1
0
mirror of https://github.com/ClusterCockpit/cc-backend synced 2025-04-09 05:35:55 +02:00

Compare commits

..

28 Commits

Author SHA1 Message Date
Christoph Kluge
96977c6183
Merge pull request from ClusterCockpit/review_logging
Review logging
2025-04-07 16:03:48 +02:00
Jan Eitzinger
73d83164fc
Merge pull request from ClusterCockpit/dependabot/go_modules/golang.org/x/net-0.36.0
Bump golang.org/x/net from 0.35.0 to 0.36.0
2025-04-04 11:05:01 +02:00
dependabot[bot]
1064f5e4a8
Bump golang.org/x/net from 0.35.0 to 0.36.0
Bumps [golang.org/x/net](https://github.com/golang/net) from 0.35.0 to 0.36.0.
- [Commits](https://github.com/golang/net/compare/v0.35.0...v0.36.0)

---
updated-dependencies:
- dependency-name: golang.org/x/net
  dependency-version: 0.36.0
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
2025-04-04 09:01:59 +00:00
Jan Eitzinger
5be98c7087
Merge pull request from ClusterCockpit/dependabot/npm_and_yarn/web/frontend/babel/runtime-7.27.0
Bump @babel/runtime from 7.26.0 to 7.27.0 in /web/frontend
2025-04-04 10:55:34 +02:00
dependabot[bot]
0d689c7dff
Bump @babel/runtime from 7.26.0 to 7.27.0 in /web/frontend
Bumps [@babel/runtime](https://github.com/babel/babel/tree/HEAD/packages/babel-runtime) from 7.26.0 to 7.27.0.
- [Release notes](https://github.com/babel/babel/releases)
- [Changelog](https://github.com/babel/babel/blob/main/CHANGELOG.md)
- [Commits](https://github.com/babel/babel/commits/v7.27.0/packages/babel-runtime)

---
updated-dependencies:
- dependency-name: "@babel/runtime"
  dependency-version: 7.27.0
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
2025-04-04 08:45:16 +00:00
Jan Eitzinger
1f24ed46a0
Merge pull request from ClusterCockpit/dependabot/go_modules/github.com/golang-jwt/jwt/v5-5.2.2
Bump github.com/golang-jwt/jwt/v5 from 5.2.1 to 5.2.2
2025-04-04 10:37:18 +02:00
dependabot[bot]
92b4159f9e
Bump github.com/golang-jwt/jwt/v5 from 5.2.1 to 5.2.2
Bumps [github.com/golang-jwt/jwt/v5](https://github.com/golang-jwt/jwt) from 5.2.1 to 5.2.2.
- [Release notes](https://github.com/golang-jwt/jwt/releases)
- [Changelog](https://github.com/golang-jwt/jwt/blob/main/VERSION_HISTORY.md)
- [Commits](https://github.com/golang-jwt/jwt/compare/v5.2.1...v5.2.2)

---
updated-dependencies:
- dependency-name: github.com/golang-jwt/jwt/v5
  dependency-version: 5.2.2
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
2025-04-04 08:35:15 +00:00
Jan Eitzinger
5817b41e29
Merge pull request from ClusterCockpit/dev
Dev
2025-03-20 13:02:23 +01:00
d6b132e3a6 Merge branch 'master' into dev 2025-03-20 12:51:23 +01:00
Jan Eitzinger
318f70f34c
Merge pull request from ClusterCockpit/split_statsTable_query
Split StatsTable DataQuery from JobMetrics Query In Job-View
2025-03-20 12:50:23 +01:00
Jan Eitzinger
e41525d40a
Merge pull request from ClusterCockpit/hotfix
fix: always return hasNextPage boolean to frontend
2025-03-20 12:49:57 +01:00
Jan Eitzinger
a102220e52
Merge pull request from ClusterCockpit/makefile-fix
Fix 'make -B', don't fail if $(VAR) already exists
2025-03-20 12:47:16 +01:00
Christoph Kluge
e9a214c5b2 fix: add nullSafe condition to monitoringStatus display on metric queryError 2025-03-19 14:57:27 +01:00
Christoph Kluge
c53f5eb144 fix: always return hasNextPage boolean to frontend
- removes dependency on uiDefaults setting
2025-03-18 18:01:37 +01:00
Christoph Kluge
9ed64e0388 Review logging, comment cleanup 2025-03-17 17:39:17 +01:00
Christoph Kluge
93040d4629 IMplement LoadNode Data, LoadNodeListData, LoadScopedStats for influxDB2 backend
- Untested
- Only Node Scope
2025-03-17 15:25:33 +01:00
Christoph Kluge
0144ad43f5 Implement NodeListData and ScopedStats for Prometheus Backend 2025-03-17 11:03:51 +01:00
Christoph Kluge
8da2fc30c3 split statsTable data from jobMetrics query, frontend refactor 2025-03-14 16:36:31 +01:00
0e27ae7795 Merge branch 'dev' of github.com:ClusterCockpit/cc-backend into dev 2025-03-14 10:52:39 +01:00
33c6cdb9fe Update test workflow 2025-03-14 10:52:27 +01:00
Christoph Kluge
f5f36427a4 split statsTable data from jobMetrics query, initial commit
- mainly backend changes
- statstable changes only for prototyping
2025-03-13 17:33:55 +01:00
Christoph Kluge
c964f09a4f Merge branch 'dev' into review_logging 2025-02-28 17:19:00 +01:00
Christoph Kluge
bd0cc69668 Review fatalf log calls and messages 2025-02-27 18:10:04 +01:00
Christoph Kluge
84fffac264 Merge branch 'dev' into review_logging 2025-02-27 15:20:46 +01:00
Christoph Kluge
fc0c76bd77 Apply new log funtion to init and main, review or add logtexts 2025-02-26 15:20:25 +01:00
Christoph Kluge
d209547968 Remove dedicated fatal loglevel, change to Fprintln for unformatted 2025-02-26 14:40:54 +01:00
Christoph Kluge
0191bc3821 Annotate and review log functions, add stdout writers 2025-02-25 10:21:48 +01:00
Michael Panzlaff
d61bf212f5 Fix 'make -B', don't fail if $(VAR) already exists 2025-02-03 17:02:13 +01:00
39 changed files with 2395 additions and 1203 deletions

