json"
"fmt"
"sync"
"time"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/lrucache"
"github.com/ClusterCockpit/cc-lib/schema"
sq "github.com/Masterminds/squirrel"
"github.com/jmoiron/sqlx"
)
var (
nodeRepoOnce sync.Once
nodeRepoInstance *NodeRepository
)
type NodeRepository struct {
DB *sqlx.DB
stmtCache *sq.StmtCache
cache *lrucache.Cache
driver string
}
func GetNodeRepository() *NodeRepository {
nodeRepoOnce.Do(func() {
db := GetConnection()
nodeRepoInstance = &NodeRepository{
DB: db.DB,
driver: db.Driver,
stmtCache: sq.NewStmtCache(db.DB),
cache: lrucache.New(1024 * 1024),
}
})
return nodeRepoInstance
}
func (r *NodeRepository) FetchMetadata(hostname string, cluster string) (map[string]string, error) {
start := time.Now()
RawMetaData := make([]byte, 0)
if err := sq.Select("node.meta_data").From("node").
Where("node.hostname = ?", hostname).
Where("node.cluster = ?", cluster).
RunWith(r.stmtCache).QueryRow().Scan(&RawMetaData); err != nil {
cclog.Warn("Error while scanning for node metadata")
return nil, err
}
if len(RawMetaData) == 0 {
return nil, nil
}
MetaData := make(map[string]string)
if err := json.Unmarshal(RawMetaData, &MetaData); err != nil {
cclog.Warn("Error while unmarshaling raw metadata json")
return nil, err
}
cclog.Debugf("Timer FetchMetadata %s", time.Since(start))
return MetaData, nil
}
//
// func (r *NodeRepository) UpdateMetadata(node *schema.Node, key, val string) (err error) {
// cachekey := fmt.Sprintf("metadata:%d", node.ID)
// r.cache.Del(cachekey)
// if node.MetaData == nil {
// if _, err = r.FetchMetadata(node); err != nil {
// cclog.Warnf("Error while fetching metadata for node, DB ID '%v'", node.ID)
// return err
// }
// }
//
// if node.MetaData != nil {
// cpy := make(map[string]string, len(node.MetaData)+1)
// maps.Copy(cpy, node.MetaData)
// cpy[key] = val
// node.MetaData = cpy
// } else {
// node.MetaData = map[string]string{key: val}
// }
//
// if node.RawMetaData, err = json.Marshal(node.MetaData); err != nil {
// cclog.Warnf("Error while marshaling metadata for node, DB ID '%v'", node.ID)
// return err
// }
//
// if _, err = sq.Update("node").
// Set("meta_data", node.RawMetaData).
// Where("node.id = ?", node.ID).
// RunWith(r.stmtCache).Exec(); err != nil {
// cclog.Warnf("Error while updating metadata for node, DB ID '%v'", node.ID)
// return err
// }
//
// r.cache.Put(cachekey, node.MetaData, len(node.RawMetaData), 24*time.Hour)
// return nil
// }
func (r *NodeRepository) GetNode(hostname string, cluster string, withMeta bool) (*schema.Node, error) {
node := &schema.Node{}
if err := sq.Select("node.hostname", "node.cluster", "node.subcluster", "node_state.node_state",
"node_state.health_state", "MAX(node_state.time_stamp) as time").
From("node_state").
Join("node ON nodes_state.node_id = node.id").
Where("node.hostname = ?", hostname).
Where("node.cluster = ?", cluster).
GroupBy("node_state.node_id").
RunWith(r.DB).
QueryRow().Scan(&node.Hostname, &node.Cluster, &node.SubCluster, &node.NodeState, &node.HealthState); err != nil {
cclog.Warnf("Error while querying node '%s' from database: %v", hostname, err)
return nil, err
}
if withMeta {
var err error
var meta map[string]string
if meta, err = r.FetchMetadata(hostname, cluster); err != nil {
cclog.Warnf("Error while fetching metadata for node '%s'", hostname)
return nil, err
}
node.MetaData = meta
}
return node, nil
}
func (r *NodeRepository) GetNodeById(id int64, withMeta bool) (*schema.Node, error) {
node := &schema.Node{}
if err := sq.Select("node.hostname", "node.cluster", "node.subcluster", "node_state.node_state",
"node_state.health_state", "MAX(node_state.time_stamp) as time").
From("node_state").
Join("node ON nodes_state.node_id = node.id").
Where("node.id = ?", id).
GroupBy("node_state.node_id").
RunWith(r.DB).
QueryRow().Scan(&node.Hostname, &node.Cluster, &node.SubCluster, &node.NodeState, &node.HealthState); err != nil {
cclog.Warnf("Error while querying node ID '%d' from database: %v", id, err)
return nil, err
}
// NEEDS METADATA BY ID
// if withMeta {
// var err error
// var meta map[string]string
// if meta, err = r.FetchMetadata(hostname, cluster); err != nil {
// cclog.Warnf("Error while fetching metadata for node '%s'", hostname)
// return nil, err
// }
// node.MetaData = meta
// }
return node, nil
}
// const NamedNodeInsert string = `
// INSERT INTO node (time_stamp, hostname, cluster, subcluster, node_state, health_state,
//
// cpus_allocated, cpus_total, memory_allocated, memory_total, gpus_allocated, gpus_total)
// VALUES (:time_stamp, :hostname, :cluster, :subcluster, :node_state, :health_state,
// :cpus_allocated, :cpus_total, :memory_allocated, :memory_total, :gpus_allocated, :gpus_total);`
const NamedNodeInsert string = `
INSERT INTO node (hostname, cluster, subcluster)
VALUES (:hostname, :cluster, :subcluster);`
// AddNode adds a Node to the node table. This can be triggered by a node collector registration or
// from a nodestate update from the job scheduler.
func (r *NodeRepository) AddNode(node *schema.NodeDB) (int64, error) {
var err error
res, err := r.DB.NamedExec(NamedNodeInsert, node)
if err != nil {
cclog.Errorf("Error while adding node '%v' to database", node.Hostname)
return 0, err
}
node.ID, err = res.LastInsertId()
if err != nil {
cclog.Errorf("Error while getting last insert id for node '%v' from database", node.Hostname)
return 0, err
}
return node.ID, nil
}
const NamedNodeStateInsert string = `
INSERT INTO node_state (time_stamp, node_state, health_state, cpus_allocated,
memory_allocated, gpus_allocated, jobs_running, node_id)
VALUES (:time_stamp, :node_state, :health_state, :cpus_allocated, :memory_allocated, :gpus_allocated, :jobs_running, :node_id);`
// TODO: Add real Monitoring Health State
// UpdateNodeState is called from the Node REST API to add a row in the node state table
func (r *NodeRepository) UpdateNodeState(hostname string, cluster string, nodeState *schema.NodeStateDB) error {
var id int64
if err := sq.Select("id").From("node").
Where("node.hostname = ?", hostname).Where("node.cluster = ?", cluster).RunWith(r.DB).
QueryRow().Scan(&id); err != nil {
if err == sql.ErrNoRows {
subcluster, err := archive.GetSubClusterByNode(cluster, hostname)
if err != nil {
cclog.Errorf("Error while getting subcluster for node '%s' in cluster '%s': %v", hostname, cluster, err)
return err
}
node := schema.NodeDB{
Hostname: hostname, Cluster: cluster, SubCluster: subcluster,
}
id, err = r.AddNode(&node)
if err != nil {
cclog.Errorf("Error while adding node '%s' to database: %v", hostname, err)
return err
}
cclog.Infof("Added node '%s' to database", hostname)
return nil
} else {
cclog.Warnf("Error while querying node '%v' from database", id)
return err
}
}
nodeState.NodeID = id
_, err := r.DB.NamedExec(NamedNodeStateInsert, nodeState)
if err != nil {
cclog.Errorf("Error while adding node state for '%v' to database", hostname)
return err
}
cclog.Infof("Updated node state for '%s' in database", hostname)
return nil
}
// func (r *NodeRepository) UpdateHealthState(hostname string, healthState *schema.MonitoringState) error {
// if _, err := sq.Update("node").Set("health_state", healthState).Where("node.id = ?", id).RunWith(r.DB).Exec(); err != nil {
// cclog.Errorf("error while updating node '%d'", id)
// return err
// }
//
// return nil
// }
func (r *NodeRepository) DeleteNode(id int64) error {
_, err := r.DB.Exec(`DELETE FROM node WHERE node.id = ?`, id)
if err != nil {
cclog.Errorf("Error while deleting node '%d' from DB", id)
return err
}
cclog.Infof("deleted node '%d' from DB", id)
return nil
}
// QueryNodes returns a list of nodes based on a node filter. It always operates
// on the last state (largest timestamp).
func (r *NodeRepository) QueryNodes(
ctx context.Context,
filters []*model.NodeFilter,
order *model.OrderByInput, // Currently unused!
) ([]*schema.Node, error) {
query, qerr := AccessCheck(ctx,
sq.Select("node.hostname", "node.cluster", "node.subcluster", "node_state.node_state",
"node_state.health_state", "MAX(node_state.time_stamp) as time").
From("node").
Join("node_state ON nodes_state.node_id = node.id"))
if qerr != nil {
return nil, qerr
}
for _, f := range filters {
if f.Hostname != nil {
query = buildStringCondition("node.hostname", f.Hostname, query)
}
if f.Cluster != nil {
query = buildStringCondition("node.cluster", f.Cluster, query)
}
if f.Subcluster != nil {
query = buildStringCondition("node.subcluster", f.Subcluster, query)
}
if f.SchedulerState != nil {
query = query.Where("node.node_state = ?", f.SchedulerState)
}
if f.HealthState != nil {
query = query.Where("node.health_state = ?", f.HealthState)
}
}
// Add Grouping after filters
query = query.GroupBy("node_state.node_id")
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
queryString, queryVars, _ := query.ToSql()
cclog.Errorf("Error while running query '%s' %v: %v", queryString, queryVars, err)
return nil, err
}
nodes := make([]*schema.Node, 0, 50)
for rows.Next() {
node := schema.Node{}
if err := rows.Scan(&node.Hostname, &node.Cluster, &node.SubCluster,
&node.NodeState, &node.HealthState); err != nil {
rows.Close()
cclog.Warn("Error while scanning rows (Nodes)")
return nil, err
}
nodes = append(nodes, &node)
}
return nodes, nil
}
func (r *NodeRepository) ListNodes(cluster string) ([]*schema.Node, error) {
q := sq.Select("node.hostname", "node.cluster", "node.subcluster", "node_state.node_state",
"node_state.health_state", "MAX(node_state.time_stamp) as time").
From("node").
Join("node_state ON node_state.node_id = node.id").
Where("node.cluster = ?", cluster).
GroupBy("node_state.node_id").
OrderBy("node.hostname ASC")
rows, err := q.RunWith(r.DB).Query()
if err != nil {
cclog.Warn("Error while querying node list")
return nil, err
}
nodeList := make([]*schema.Node, 0, 100)
defer rows.Close()
for rows.Next() {
node := &schema.Node{}
if err := rows.Scan(&node.Hostname, &node.Cluster,
&node.SubCluster, &node.NodeState, &node.HealthState); err != nil {
cclog.Warn("Error while scanning node list")
return nil, err
}
nodeList = append(nodeList, node)
}
return nodeList, nil
}
func (r *NodeRepository) CountNodeStates(ctx context.Context, filters []*model.NodeFilter) ([]*model.NodeStates, error) {
query, qerr := AccessCheck(ctx, sq.Select("hostname", "node_state", "MAX(time_stamp) as time").From("node"))
if qerr != nil {
return nil, qerr
}
query = query.Join("node_state ON node_state.node_id = node.id")
for _, f := range filters {
if f.Hostname != nil {
query = buildStringCondition("hostname", f.Hostname, query)
}
if f.Cluster != nil {
query = buildStringCondition("cluster", f.Cluster, query)
}
if f.Subcluster != nil {
query = buildStringCondition("subcluster", f.Subcluster, query)
}
if f.SchedulerState != nil {
query = query.Where("node_state = ?", f.SchedulerState)
}
if f.HealthState != nil {
query = query.Where("health_state = ?", f.HealthState)
}
}
// Add Group and Order
query = query.GroupBy("hostname").OrderBy("hostname DESC")
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
queryString, queryVars, _ := query.ToSql()
cclog.Errorf("Error while running query '%s' %v: %v", queryString, queryVars, err)
return nil, err
}
stateMap := map[string]int{}
for rows.Next() {
var hostname, node_state string
var timestamp int64
if err := rows.Scan(&hostname, &node_state, &timestamp); err != nil {
rows.Close()
cclog.Warn("Error while scanning rows (NodeStates)")
return nil, err
}
stateMap[node_state] += 1
}
nodes := make([]*model.NodeStates, 0)
for state, counts := range stateMap {
node := model.NodeStates{State: state, Count: counts}
nodes = append(nodes, &node)
}
return nodes, nil
}
func (r *NodeRepository) CountHealthStates(ctx context.Context, filters []*model.NodeFilter) ([]*model.NodeStates, error) {
query, qerr := AccessCheck(ctx, sq.Select("hostname", "health_state", "MAX(time_stamp) as time").From("node"))
if qerr != nil {
return nil, qerr
}
query = query.Join("node_state ON node_state.node_id = node.id")
for _, f := range filters {
if f.Hostname != nil {
query = buildStringCondition("hostname", f.Hostname, query)
}
if f.Cluster != nil {
query = buildStringCondition("cluster", f.Cluster, query)
}
if f.Subcluster != nil {
query = buildStringCondition("subcluster", f.Subcluster, query)