@ -1,331 +0,0 @@
# 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 }}

@ -7,7 +7,7 @@ jobs:
- name: Install Go
uses: actions/setup-go@v4
with:
go-version: 1.22.x
go-version: 1.24.x
- name: Checkout code
uses: actions/checkout@v3
- name: Build, Vet & Test

@ -137,11 +137,6 @@ type JobMetricWithName {
metric: JobMetric!
}
type JobMetricStatWithName {
name: String!
stats: MetricStatistics!
}
type JobMetric {
unit: Unit
timestep: Int!
@ -156,6 +151,30 @@ type Series {
data: [NullableFloat!]!
}
type StatsSeries {
mean: [NullableFloat!]!
median: [NullableFloat!]!
min: [NullableFloat!]!
max: [NullableFloat!]!
}
type JobStatsWithScope {
name: String!
scope: MetricScope!
stats: [ScopedStats!]!
}
type ScopedStats {
hostname: String!
id: String
data: MetricStatistics!
}
type JobStats {
name: String!
stats: MetricStatistics!
}
type Unit {
base: String!
prefix: String
@ -167,13 +186,6 @@ type MetricStatistics {
max: Float!
}
type StatsSeries {
mean: [NullableFloat!]!
median: [NullableFloat!]!
min: [NullableFloat!]!
max: [NullableFloat!]!
}
type MetricFootprints {
metric: String!
data: [NullableFloat!]!
@ -247,7 +259,8 @@ type Query {
job(id: ID!): Job
jobMetrics(id: ID!, metrics: [String!], scopes: [MetricScope!], resolution: Int): [JobMetricWithName!]!
jobMetricStats(id: ID!, metrics: [String!]): [JobMetricStatWithName!]!
jobStats(id: ID!, metrics: [String!]): [JobStats!]!
scopedJobStats(id: ID!, metrics: [String!], scopes: [MetricScope!]): [JobStatsWithScope!]!
jobsFootprints(filter: [JobFilter!], metrics: [String!]!): Footprints
jobs(filter: [JobFilter!], page: PageRequest, order: OrderByInput): JobResultList!

@ -12,7 +12,7 @@ var (
)
func cliInit() {
flag.BoolVar(&flagInit, "init", false, "Setup var directory, initialize swlite database file, config.json and .env")
flag.BoolVar(&flagInit, "init", false, "Setup var directory, initialize sqlite 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")
@ -24,10 +24,10 @@ func cliInit() {
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 user by `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 a existing user. Argument format: <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,warn (default),err,fatal,crit]`")
flag.StringVar(&flagLogLevel, "loglevel", "warn", "Sets the logging level: `[debug, info (default), warn, err, crit]`")
flag.Parse()
}

@ -5,7 +5,6 @@
package main
import (
"fmt"
"os"
"github.com/ClusterCockpit/cc-backend/internal/repository"
@ -62,24 +61,23 @@ const configString = `
func initEnv() {
if util.CheckFileExists("var") {
fmt.Print("Directory ./var already exists. Exiting!\n")
os.Exit(0)
log.Exit("Directory ./var already exists. Cautiously exiting application initialization.")
}
if err := os.WriteFile("config.json", []byte(configString), 0o666); err != nil {
log.Fatalf("Writing config.json failed: %s", err.Error())
log.Abortf("Could not write default ./config.json with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
}
if err := os.WriteFile(".env", []byte(envString), 0o666); err != nil {
log.Fatalf("Writing .env failed: %s", err.Error())
log.Abortf("Could not write default ./.env file with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
}
if err := os.Mkdir("var", 0o777); err != nil {
log.Fatalf("Mkdir var failed: %s", err.Error())
log.Abortf("Could not create default ./var folder with permissions '0o777'. Application initialization failed, exited.\nError: %s\n", err.Error())
}
err := repository.MigrateDB("sqlite3", "./var/job.db")
if err != nil {
log.Fatalf("Initialize job.db failed: %s", err.Error())
log.Abortf("Could not initialize default sqlite3 database as './var/job.db'. Application initialization failed, exited.\nError: %s\n", err.Error())
}
}

@ -61,15 +61,23 @@ func main() {
// Apply config flags for pkg/log
log.Init(flagLogLevel, flagLogDateTime)
// If init flag set, run tasks here before any file dependencies cause errors
if flagInit {
initEnv()
log.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.")
}
// See https://github.com/google/gops (Runtime overhead is almost zero)
if flagGops {
if err := agent.Listen(agent.Options{}); err != nil {
log.Fatalf("gops/agent.Listen failed: %s", err.Error())
log.Abortf("Could not start gops agent with 'gops/agent.Listen(agent.Options{})'. 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())
log.Abortf("Could not parse existing .env file at location './.env'. Application startup failed, exited.\nError: %s\n", err.Error())
}
// Initialize sub-modules and handle command line flags.
@ -87,37 +95,29 @@ func main() {
if flagMigrateDB {
err := repository.MigrateDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
log.Fatal(err)
log.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())
}
os.Exit(0)
log.Exitf("MigrateDB Success: Migrated '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, repository.Version)
}
if flagRevertDB {
err := repository.RevertDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
log.Fatal(err)
log.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())
}
os.Exit(0)
log.Exitf("RevertDB Success: Reverted '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, (repository.Version - 1))
}
if flagForceDB {
err := repository.ForceDB(config.Keys.DBDriver, config.Keys.DB)
if err != nil {
log.Fatal(err)
log.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())
}
os.Exit(0)
log.Exitf("ForceDB Success: Forced '%s' database at location '%s' to version %d.\n", config.Keys.DBDriver, config.Keys.DB, repository.Version)
}
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 {
auth.Init()
@ -125,20 +125,27 @@ func main() {
if flagNewUser != "" {
parts := strings.SplitN(flagNewUser, ":", 3)
if len(parts) != 3 || len(parts[0]) == 0 {
log.Fatal("invalid argument format for user creation")
log.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)
}
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 {
log.Fatalf("adding '%s' user authentication failed: %v", parts[0], err)
log.Abortf("Add User: Could not add new user authentication for '%s' and roles '%s'.\nError: %s\n", parts[0], parts[1], err.Error())
} else {
log.Printf("Add User: Added new user '%s' with roles '%s'.\n", parts[0], parts[1])
}
}
if flagDelUser != "" {
ur := repository.GetUserRepository()
if err := ur.DelUser(flagDelUser); err != nil {
log.Fatalf("deleting user failed: %v", err)
log.Abortf("Delete User: Could not delete user '%s' from DB.\nError: %s\n", flagDelUser, err.Error())
} else {
log.Printf("Delete User: Deleted user '%s' from DB.\n", flagDelUser)
}
}
@ -146,60 +153,64 @@ func main() {
if flagSyncLDAP {
if authHandle.LdapAuth == nil {
log.Fatal("cannot sync: LDAP authentication is not configured")
log.Abort("Sync LDAP: LDAP authentication is not configured, could not synchronize. No changes, exited.")
}
if err := authHandle.LdapAuth.Sync(); err != nil {
log.Fatalf("LDAP sync failed: %v", err)
log.Abortf("Sync LDAP: Could not synchronize, failed with error.\nError: %s\n", err.Error())
}
log.Info("LDAP sync successfull")
log.Print("Sync LDAP: LDAP synchronization successfull.")
}
if flagGenJWT != "" {
ur := repository.GetUserRepository()
user, err := ur.GetUser(flagGenJWT)
if err != nil {
log.Fatalf("could not get user from JWT: %v", err)
log.Abortf("JWT: Could not get supplied user '%s' from DB. No changes, exited.\nError: %s\n", flagGenJWT, err.Error())
}
if !user.HasRole(schema.RoleApi) {
log.Warnf("user '%s' does not have the API role", user.Username)
log.Warnf("JWT: User '%s' does not have the role 'api'. REST API endpoints will return error!\n", user.Username)
}
jwt, err := authHandle.JwtAuth.ProvideJWT(user)
if err != nil {
log.Fatalf("failed to provide JWT to user '%s': %v", user.Username, err)
log.Abortf("JWT: User '%s' found in DB, but failed to provide JWT.\nError: %s\n", user.Username, err.Error())
}
fmt.Printf("MAIN > JWT for '%s': %s\n", user.Username, jwt)
log.Printf("JWT: Successfully generated JWT for user '%s': %s\n", user.Username, jwt)
}
} else if flagNewUser != "" || flagDelUser != "" {
log.Fatal("arguments --add-user and --del-user can only be used if authentication is enabled")
log.Abort("Error: Arguments '--add-user' and '--del-user' can only be used if authentication is enabled. No changes, exited.")
}
if err := archive.Init(config.Keys.Archive, config.Keys.DisableArchive); err != nil {
log.Fatalf("failed to initialize archive: %s", err.Error())
log.Abortf("Init: Failed to initialize archive.\nError: %s\n", err.Error())
}
if err := metricdata.Init(); err != nil {
log.Fatalf("failed to initialize metricdata repository: %s", err.Error())
log.Abortf("Init: Failed to initialize metricdata repository.\nError %s\n", err.Error())
}
if flagReinitDB {
if err := importer.InitDB(); err != nil {
log.Fatalf("failed to re-initialize repository DB: %s", err.Error())
log.Abortf("Init DB: Failed to re-initialize repository DB.\nError: %s\n", err.Error())
} else {
log.Print("Init DB: Sucessfully re-initialized repository DB.")
}
}
if flagImportJob != "" {
if err := importer.HandleImportFlag(flagImportJob); err != nil {
log.Fatalf("job import failed: %s", err.Error())
log.Abortf("Import Job: Job import failed.\nError: %s\n", err.Error())
} else {
log.Printf("Import Job: Imported Job '%s' into DB.\n", flagImportJob)
}
}
if !flagServer {
return
log.Exit("No errors, server flag not set. Exiting cc-backend.")
}
archiver.Start(repository.GetJobRepository())