}
if f.SchedulerState != nil {
query = query.Where("node_state = ?", f.SchedulerState)
}
if f.HealthState != nil {
query = query.Where("health_state = ?", f.HealthState)
}
}
// Add Group and Order
query = query.GroupBy("hostname").OrderBy("hostname DESC")
rows, err := query.RunWith(r.stmtCache).Query()
if err != nil {
queryString, queryVars, _ := query.ToSql()
cclog.Errorf("Error while running query '%s' %v: %v", queryString, queryVars, err)
return nil, err
}
stateMap := map[string]int{}
for rows.Next() {
var hostname, health_state string
var timestamp int64
if err := rows.Scan(&hostname, &health_state, &timestamp); err != nil {
rows.Close()
cclog.Warn("Error while scanning rows (NodeStates)")
return nil, err
}
stateMap[health_state] += 1
}
nodes := make([]*model.NodeStates, 0)
for state, counts := range stateMap {
node := model.NodeStates{State: state, Count: counts}
nodes = append(nodes, &node)
}
return nodes, nil
}
func AccessCheck(ctx context.Context, query sq.SelectBuilder) (sq.SelectBuilder, error) {
user := GetUserFromContext(ctx)
return AccessCheckWithUser(user, query)
}
func AccessCheckWithUser(user *schema.User, query sq.SelectBuilder) (sq.SelectBuilder, error) {
if user == nil {
var qnil sq.SelectBuilder
return qnil, fmt.Errorf("user context is nil")
}
switch {
// case len(user.Roles) == 1 && user.HasRole(schema.RoleApi): // API-User : Access NodeInfos
// return query, nil
case user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport}): // Admin & Support : Access NodeInfos
return query, nil
default: // No known Role: No Access, return error
var qnil sq.SelectBuilder
return qnil, fmt.Errorf("user has no or unknown roles")
}
}

View File

@@ -1,190 +0,0 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
"encoding/json"
"fmt"
"os"
"path/filepath"
"testing"
"time"
"github.com/ClusterCockpit/cc-backend/internal/config"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
_ "github.com/mattn/go-sqlite3"
)
func nodeTestSetup(t *testing.T) {
const testconfig = `{
"main": {
"addr": "0.0.0.0:8080",
"validate": false,
"apiAllowedIPs": [
"*"
]
},
"archive": {
"kind": "file",
"path": "./var/job-archive"
},
"auth": {
"jwts": {
"max-age": "2m"
}
},
"clusters": [
{
"name": "testcluster",
"metricDataRepository": {"kind": "test", "url": "bla:8081"},
"filterRanges": {
"numNodes": { "from": 1, "to": 64 },
"duration": { "from": 0, "to": 86400 },
"startTime": { "from": "2022-01-01T00:00:00Z", "to": null }
}
}
]
}`
const testclusterJSON = `{
"name": "testcluster",
"subClusters": [
{
"name": "sc1",
"nodes": "host123,host124,host125",
"processorType": "Intel Core i7-4770",
"socketsPerNode": 1,
"coresPerSocket": 4,
"threadsPerCore": 2,
"flopRateScalar": {
"unit": {
"prefix": "G",
"base": "F/s"
},
"value": 14
},
"flopRateSimd": {
"unit": {
"prefix": "G",
"base": "F/s"
},
"value": 112
},
"memoryBandwidth": {
"unit": {
"prefix": "G",
"base": "B/s"
},
"value": 24
},
"numberOfNodes": 70,
"topology": {
"node": [0, 1, 2, 3, 4, 5, 6, 7],
"socket": [[0, 1, 2, 3, 4, 5, 6, 7]],
"memoryDomain": [[0, 1, 2, 3, 4, 5, 6, 7]],
"die": [[0, 1, 2, 3, 4, 5, 6, 7]],
"core": [[0], [1], [2], [3], [4], [5], [6], [7]]
}
}
],
"metricConfig": [
{
"name": "load_one",
"unit": { "base": ""},
"scope": "node",
"timestep": 60,
"aggregation": "avg",
"peak": 8,
"normal": 0,
"caution": 0,
"alert": 0
}
]
}`
cclog.Init("debug", true)
tmpdir := t.TempDir()
jobarchive := filepath.Join(tmpdir, "job-archive")
if err := os.Mkdir(jobarchive, 0o777); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"),
fmt.Appendf(nil, "%d", 2), 0o666); err != nil {
t.Fatal(err)
}
if err := os.Mkdir(filepath.Join(jobarchive, "testcluster"),
0o777); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(filepath.Join(jobarchive, "testcluster", "cluster.json"),
[]byte(testclusterJSON), 0o666); err != nil {
t.Fatal(err)
}
dbfilepath := filepath.Join(tmpdir, "test.db")
err := MigrateDB("sqlite3", dbfilepath)
if err != nil {
t.Fatal(err)
}
cfgFilePath := filepath.Join(tmpdir, "config.json")
if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0o666); err != nil {
t.Fatal(err)
}
ccconf.Init(cfgFilePath)
// Load and check main configuration
if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
config.Init(cfg, clustercfg)
} else {
cclog.Abort("Cluster configuration must be present")
}
} else {
cclog.Abort("Main configuration must be present")
}
archiveCfg := fmt.Sprintf("{\"kind\": \"file\",\"path\": \"%s\"}", jobarchive)
Connect("sqlite3", dbfilepath)
if err := archive.Init(json.RawMessage(archiveCfg), config.Keys.DisableArchive); err != nil {
t.Fatal(err)
}
}
func TestUpdateNodeState(t *testing.T) {
nodeTestSetup(t)
nodeState := schema.NodeStateDB{
TimeStamp: time.Now().Unix(), NodeState: "allocated",
CpusAllocated: 72,
MemoryAllocated: 480,
GpusAllocated: 0,
HealthState: schema.MonitoringStateFull,
JobsRunning: 1,
}
repo := GetNodeRepository()
err := repo.UpdateNodeState("host124", "testcluster", &nodeState)
if err != nil {
return
}
node, err := repo.GetNode("host124", "testcluster", false)
if err != nil {
return
}
if node.NodeState != "allocated" {
t.Errorf("wrong node state\ngot: %s \nwant: allocated ", node.NodeState)
}
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -9,8 +9,8 @@ import (
"testing"
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
_ "github.com/mattn/go-sqlite3"
)
@@ -65,7 +65,7 @@ func BenchmarkDB_FindJobById(b *testing.B) {
func BenchmarkDB_FindJob(b *testing.B) {
var jobId int64 = 107266
var startTime int64 = 1657557241
cluster := "fritz"
var cluster = "fritz"
b.Run("FindJob", func(b *testing.B) {
db := setup(b)
@@ -147,7 +147,7 @@ func getContext(tb testing.TB) context.Context {
func setup(tb testing.TB) *JobRepository {
tb.Helper()
cclog.Init("warn", true)
log.Init("warn", true)
dbfile := "testdata/job.db"
err := MigrateDB("sqlite3", dbfile)
noErr(tb, err)

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -14,22 +14,20 @@ import (
"github.com/ClusterCockpit/cc-backend/internal/graph/model"
"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
// GraphQL validation should make sure that no unkown values can be specified.