@ -64,7 +64,7 @@ func serverInit() {
case string:
return fmt.Errorf("MAIN > Panic: %s", e)
case error:
return fmt.Errorf("MAIN > Panic caused by: %w", e)
return fmt.Errorf("MAIN > Panic caused by: %s", e.Error())
}
return errors.New("MAIN > Internal server error (panic)")
@ -268,7 +268,7 @@ func serverStart() {
// Start http or https server
listener, err := net.Listen("tcp", config.Keys.Addr)
if err != nil {
log.Fatalf("starting http listener failed: %v", err)
log.Abortf("Server Start: Starting http listener on '%s' failed.\nError: %s\n", config.Keys.Addr, err.Error())
}
if !strings.HasSuffix(config.Keys.Addr, ":80") && config.Keys.RedirectHttpTo != "" {
@ -281,7 +281,7 @@ func serverStart() {
cert, err := tls.LoadX509KeyPair(
config.Keys.HttpsCertFile, config.Keys.HttpsKeyFile)
if err != nil {
log.Fatalf("loading X509 keypair failed: %v", err)
log.Abortf("Server Start: Loading X509 keypair failed. Check options 'https-cert-file' and 'https-key-file' in 'config.json'.\nError: %s\n", err.Error())
}
listener = tls.NewListener(listener, &tls.Config{
Certificates: []tls.Certificate{cert},
@ -292,20 +292,20 @@ func serverStart() {
MinVersion: tls.VersionTLS12,
PreferServerCipherSuites: true,
})
fmt.Printf("HTTPS server listening at %s...", config.Keys.Addr)
log.Printf("HTTPS server listening at %s...\n", config.Keys.Addr)
} else {
fmt.Printf("HTTP server listening at %s...", config.Keys.Addr)
log.Printf("HTTP server listening at %s...\n", 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 {
log.Fatalf("error while preparing server start: %s", err.Error())
log.Abortf("Server Start: Error while preparing server start.\nError: %s\n", err.Error())
}
if err = server.Serve(listener); err != nil && err != http.ErrServerClosed {
log.Fatalf("starting server failed: %v", err)
log.Abortf("Server Start: Starting server failed.\nError: %s\n", err.Error())
}
}

5
go.mod

@ -1,6 +1,7 @@
module github.com/ClusterCockpit/cc-backend
go 1.23.5
toolchain go1.24.1
require (
github.com/99designs/gqlgen v0.17.66
@ -10,7 +11,7 @@ require (
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 v5.2.1
github.com/golang-jwt/jwt/v5 v5.2.2
github.com/golang-migrate/migrate/v4 v4.18.2
github.com/google/gops v0.3.28
github.com/gorilla/handlers v1.5.2
@ -78,7 +79,7 @@ require (
github.com/xrash/smetrics v0.0.0-20240521201337-686a1a2994c1 // indirect
go.uber.org/atomic v1.11.0 // indirect
golang.org/x/mod v0.23.0 // indirect
golang.org/x/net v0.35.0 // indirect
golang.org/x/net v0.36.0 // 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

8
go.sum

@ -83,8 +83,8 @@ github.com/go-viper/mapstructure/v2 v2.2.1 h1:ZAaOCxANMuZx5RCeg0mBdEZk7DZasvvZIx
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 v5.2.1 h1:OuVbFODueb089Lh128TAcimifWaLhJwVflnrgM17wHk=
github.com/golang-jwt/jwt/v5 v5.2.1/go.mod h1:pqrtFR0X4osieyHYxtmOUWsAWrfe1Q5UVIyoH402zdk=
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/google/go-cmp v0.5.9/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
@ -279,8 +279,8 @@ 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 v0.35.0 h1:T5GQRQb2y08kTAByq9L4/bz8cipCdA8FbRTXewonqY8=
golang.org/x/net v0.35.0/go.mod h1:EglIi67kWsHKlRzzVMUD93VMSWGFOMSZgxFjparz1Qk=
golang.org/x/net v0.36.0 h1:vWF2fRbw4qslQsQzgFqZff+BItCvGFQqKzKIzx1rmoA=
golang.org/x/net v0.36.0/go.mod h1:bFmbeoIPfrw4sMHNhb4J9f6+tPziuGjq7Jk/38fxi1I=
golang.org/x/oauth2 v0.27.0 h1:da9Vo7/tDv5RH/7nZDz1eMGS/q1Vv1N/7FCrBhI9I3M=
golang.org/x/oauth2 v0.27.0/go.mod h1:onh5ek6nERTohokkhCD/y2cV4Do3fxFHFuAejCkRWT8=
golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=

@ -1423,8 +1423,6 @@ func (api *RestApi) updateConfiguration(rw http.ResponseWriter, r *http.Request)
rw.Header().Set("Content-Type", "text/plain")
key, value := r.FormValue("key"), r.FormValue("value")
// fmt.Printf("REST > KEY: %#v\nVALUE: %#v\n", key, value)
if err := repository.GetUserCfgRepo().UpdateConfig(key, value, repository.GetUserFromContext(r.Context())); err != nil {
http.Error(rw, err.Error(), http.StatusUnprocessableEntity)
return

@ -7,9 +7,9 @@ package config
import (
"bytes"
"encoding/json"
"log"
"os"
"github.com/ClusterCockpit/cc-backend/pkg/log"
"github.com/ClusterCockpit/cc-backend/pkg/schema"
)
@ -53,20 +53,20 @@ func Init(flagConfigFile string) {
raw, err := os.ReadFile(flagConfigFile)
if err != nil {
if !os.IsNotExist(err) {
log.Fatalf("CONFIG ERROR: %v", err)
log.Abortf("Config Init: Could not read config file '%s'.\nError: %s\n", flagConfigFile, err.Error())
}
} else {
if err := schema.Validate(schema.Config, bytes.NewReader(raw)); err != nil {
log.Fatalf("Validate config: %v\n", err)
log.Abortf("Config Init: Could not validate config file '%s'.\nError: %s\n", flagConfigFile, err.Error())
}
dec := json.NewDecoder(bytes.NewReader(raw))
dec.DisallowUnknownFields()
if err := dec.Decode(&Keys); err != nil {
log.Fatalf("could not decode: %v", err)
log.Abortf("Config Init: Could not decode config file '%s'.\nError: %s\n", flagConfigFile, err.Error())
}
if Keys.Clusters == nil || len(Keys.Clusters) < 1 {
log.Fatal("At least one cluster required in config!")
log.Abort("Config Init: At least one cluster required in config. Exited with error.")
}
}
}

File diff suppressed because it is too large Load Diff

@ -81,11 +81,6 @@ type JobLinkResultList struct {
Count *int `json:"count,omitempty"`
}
type JobMetricStatWithName struct {
Name string `json:"name"`
Stats *schema.MetricStatistics `json:"stats"`
}
type JobMetricWithName struct {
Name string `json:"name"`
Scope schema.MetricScope `json:"scope"`
@ -100,6 +95,17 @@ type JobResultList struct {
HasNextPage *bool `json:"hasNextPage,omitempty"`
}
type JobStats struct {
Name string `json:"name"`
Stats *schema.MetricStatistics `json:"stats"`
}
type JobStatsWithScope struct {
Name string `json:"name"`
Scope schema.MetricScope `json:"scope"`
Stats []*ScopedStats `json:"stats"`
}
type JobsStatistics struct {
ID string `json:"id"`
Name string `json:"name"`
@ -173,6 +179,12 @@ 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 StringInput struct {
Eq *string `json:"eq,omitempty"`
Neq *string `json:"neq,omitempty"`

@ -301,24 +301,23 @@ func (r *queryResolver) JobMetrics(ctx context.Context, id string, metrics []str
return res, err
}
// JobMetricStats is the resolver for the jobMetricStats field.
func (r *queryResolver) JobMetricStats(ctx context.Context, id string, metrics []string) ([]*model.JobMetricStatWithName, error) {
// JobStats is the resolver for the jobStats field.
func (r *queryResolver) JobStats(ctx context.Context, id string, metrics []string) ([]*model.JobStats, error) {
job, err := r.Query().Job(ctx, id)
if err != nil {
log.Warn("Error while querying job for metrics")
log.Warnf("Error while querying job %s for metadata", id)
return nil, err
}
data, err := metricDataDispatcher.LoadStatData(job, metrics, ctx)
data, err := metricDataDispatcher.LoadJobStats(job, metrics, ctx)
if err != nil {
log.Warn("Error while loading job stat data")
log.Warnf("Error while loading jobStats data for job id %s", id)
return nil, err
}
res := []*model.JobMetricStatWithName{}
res := []*model.JobStats{}
for name, md := range data {
res = append(res, &model.JobMetricStatWithName{
res = append(res, &model.JobStats{
Name: name,
Stats: &md,
})
@ -327,6 +326,44 @@ func (r *queryResolver) JobMetricStats(ctx context.Context, id string, metrics [
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.JobStatsWithScope, error) {
job, err := r.Query().Job(ctx, id)
if err != nil {
log.Warnf("Error while querying job %s for metadata", id)
return nil, err
}
data, err := metricDataDispatcher.LoadScopedJobStats(job, metrics, scopes, ctx)
if err != nil {
log.Warnf("Error while loading scopedJobStats data for job id %s", id)
return nil, err
}
res := make([]*model.JobStatsWithScope, 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.JobStatsWithScope{
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) {
// NOTE: Legacy Naming! This resolver is for normalized histograms in analysis view only - *Not* related to DB "footprint" column!
@ -354,30 +391,28 @@ func (r *queryResolver) Jobs(ctx context.Context, filter []*model.JobFilter, pag
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),
}
nextJobs, err := r.Repo.QueryJobs(ctx, filter, nextPage, order)
if err != nil {
log.Warn("Error while querying next jobs")
return nil, err
}
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
// 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 {
log.Warn("Error while querying next jobs")
return nil, err
}
hasNextPage := false
if len(nextJobs) == 1 {
hasNextPage = true
}
return &model.JobResultList{Items: jobs, Count: &count, HasNextPage: &hasNextPage}, nil
}
// JobsStatistics is the resolver for the jobsStatistics field.