var groupBy2column = map[model.Aggregate]string{
model.AggregateUser: "job.hpc_user",
model.AggregateProject: "job.project",
model.AggregateCluster: "job.cluster",
model.AggregateSubcluster: "job.subcluster",
model.AggregateUser: "job.hpc_user",
model.AggregateProject: "job.project",
model.AggregateCluster: "job.cluster",
}
var sortBy2column = map[model.SortByAggregate]string{
model.SortByAggregateTotaljobs: "totalJobs",
model.SortByAggregateTotalusers: "totalUsers",
model.SortByAggregateTotalwalltime: "totalWalltime",
model.SortByAggregateTotalnodes: "totalNodes",
model.SortByAggregateTotalnodehours: "totalNodeHours",
@@ -78,12 +76,8 @@ func (r *JobRepository) buildStatsQuery(
// fmt.Sprintf(`CAST(ROUND((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END) / 3600) as %s) as value`, time.Now().Unix(), castType)
if col != "" {
// Scan columns: id, name, totalJobs, totalUsers, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours
query = sq.Select(
col,
"name",
"COUNT(job.id) as totalJobs",
"COUNT(DISTINCT job.hpc_user) AS totalUsers",
// Scan columns: id, totalJobs, name, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours
query = sq.Select(col, "COUNT(job.id) as totalJobs", "name",
fmt.Sprintf(`CAST(ROUND(SUM((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END)) / 3600) as %s) as totalWalltime`, time.Now().Unix(), castType),
fmt.Sprintf(`CAST(SUM(job.num_nodes) as %s) as totalNodes`, castType),
fmt.Sprintf(`CAST(ROUND(SUM((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END) * job.num_nodes) / 3600) as %s) as totalNodeHours`, time.Now().Unix(), castType),
@@ -93,10 +87,8 @@ func (r *JobRepository) buildStatsQuery(
fmt.Sprintf(`CAST(ROUND(SUM((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END) * job.num_acc) / 3600) as %s) as totalAccHours`, time.Now().Unix(), castType),
).From("job").LeftJoin("hpc_user ON hpc_user.username = job.hpc_user").GroupBy(col)
} else {
// Scan columns: totalJobs, totalUsers, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours
query = sq.Select(
"COUNT(job.id) as totalJobs",
"COUNT(DISTINCT job.hpc_user) AS totalUsers",
// Scan columns: totalJobs, name, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours
query = sq.Select("COUNT(job.id)",
fmt.Sprintf(`CAST(ROUND(SUM((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END)) / 3600) as %s)`, time.Now().Unix(), castType),
fmt.Sprintf(`CAST(SUM(job.num_nodes) as %s)`, castType),
fmt.Sprintf(`CAST(ROUND(SUM((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END) * job.num_nodes) / 3600) as %s)`, time.Now().Unix(), castType),
@@ -166,7 +158,7 @@ func (r *JobRepository) JobsStatsGrouped(
rows, err := query.RunWith(r.DB).Query()
if err != nil {
cclog.Warn("Error while querying DB for job statistics")
log.Warn("Error while querying DB for job statistics")
return nil, err
}
@@ -175,14 +167,14 @@ func (r *JobRepository) JobsStatsGrouped(
for rows.Next() {
var id sql.NullString
var name sql.NullString
var jobs, users, walltime, nodes, nodeHours, cores, coreHours, accs, accHours sql.NullInt64
if err := rows.Scan(&id, &name, &jobs, &users, &walltime, &nodes, &nodeHours, &cores, &coreHours, &accs, &accHours); err != nil {
cclog.Warnf("Error while scanning rows: %s", err.Error())
var jobs, walltime, nodes, nodeHours, cores, coreHours, accs, accHours sql.NullInt64
if err := rows.Scan(&id, &jobs, &name, &walltime, &nodes, &nodeHours, &cores, &coreHours, &accs, &accHours); err != nil {
log.Warn("Error while scanning rows")
return nil, err
}
if id.Valid {
var totalJobs, totalUsers, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours int
var totalJobs, totalWalltime, totalNodes, totalNodeHours, totalCores, totalCoreHours, totalAccs, totalAccHours int
var personName string
if name.Valid {
@@ -193,10 +185,6 @@ func (r *JobRepository) JobsStatsGrouped(
totalJobs = int(jobs.Int64)
}
if users.Valid {
totalUsers = int(users.Int64)
}
if walltime.Valid {
totalWalltime = int(walltime.Int64)
}
@@ -240,9 +228,8 @@ func (r *JobRepository) JobsStatsGrouped(
stats = append(stats,
&model.JobsStatistics{
ID: id.String,
TotalJobs: totalJobs,
TotalUsers: totalUsers,
TotalWalltime: totalWalltime,
TotalJobs: int(jobs.Int64),
TotalWalltime: int(walltime.Int64),
TotalNodes: totalNodes,
TotalNodeHours: totalNodeHours,
TotalCores: totalCores,
@@ -254,7 +241,7 @@ func (r *JobRepository) JobsStatsGrouped(
}
}
cclog.Debugf("Timer JobsStatsGrouped %s", time.Since(start))
log.Debugf("Timer JobsStatsGrouped %s", time.Since(start))
return stats, nil
}
@@ -272,9 +259,9 @@ func (r *JobRepository) JobsStats(
row := query.RunWith(r.DB).QueryRow()
stats := make([]*model.JobsStatistics, 0, 1)
var jobs, users, walltime, nodes, nodeHours, cores, coreHours, accs, accHours sql.NullInt64
if err := row.Scan(&jobs, &users, &walltime, &nodes, &nodeHours, &cores, &coreHours, &accs, &accHours); err != nil {
cclog.Warn("Error while scanning rows")
var jobs, walltime, nodes, nodeHours, cores, coreHours, accs, accHours sql.NullInt64
if err := row.Scan(&jobs, &walltime, &nodes, &nodeHours, &cores, &coreHours, &accs, &accHours); err != nil {
log.Warn("Error while scanning rows")
return nil, err
}
@@ -293,7 +280,6 @@ func (r *JobRepository) JobsStats(
stats = append(stats,
&model.JobsStatistics{
TotalJobs: int(jobs.Int64),
TotalUsers: int(users.Int64),
TotalWalltime: int(walltime.Int64),
TotalNodeHours: totalNodeHours,
TotalCoreHours: totalCoreHours,
@@ -301,11 +287,11 @@ func (r *JobRepository) JobsStats(
})
}
cclog.Debugf("Timer JobStats %s", time.Since(start))
log.Debugf("Timer JobStats %s", time.Since(start))
return stats, nil
}
func LoadJobStat(job *schema.Job, metric string, statType string) float64 {
func LoadJobStat(job *schema.JobMeta, metric string, statType string) float64 {
if stats, ok := job.Statistics[metric]; ok {
switch statType {
case "avg":
@@ -315,7 +301,7 @@ func LoadJobStat(job *schema.Job, metric string, statType string) float64 {
case "min":
return stats.Min
default:
cclog.Errorf("Unknown stat type %s", statType)
log.Errorf("Unknown stat type %s", statType)
}
}
@@ -336,7 +322,7 @@ func (r *JobRepository) JobCountGrouped(
}
rows, err := query.RunWith(r.DB).Query()
if err != nil {
cclog.Warn("Error while querying DB for job statistics")