@ -224,8 +224,34 @@ func LoadAverages(
return nil
}
// Used for polar plots in frontend
func LoadStatData(
// 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 {
log.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,
@ -237,12 +263,12 @@ func LoadStatData(
data := make(map[string]schema.MetricStatistics, len(metrics))
repo, err := metricdata.GetMetricDataRepo(job.Cluster)
if err != nil {
return data, fmt.Errorf("METRICDATA/METRICDATA > no metric data repository configured for '%s'", job.Cluster)
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 {
log.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 %d (User %s, Project %s)", job.JobID, job.User, job.Project)
return data, err
}

@ -129,13 +129,13 @@ func (ccms *CCMetricStore) doRequest(
) (*ApiQueryResponse, error) {
buf := &bytes.Buffer{}
if err := json.NewEncoder(buf).Encode(body); err != nil {
log.Warn("Error while encoding request body")
log.Errorf("Error while encoding request body: %s", err.Error())
return nil, err
}
req, err := http.NewRequestWithContext(ctx, http.MethodGet, ccms.queryEndpoint, buf)
if err != nil {
log.Warn("Error while building request body")
log.Errorf("Error while building request body: %s", err.Error())
return nil, err
}
if ccms.jwt != "" {
@ -151,7 +151,7 @@ func (ccms *CCMetricStore) doRequest(
res, err := ccms.client.Do(req)
if err != nil {
log.Error("Error while performing request")
log.Errorf("Error while performing request: %s", err.Error())
return nil, err
}
@ -161,7 +161,7 @@ func (ccms *CCMetricStore) doRequest(
var resBody ApiQueryResponse
if err := json.NewDecoder(bufio.NewReader(res.Body)).Decode(&resBody); err != nil {
log.Warn("Error while decoding result body")
log.Errorf("Error while decoding result body: %s", err.Error())
return nil, err
}
@ -177,7 +177,7 @@ func (ccms *CCMetricStore) LoadData(
) (schema.JobData, error) {
queries, assignedScope, err := ccms.buildQueries(job, metrics, scopes, resolution)
if err != nil {
log.Warn("Error while building queries")
log.Errorf("Error while building queries for jobId %d, Metrics %v, Scopes %v: %s", job.JobID, metrics, scopes, err.Error())
return nil, err
}
@ -192,7 +192,7 @@ func (ccms *CCMetricStore) LoadData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
log.Error("Error while performing request")
log.Errorf("Error while performing request: %s", err.Error())
return nil, err
}
@ -557,16 +557,9 @@ 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 {
log.Warn("Error while building query")
log.Errorf("Error while building queries for jobId %d, Metrics %v: %s", job.JobID, metrics, err.Error())
return nil, err
}
@ -581,7 +574,7 @@ func (ccms *CCMetricStore) LoadStats(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
log.Error("Error while performing request")
log.Errorf("Error while performing request: %s", err.Error())
return nil, err
}
@ -591,9 +584,8 @@ func (ccms *CCMetricStore) LoadStats(
metric := ccms.toLocalName(query.Metric)
data := res[0]
if data.Error != nil {
log.Infof("fetching %s for node %s failed: %s", metric, query.Hostname, *data.Error)
log.Errorf("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]
@ -603,9 +595,8 @@ func (ccms *CCMetricStore) LoadStats(
}
if data.Avg.IsNaN() || data.Min.IsNaN() || data.Max.IsNaN() {
log.Infof("fetching %s for node %s failed: one of avg/min/max is NaN", metric, query.Hostname)
log.Warnf("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{
@ -618,7 +609,98 @@ func (ccms *CCMetricStore) LoadStats(
return stats, nil
}
// TODO: Support sub-node-scope metrics! For this, the partition of a node needs to be known!
// 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 {
log.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.Unix(),
To: job.StartTime.Add(time.Duration(job.Duration) * time.Second).Unix(),
Queries: queries,
WithStats: true,
WithData: false,
}
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
log.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
func (ccms *CCMetricStore) LoadNodeData(
cluster string,
metrics, nodes []string,
@ -652,7 +734,7 @@ func (ccms *CCMetricStore) LoadNodeData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
log.Error(fmt.Sprintf("Error while performing request %#v\n", err))
log.Errorf("Error while performing request: %s", err.Error())
return nil, err
}
@ -710,6 +792,7 @@ func (ccms *CCMetricStore) LoadNodeData(
return data, nil
}
// Used for Systems-View Node-List
func (ccms *CCMetricStore) LoadNodeListData(
cluster, subCluster, nodeFilter string,
metrics []string,
@ -768,7 +851,7 @@ func (ccms *CCMetricStore) LoadNodeListData(
queries, assignedScope, err := ccms.buildNodeQueries(cluster, subCluster, nodes, metrics, scopes, resolution)
if err != nil {
log.Warn("Error while building queries")
log.Errorf("Error while building node queries for Cluster %s, SubCLuster %s, Metrics %v, Scopes %v: %s", cluster, subCluster, metrics, scopes, err.Error())
return nil, totalNodes, hasNextPage, err
}
@ -783,7 +866,7 @@ func (ccms *CCMetricStore) LoadNodeListData(
resBody, err := ccms.doRequest(ctx, &req)
if err != nil {
log.Error(fmt.Sprintf("Error while performing request %#v\n", err))
log.Errorf("Error while performing request: %s", err.Error())
return nil, totalNodes, hasNextPage, err
}
@ -888,7 +971,7 @@ func (ccms *CCMetricStore) buildNodeQueries(
if subCluster != "" {
subClusterTopol, scterr = archive.GetSubCluster(cluster, subCluster)
if scterr != nil {
// TODO: Log
log.Errorf("could not load cluster %s subCluster %s topology: %s", cluster, subCluster, scterr.Error())
return nil, nil, scterr
}
}
@ -898,7 +981,7 @@ 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)
log.Infof("metric '%s' is not specified for cluster '%s'", metric, cluster)
log.Warnf("metric '%s' is not specified for cluster '%s'", metric, cluster)
continue
}

@ -10,6 +10,8 @@ import (
"encoding/json"
"errors"
"fmt"
"math"
"sort"
"strings"
"time"
@ -64,6 +66,8 @@ func (idb *InfluxDBv2DataRepository) LoadData(
ctx context.Context,
resolution int) (schema.JobData, error) {
log.Infof("InfluxDB 2 Backend: Resolution Scaling not Implemented, will return default timestep. Requested Resolution %d", resolution)
measurementsConds := make([]string, 0, len(metrics))
for _, m := range metrics {
measurementsConds = append(measurementsConds, fmt.Sprintf(`r["_measurement"] == "%s"`, m))
@ -86,7 +90,7 @@ func (idb *InfluxDBv2DataRepository) LoadData(
query := ""
switch scope {
case "node":
// Get Finest Granularity, Groupy By Measurement and Hostname (== Metric / Node), Calculate Mean for 60s windows
// Get Finest Granularity, Groupy By Measurement and Hostname (== Metric / Node), Calculate Mean for 60s windows <-- Resolution could be added here?
// log.Info("Scope 'node' requested. ")
query = fmt.Sprintf(`
from(bucket: "%s")
@ -116,6 +120,12 @@ func (idb *InfluxDBv2DataRepository) LoadData(
// idb.bucket,
// idb.formatTime(job.StartTime), idb.formatTime(idb.epochToTime(job.StartTimeUnix + int64(job.Duration) + int64(1) )),
// measurementsCond, hostsCond)
case "hwthread":
log.Info(" Scope 'hwthread' requested, but not yet supported: Will return 'node' scope only. ")
continue
case "accelerator":
log.Info(" Scope 'accelerator' requested, but not yet supported: Will return 'node' scope only. ")
continue
default:
log.Infof("Unknown scope '%s' requested: Will return 'node' scope.", scope)
continue
@ -173,6 +183,11 @@ func (idb *InfluxDBv2DataRepository) LoadData(
}
case "socket":
continue
case "accelerator":
continue
case "hwthread":
// See below @ core
continue
case "core":
continue
// Include Series.Id in hostSeries
@ -301,6 +316,53 @@ func (idb *InfluxDBv2DataRepository) LoadStats(
return stats, nil
}
// Used in Job-View StatsTable
// UNTESTED
func (idb *InfluxDBv2DataRepository) LoadScopedStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context) (schema.ScopedJobStats, error) {
// Assumption: idb.loadData() only returns series node-scope - use node scope for statsTable
scopedJobStats := make(schema.ScopedJobStats)
data, err := idb.LoadData(job, metrics, []schema.MetricScope{schema.MetricScopeNode}, ctx, 0 /*resolution here*/)
if err != nil {
log.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() {
log.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
}
// Used in Systems-View @ Node-Overview
// UNTESTED
func (idb *InfluxDBv2DataRepository) LoadNodeData(
cluster string,
metrics, nodes []string,
@ -308,12 +370,123 @@ func (idb *InfluxDBv2DataRepository) LoadNodeData(
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)
// Note: scopes[] Array will be ignored, only return node scope
return nil, errors.New("METRICDATA/INFLUXV2 > unimplemented for InfluxDBv2DataRepository")
// CONVERT ARGS TO INFLUX
measurementsConds := make([]string, 0)
for _, m := range metrics {
measurementsConds = append(measurementsConds, fmt.Sprintf(`r["_measurement"] == "%s"`, m))
}
measurementsCond := strings.Join(measurementsConds, " or ")
hostsConds := make([]string, 0)
if nodes == nil {
var allNodes []string
subClusterNodeLists := archive.NodeLists[cluster]
for _, nodeList := range subClusterNodeLists {
allNodes = append(nodes, nodeList.PrintList()...)
}
for _, node := range allNodes {
nodes = append(nodes, node)
hostsConds = append(hostsConds, fmt.Sprintf(`r["hostname"] == "%s"`, node))
}
} else {
for _, node := range nodes {
hostsConds = append(hostsConds, fmt.Sprintf(`r["hostname"] == "%s"`, node))
}
}
hostsCond := strings.Join(hostsConds, " or ")
// BUILD AND PERFORM QUERY
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(from), idb.formatTime(to),
measurementsCond, hostsCond)
rows, err := idb.queryClient.Query(ctx, query)
if err != nil {
log.Error("Error while performing query")
return nil, err
}
// HANDLE QUERY RETURN