log.Warn("Error while querying DB for job statistics")
return nil, err
}
@@ -346,7 +332,7 @@ func (r *JobRepository) JobCountGrouped(
var id sql.NullString
var cnt sql.NullInt64
if err := rows.Scan(&id, &cnt); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
if id.Valid {
@@ -358,7 +344,7 @@ func (r *JobRepository) JobCountGrouped(
}
}
cclog.Debugf("Timer JobCountGrouped %s", time.Since(start))
log.Debugf("Timer JobCountGrouped %s", time.Since(start))
return stats, nil
}
@@ -378,7 +364,7 @@ func (r *JobRepository) AddJobCountGrouped(
}
rows, err := query.RunWith(r.DB).Query()
if err != nil {
cclog.Warn("Error while querying DB for job statistics")
log.Warn("Error while querying DB for job statistics")
return nil, err
}
@@ -388,7 +374,7 @@ func (r *JobRepository) AddJobCountGrouped(
var id sql.NullString
var cnt sql.NullInt64
if err := rows.Scan(&id, &cnt); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
if id.Valid {
@@ -407,7 +393,7 @@ func (r *JobRepository) AddJobCountGrouped(
}
}
cclog.Debugf("Timer AddJobCountGrouped %s", time.Since(start))
log.Debugf("Timer AddJobCountGrouped %s", time.Since(start))
return stats, nil
}
@@ -425,7 +411,7 @@ func (r *JobRepository) AddJobCount(
}
rows, err := query.RunWith(r.DB).Query()
if err != nil {
cclog.Warn("Error while querying DB for job statistics")
log.Warn("Error while querying DB for job statistics")
return nil, err
}
@@ -434,7 +420,7 @@ func (r *JobRepository) AddJobCount(
for rows.Next() {
var cnt sql.NullInt64
if err := rows.Scan(&cnt); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
@@ -452,7 +438,7 @@ func (r *JobRepository) AddJobCount(
}
}
cclog.Debugf("Timer AddJobCount %s", time.Since(start))
log.Debugf("Timer AddJobCount %s", time.Since(start))
return stats, nil
}
@@ -493,29 +479,29 @@ func (r *JobRepository) AddHistograms(
value := fmt.Sprintf(`CAST(ROUND(((CASE WHEN job.job_state = "running" THEN %d - job.start_time ELSE job.duration END) / %d) + 1) as %s) as value`, time.Now().Unix(), targetBinSize, castType)
stat.HistDuration, err = r.jobsDurationStatisticsHistogram(ctx, value, filter, targetBinSize, &targetBinCount)
if err != nil {
cclog.Warn("Error while loading job statistics histogram: job duration")
log.Warn("Error while loading job statistics histogram: job duration")
return nil, err
}
stat.HistNumNodes, err = r.jobsStatisticsHistogram(ctx, "job.num_nodes as value", filter)
if err != nil {
cclog.Warn("Error while loading job statistics histogram: num nodes")
log.Warn("Error while loading job statistics histogram: num nodes")
return nil, err
}
stat.HistNumCores, err = r.jobsStatisticsHistogram(ctx, "job.num_hwthreads as value", filter)
if err != nil {
cclog.Warn("Error while loading job statistics histogram: num hwthreads")
log.Warn("Error while loading job statistics histogram: num hwthreads")
return nil, err
}
stat.HistNumAccs, err = r.jobsStatisticsHistogram(ctx, "job.num_acc as value", filter)
if err != nil {
cclog.Warn("Error while loading job statistics histogram: num acc")
log.Warn("Error while loading job statistics histogram: num acc")
return nil, err
}
cclog.Debugf("Timer AddHistograms %s", time.Since(start))
log.Debugf("Timer AddHistograms %s", time.Since(start))
return stat, nil
}
@@ -534,7 +520,7 @@ func (r *JobRepository) AddMetricHistograms(
if f.State != nil {
if len(f.State) == 1 && f.State[0] == "running" {
stat.HistMetrics = r.runningJobsMetricStatisticsHistogram(ctx, metrics, filter, targetBinCount)
cclog.Debugf("Timer AddMetricHistograms %s", time.Since(start))
log.Debugf("Timer AddMetricHistograms %s", time.Since(start))
return stat, nil
}
}
@@ -544,13 +530,13 @@ func (r *JobRepository) AddMetricHistograms(
for _, m := range metrics {
metricHisto, err := r.jobsMetricStatisticsHistogram(ctx, m, filter, targetBinCount)
if err != nil {
cclog.Warnf("Error while loading job metric statistics histogram: %s", m)
log.Warnf("Error while loading job metric statistics histogram: %s", m)
continue
}
stat.HistMetrics = append(stat.HistMetrics, metricHisto)
}
cclog.Debugf("Timer AddMetricHistograms %s", time.Since(start))
log.Debugf("Timer AddMetricHistograms %s", time.Since(start))
return stat, nil
}
@@ -574,7 +560,7 @@ func (r *JobRepository) jobsStatisticsHistogram(
rows, err := query.GroupBy("value").RunWith(r.DB).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -583,13 +569,13 @@ func (r *JobRepository) jobsStatisticsHistogram(
for rows.Next() {
point := model.HistoPoint{}
if err := rows.Scan(&point.Value, &point.Count); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
points = append(points, &point)
}
cclog.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
log.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
return points, nil
}
@@ -621,7 +607,7 @@ func (r *JobRepository) jobsDurationStatisticsHistogram(
rows, err := query.GroupBy("value").RunWith(r.DB).Query()
if err != nil {
cclog.Error("Error while running query")
log.Error("Error while running query")
return nil, err
}
@@ -629,7 +615,7 @@ func (r *JobRepository) jobsDurationStatisticsHistogram(
for rows.Next() {
point := model.HistoPoint{}
if err := rows.Scan(&point.Value, &point.Count); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
@@ -644,7 +630,7 @@ func (r *JobRepository) jobsDurationStatisticsHistogram(
}
}
cclog.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
log.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
return points, nil
}
@@ -666,7 +652,7 @@ func (r *JobRepository) jobsMetricStatisticsHistogram(
peak = metricConfig.Peak
unit = metricConfig.Unit.Prefix + metricConfig.Unit.Base
footprintStat = metricConfig.Footprint
cclog.Debugf("Cluster %s filter found with peak %f for %s", *f.Cluster.Eq, peak, metric)
log.Debugf("Cluster %s filter found with peak %f for %s", *f.Cluster.Eq, peak, metric)
}
}
@@ -688,32 +674,57 @@ func (r *JobRepository) jobsMetricStatisticsHistogram(
}
}
// cclog.Debugf("Metric %s, Peak %f, Unit %s", metric, peak, unit)
// Make bins, see https://jereze.com/code/sql-histogram/ (Modified here)
start := time.Now()
// log.Debugf("Metric %s, Peak %f, Unit %s, Aggregation %s", metric, peak, unit, aggreg)
// Make bins, see https://jereze.com/code/sql-histogram/
// Find Jobs' Value Bin Number: Divide Value by Peak, Multiply by RequestedBins, then CAST to INT: Gets Bin-Number of Job
binQuery := fmt.Sprintf(`CAST(
((case when json_extract(footprint, "$.%s") = %f then %f*0.999999999 else json_extract(footprint, "$.%s") end) / %f)
* %v as INTEGER )`,
(metric + "_" + footprintStat), peak, peak, (metric + "_" + footprintStat), peak, *bins)
start := time.Now()
jm := fmt.Sprintf(`json_extract(footprint, "$.%s")`, (metric + "_" + footprintStat))
crossJoinQuery := sq.Select(
fmt.Sprintf(`max(%s) as max`, jm),
fmt.Sprintf(`min(%s) as min`, jm),
).From("job").Where(
"JSON_VALID(footprint)",
).Where(
fmt.Sprintf(`%s is not null`, jm),
).Where(
fmt.Sprintf(`%s <= %f`, jm, peak),
)
crossJoinQuery, cjqerr := SecurityCheck(ctx, crossJoinQuery)
if cjqerr != nil {
return nil, cjqerr
}
for _, f := range filters {
crossJoinQuery = BuildWhereClause(f, crossJoinQuery)
}
crossJoinQuerySql, crossJoinQueryArgs, sqlerr := crossJoinQuery.ToSql()
if sqlerr != nil {
return nil, sqlerr
}
binQuery := fmt.Sprintf(`CAST( (case when %s = value.max
then value.max*0.999999999 else %s end - value.min) / (value.max -
value.min) * %v as INTEGER )`, jm, jm, *bins)
mainQuery := sq.Select(
fmt.Sprintf(`%s + 1 as bin`, binQuery),
`count(*) as count`,
// For Debug: // fmt.Sprintf(`CAST((%f / %d) as INTEGER ) * %s as min`, peak, *bins, binQuery),
// For Debug: // fmt.Sprintf(`CAST((%f / %d) as INTEGER ) * (%s + 1) as max`, peak, *bins, binQuery),
).From("job").Where(
"JSON_VALID(footprint)",
).Where(fmt.Sprintf(`json_extract(footprint, "$.%s") is not null and json_extract(footprint, "$.%s") <= %f`, (metric + "_" + footprintStat), (metric + "_" + footprintStat), peak))
fmt.Sprintf(`count(%s) as count`, jm),
fmt.Sprintf(`CAST(((value.max / %d) * (%v )) as INTEGER ) as min`, *bins, binQuery),
fmt.Sprintf(`CAST(((value.max / %d) * (%v + 1 )) as INTEGER ) as max`, *bins, binQuery),
).From("job").CrossJoin(
fmt.Sprintf(`(%s) as value`, crossJoinQuerySql), crossJoinQueryArgs...,
).Where(fmt.Sprintf(`%s is not null and %s <= %f`, jm, jm, peak))
// Only accessible Jobs...
mainQuery, qerr := SecurityCheck(ctx, mainQuery)
if qerr != nil {
return nil, qerr
}
// Filters...
for _, f := range filters {
mainQuery = BuildWhereClause(f, mainQuery)
}
@@ -723,38 +734,36 @@ func (r *JobRepository) jobsMetricStatisticsHistogram(
rows, err := mainQuery.RunWith(r.DB).Query()
if err != nil {
cclog.Errorf("Error while running mainQuery: %s", err)
log.Errorf("Error while running mainQuery: %s", err)
return nil, err
}
// Setup Return Array With Bin-Numbers for Match and Min/Max based on Peak
// Setup Array
points := make([]*model.MetricHistoPoint, 0)
binStep := int(peak) / *bins
for i := 1; i <= *bins; i++ {
binMin := (binStep * (i - 1))
binMax := (binStep * i)
epoint := model.MetricHistoPoint{Bin: &i, Count: 0, Min: &binMin, Max: &binMax}
points = append(points, &epoint)
binMax := ((int(peak) / *bins) * i)
binMin := ((int(peak) / *bins) * (i - 1))
point := model.MetricHistoPoint{Bin: &i, Count: 0, Min: &binMin, Max: &binMax}
points = append(points, &point)
}
for rows.Next() { // Fill Count if Bin-No. Matches (Not every Bin exists in DB!)
rpoint := model.MetricHistoPoint{}
if err := rows.Scan(&rpoint.Bin, &rpoint.Count); err != nil { // Required for Debug: &rpoint.Min, &rpoint.Max
cclog.Warnf("Error while scanning rows for %s", metric)
return nil, err // FIXME: Totally bricks cc-backend if returned and if all metrics requested?
for rows.Next() {
point := model.MetricHistoPoint{}
if err := rows.Scan(&point.Bin, &point.Count, &point.Min, &point.Max); err != nil {
log.Warnf("Error while scanning rows for %s", jm)
return nil, err // Totally bricks cc-backend if returned and if all metrics requested?
}
for _, e := range points {
if e.Bin != nil && rpoint.Bin != nil {
if *e.Bin == *rpoint.Bin {
e.Count = rpoint.Count
// Only Required For Debug: Check DB returned Min/Max against Backend Init above
// if rpoint.Min != nil {
// cclog.Warnf(">>>> Bin %d Min Set For %s to %d (Init'd with: %d)", *e.Bin, metric, *rpoint.Min, *e.Min)
// }
// if rpoint.Max != nil {
// cclog.Warnf(">>>> Bin %d Max Set For %s to %d (Init'd with: %d)", *e.Bin, metric, *rpoint.Max, *e.Max)
// }
if e.Bin != nil && point.Bin != nil {
if *e.Bin == *point.Bin {
e.Count = point.Count
if point.Min != nil {
e.Min = point.Min
}
if point.Max != nil {
e.Max = point.Max
}
break
}
}
@@ -763,7 +772,7 @@ func (r *JobRepository) jobsMetricStatisticsHistogram(
result := model.MetricHistoPoints{Metric: metric, Unit: unit, Stat: &footprintStat, Data: points}
cclog.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
log.Debugf("Timer jobsStatisticsHistogram %s", time.Since(start))
return &result, nil
}
@@ -773,14 +782,15 @@ func (r *JobRepository) runningJobsMetricStatisticsHistogram(
filters []*model.JobFilter,
bins *int,
) []*model.MetricHistoPoints {
// Get Jobs
jobs, err := r.QueryJobs(ctx, filters, &model.PageRequest{Page: 1, ItemsPerPage: 500 + 1}, nil)
if err != nil {
cclog.Errorf("Error while querying jobs for footprint: %s", err)
log.Errorf("Error while querying jobs for footprint: %s", err)
return nil
}
if len(jobs) > 500 {
cclog.Errorf("too many jobs matched (max: %d)", 500)
log.Errorf("too many jobs matched (max: %d)", 500)
return nil
}
@@ -796,7 +806,7 @@ func (r *JobRepository) runningJobsMetricStatisticsHistogram(
}
if err := metricDataDispatcher.LoadAverages(job, metrics, avgs, ctx); err != nil {
cclog.Errorf("Error while loading averages for histogram: %s", err)
log.Errorf("Error while loading averages for histogram: %s", err)
return nil
}
}

View File

@@ -1,8 +1,7 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
import (
@@ -20,6 +19,7 @@ func TestBuildJobStatsQuery(t *testing.T) {
noErr(t, err)
fmt.Printf("SQL: %s\n", sql)
}
func TestJobStats(t *testing.T) {
@@ -29,7 +29,7 @@ func TestJobStats(t *testing.T) {
stats, err := r.JobsStats(getContext(t), []*model.JobFilter{filter})
noErr(t, err)
if stats[0].TotalJobs != 544 {
t.Fatalf("Want 544, Got %d", stats[0].TotalJobs)
if stats[0].TotalJobs != 6 {
t.Fatalf("Want 98, Got %d", stats[0].TotalJobs)
}
}

View File

@@ -1,5 +1,5 @@
// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
// All rights reserved. This file is part of cc-backend.
// All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package repository
@@ -9,8 +9,8 @@ import (
"strings"
"github.com/ClusterCockpit/cc-backend/pkg/archive"
cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
"github.com/ClusterCockpit/cc-lib/schema"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
sq "github.com/Masterminds/squirrel"
)
@@ -18,7 +18,7 @@ import (
func (r *JobRepository) AddTag(user *schema.User, job int64, tag int64) ([]*schema.Tag, error) {
j, err := r.FindByIdWithUser(user, job)
if err != nil {
cclog.Warn("Error while finding job by id")
log.Warn("Error while finding job by id")
return nil, err
}
@@ -26,61 +26,30 @@ func (r *JobRepository) AddTag(user *schema.User, job int64, tag int64) ([]*sche
if _, err := q.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error adding tag with %s: %v", s, err)
log.Errorf("Error adding tag with %s: %v", s, err)
return nil, err
}
tags, err := r.GetTags(user, &job)
if err != nil {
cclog.Warn("Error while getting tags for job")
log.Warn("Error while getting tags for job")
return nil, err
}
archiveTags, err := r.getArchiveTags(&job)
if err != nil {
cclog.Warn("Error while getting tags for job")
log.Warn("Error while getting tags for job")
return nil, err
}
return tags, archive.UpdateTags(j, archiveTags)
}
func (r *JobRepository) AddTagDirect(job int64, tag int64) ([]*schema.Tag, error) {
j, err := r.FindByIdDirect(job)
if err != nil {
cclog.Warn("Error while finding job by id")
return nil, err
}
q := sq.Insert("jobtag").Columns("job_id", "tag_id").Values(job, tag)
if _, err := q.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error adding tag with %s: %v", s, err)
return nil, err
}
tags, err := r.GetTagsDirect(&job)
if err != nil {
cclog.Warn("Error while getting tags for job")
return nil, err
}
archiveTags, err := r.getArchiveTags(&job)
if err != nil {
cclog.Warn("Error while getting tags for job")
return nil, err
}
return tags, archive.UpdateTags(j, archiveTags)
}
// Removes a tag from a job by tag id.
// Used by GraphQL API
// Removes a tag from a job
func (r *JobRepository) RemoveTag(user *schema.User, job, tag int64) ([]*schema.Tag, error) {
j, err := r.FindByIdWithUser(user, job)
if err != nil {
cclog.Warn("Error while finding job by id")
log.Warn("Error while finding job by id")
return nil, err
}
@@ -88,129 +57,25 @@ func (r *JobRepository) RemoveTag(user *schema.User, job, tag int64) ([]*schema.
if _, err := q.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error removing tag with %s: %v", s, err)
log.Errorf("Error removing tag with %s: %v", s, err)
return nil, err
}
tags, err := r.GetTags(user, &job)
if err != nil {
cclog.Warn("Error while getting tags for job")
log.Warn("Error while getting tags for job")
return nil, err
}
archiveTags, err := r.getArchiveTags(&job)
if err != nil {
cclog.Warn("Error while getting tags for job")
log.Warn("Error while getting tags for job")
return nil, err
}
return tags, archive.UpdateTags(j, archiveTags)
}
// Removes a tag from a job by tag info
// Used by REST API
func (r *JobRepository) RemoveJobTagByRequest(user *schema.User, job int64, tagType string, tagName string, tagScope string) ([]*schema.Tag, error) {
// Get Tag ID to delete
tagID, exists := r.TagId(tagType, tagName, tagScope)
if !exists {
cclog.Warnf("Tag does not exist (name, type, scope): %s, %s, %s", tagName, tagType, tagScope)
return nil, fmt.Errorf("tag does not exist (name, type, scope): %s, %s, %s", tagName, tagType, tagScope)
}
// Get Job
j, err := r.FindByIdWithUser(user, job)
if err != nil {
cclog.Warn("Error while finding job by id")
return nil, err
}
// Handle Delete
q := sq.Delete("jobtag").Where("jobtag.job_id = ?", job).Where("jobtag.tag_id = ?", tagID)
if _, err := q.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error removing tag from table 'jobTag' with %s: %v", s, err)
return nil, err
}
tags, err := r.GetTags(user, &job)
if err != nil {
cclog.Warn("Error while getting tags for job")
return nil, err
}
archiveTags, err := r.getArchiveTags(&job)
if err != nil {
cclog.Warn("Error while getting tags for job")
return nil, err
}
return tags, archive.UpdateTags(j, archiveTags)
}
func (r *JobRepository) removeTagFromArchiveJobs(jobIds []int64) {
for _, j := range jobIds {
tags, err := r.getArchiveTags(&j)
if err != nil {
cclog.Warnf("Error while getting tags for job %d", j)
continue
}
job, err := r.FindByIdDirect(j)
if err != nil {
cclog.Warnf("Error while getting job %d", j)
continue
}
archive.UpdateTags(job, tags)
}
}
// Removes a tag from db by tag info
// Used by REST API. Does not update tagged jobs in Job archive.
func (r *JobRepository) RemoveTagByRequest(tagType string, tagName string, tagScope string) error {
// Get Tag ID to delete
tagID, exists := r.TagId(tagType, tagName, tagScope)
if !exists {
cclog.Warnf("Tag does not exist (name, type, scope): %s, %s, %s", tagName, tagType, tagScope)
return fmt.Errorf("tag does not exist (name, type, scope): %s, %s, %s", tagName, tagType, tagScope)
}
return r.RemoveTagById(tagID)
}
// Removes a tag from db by tag id
// Used by GraphQL API.
func (r *JobRepository) RemoveTagById(tagID int64) error {
jobIds, err := r.FindJobIdsByTag(tagID)
if err != nil {
return err
}
// Handle Delete JobTagTable
qJobTag := sq.Delete("jobtag").Where("jobtag.tag_id = ?", tagID)
if _, err := qJobTag.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := qJobTag.ToSql()
cclog.Errorf("Error removing tag from table 'jobTag' with %s: %v", s, err)
return err
}
// Handle Delete TagTable
qTag := sq.Delete("tag").Where("tag.id = ?", tagID)
if _, err := qTag.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := qTag.ToSql()
cclog.Errorf("Error removing tag from table 'tag' with %s: %v", s, err)
return err
}
// asynchronously update archive jobs
go r.removeTagFromArchiveJobs(jobIds)
return nil
}
// CreateTag creates a new tag with the specified type and name and returns its database id.
func (r *JobRepository) CreateTag(tagType string, tagName string, tagScope string) (tagId int64, err error) {
// Default to "Global" scope if none defined
@@ -223,7 +88,7 @@ func (r *JobRepository) CreateTag(tagType string, tagName string, tagScope strin
res, err := q.RunWith(r.stmtCache).Exec()
if err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error inserting tag with %s: %v", s, err)
log.Errorf("Error inserting tag with %s: %v", s, err)
return 0, err
}
@@ -272,7 +137,7 @@ func (r *JobRepository) CountTags(user *schema.User) (tags []schema.Tag, counts
// Handle Job Ownership
if user != nil && user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport}) { // ADMIN || SUPPORT: Count all jobs
// cclog.Debug("CountTags: User Admin or Support -> Count all Jobs for Tags")
// log.Debug("CountTags: User Admin or Support -> Count all Jobs for Tags")
// Unchanged: Needs to be own case still, due to UserRole/NoRole compatibility handling in else case
} else if user != nil && user.HasRole(schema.RoleManager) { // MANAGER: Count own jobs plus project's jobs
// Build ("project1", "project2", ...) list of variable length directly in SQL string
@@ -333,38 +198,6 @@ func (r *JobRepository) AddTagOrCreate(user *schema.User, jobId int64, tagType s
return tagId, nil
}
// used in auto tagger plugins
func (r *JobRepository) AddTagOrCreateDirect(jobId int64, tagType string, tagName string) (tagId int64, err error) {
tagScope := "global"
tagId, exists := r.TagId(tagType, tagName, tagScope)
if !exists {
tagId, err = r.CreateTag(tagType, tagName, tagScope)
if err != nil {
return 0, err
}
}
if _, err := r.AddTagDirect(jobId, tagId); err != nil {
return 0, err
}
return tagId, nil
}
func (r *JobRepository) HasTag(jobId int64, tagType string, tagName string) bool {
var id int64
q := sq.Select("id").From("tag").Join("jobtag ON jobtag.tag_id = tag.id").
Where("jobtag.job_id = ?", jobId).Where("tag.tag_type = ?", tagType).
Where("tag.tag_name = ?", tagName)
err := q.RunWith(r.stmtCache).QueryRow().Scan(&id)
if err != nil {
return false
} else {
return true
}
}
// TagId returns the database id of the tag with the specified type and name.
func (r *JobRepository) TagId(tagType string, tagName string, tagScope string) (tagId int64, exists bool) {
exists = true
@@ -376,16 +209,6 @@ func (r *JobRepository) TagId(tagType string, tagName string, tagScope string) (
return
}
// TagInfo returns the database infos of the tag with the specified id.
func (r *JobRepository) TagInfo(tagId int64) (tagType string, tagName string, tagScope string, exists bool) {
exists = true
if err := sq.Select("tag.tag_type", "tag.tag_name", "tag.tag_scope").From("tag").Where("tag.id = ?", tagId).
RunWith(r.stmtCache).QueryRow().Scan(&tagType, &tagName, &tagScope); err != nil {
exists = false
}
return
}
// GetTags returns a list of all scoped tags if job is nil or of the tags that the job with that database ID has.
func (r *JobRepository) GetTags(user *schema.User, job *int64) ([]*schema.Tag, error) {
q := sq.Select("id", "tag_type", "tag_name", "tag_scope").From("tag")
@@ -396,7 +219,7 @@ func (r *JobRepository) GetTags(user *schema.User, job *int64) ([]*schema.Tag, e
rows, err := q.RunWith(r.stmtCache).Query()
if err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error get tags with %s: %v", s, err)
log.Errorf("Error get tags with %s: %v", s, err)
return nil, err
}
@@ -404,7 +227,7 @@ func (r *JobRepository) GetTags(user *schema.User, job *int64) ([]*schema.Tag, e
for rows.Next() {
tag := &schema.Tag{}
if err := rows.Scan(&tag.ID, &tag.Type, &tag.Name, &tag.Scope); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
// Handle Scope Filtering: Tag Scope is Global, Private (== Username) or User is auth'd to view Admin Tags
@@ -420,32 +243,6 @@ func (r *JobRepository) GetTags(user *schema.User, job *int64) ([]*schema.Tag, e
return tags, nil
}
func (r *JobRepository) GetTagsDirect(job *int64) ([]*schema.Tag, error) {
q := sq.Select("id", "tag_type", "tag_name", "tag_scope").From("tag")
if job != nil {
q = q.Join("jobtag ON jobtag.tag_id = tag.id").Where("jobtag.job_id = ?", *job)
}
rows, err := q.RunWith(r.stmtCache).Query()
if err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error get tags with %s: %v", s, err)
return nil, err
}
tags := make([]*schema.Tag, 0)
for rows.Next() {
tag := &schema.Tag{}
if err := rows.Scan(&tag.ID, &tag.Type, &tag.Name, &tag.Scope); err != nil {
cclog.Warn("Error while scanning rows")
return nil, err
}
tags = append(tags, tag)
}
return tags, nil
}
// GetArchiveTags returns a list of all tags *regardless of scope* for archiving if job is nil or of the tags that the job with that database ID has.
func (r *JobRepository) getArchiveTags(job *int64) ([]*schema.Tag, error) {
q := sq.Select("id", "tag_type", "tag_name", "tag_scope").From("tag")
@@ -456,7 +253,7 @@ func (r *JobRepository) getArchiveTags(job *int64) ([]*schema.Tag, error) {
rows, err := q.RunWith(r.stmtCache).Query()
if err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error get tags with %s: %v", s, err)
log.Errorf("Error get tags with %s: %v", s, err)
return nil, err
}
@@ -464,7 +261,7 @@ func (r *JobRepository) getArchiveTags(job *int64) ([]*schema.Tag, error) {
for rows.Next() {
tag := &schema.Tag{}
if err := rows.Scan(&tag.ID, &tag.Type, &tag.Name, &tag.Scope); err != nil {
cclog.Warn("Error while scanning rows")
log.Warn("Error while scanning rows")
return nil, err
}
tags = append(tags, tag)
@@ -488,7 +285,7 @@ func (r *JobRepository) ImportTag(jobId int64, tagType string, tagName string, t
if _, err := q.RunWith(r.stmtCache).Exec(); err != nil {
s, _, _ := q.ToSql()
cclog.Errorf("Error adding tag on import with %s: %v", s, err)
log.Errorf("Error adding tag on import with %s: %v", s, err)
return err
}

Some files were not shown because too many files have changed in this diff Show More