// Collect Float Arrays for Node@Metric -> No Scope Handling!
influxData := make(map[string]map[string][]schema.Float)
for rows.Next() {
row := rows.Record()
host, field := row.ValueByKey("hostname").(string), row.Measurement()
influxHostData, ok := influxData[host]
if !ok {
influxHostData = make(map[string][]schema.Float)
influxData[host] = influxHostData
}
influxFieldData, ok := influxData[host][field]
if !ok {
influxFieldData = make([]schema.Float, 0)
influxData[host][field] = influxFieldData
}
val, ok := row.Value().(float64)
if ok {
influxData[host][field] = append(influxData[host][field], schema.Float(val))
} else {
influxData[host][field] = append(influxData[host][field], schema.Float(0))
}
}
// BUILD FUNCTION RETURN
data := make(map[string]map[string][]*schema.JobMetric)
for node, metricData := range influxData {
nodeData, ok := data[node]
if !ok {
nodeData = make(map[string][]*schema.JobMetric)
data[node] = nodeData
}
for metric, floatArray := range metricData {
avg, min, max := 0.0, 0.0, 0.0
for _, val := range floatArray {
avg += float64(val)
min = math.Min(min, float64(val))
max = math.Max(max, float64(val))
}
stats := schema.MetricStatistics{
Avg: (math.Round((avg/float64(len(floatArray)))*100) / 100),
Min: (math.Round(min*100) / 100),
Max: (math.Round(max*100) / 100),
}
mc := archive.GetMetricConfig(cluster, metric)
nodeData[metric] = append(nodeData[metric], &schema.JobMetric{
Unit: mc.Unit,
Timestep: mc.Timestep,
Series: []schema.Series{
{
Hostname: node,
Statistics: stats,
Data: floatArray,
},
},
})
}
}
return data, nil
}
// Used in Systems-View @ Node-List
// UNTESTED
func (idb *InfluxDBv2DataRepository) LoadNodeListData(
cluster, subCluster, nodeFilter string,
metrics []string,
@ -324,10 +497,79 @@ func (idb *InfluxDBv2DataRepository) LoadNodeListData(
ctx context.Context,
) (map[string]schema.JobData, int, bool, error) {
// Assumption: idb.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 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")
// 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 And Convert Data, use idb.LoadNodeData() for query
rawNodeData, err := idb.LoadNodeData(cluster, metrics, nodes, scopes, from, to, ctx)
if err != nil {
log.Error(fmt.Sprintf("Error while loading influx nodeData for nodeListData %#v\n", err))
return nil, totalNodes, hasNextPage, err
}
data := make(map[string]schema.JobData)
for node, nodeData := range rawNodeData {
// Init Nested Map Data Structures If Not Found
hostData, ok := data[node]
if !ok {
hostData = make(schema.JobData)
data[node] = hostData
}
for metric, nodeMetricData := range nodeData {
metricData, ok := hostData[metric]
if !ok {
metricData = make(map[schema.MetricScope]*schema.JobMetric)
data[node][metric] = metricData
}
data[node][metric][schema.MetricScopeNode] = nodeMetricData[0] // Only Node Scope Returned from loadNodeData
}
}
return data, totalNodes, hasNextPage, nil
}

@ -24,9 +24,12 @@ 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 assumed for now.
// Return a map of metrics to a map of nodes to the metric statistics of the job. node scope only.
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)

@ -448,6 +448,51 @@ func (pdb *PrometheusDataRepository) LoadNodeData(
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 {
log.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() {
log.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,
@ -458,10 +503,132 @@ func (pdb *PrometheusDataRepository) LoadNodeListData(
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)
return nil, totalNodes, hasNextPage, errors.New("METRICDATA/INFLUXV2 > unimplemented for PrometheusDataRepository")
// 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() {
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 {
log.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 {
log.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 {
log.Errorf("Prometheus query error in LoadNodeData: %v\n", err)
return nil, totalNodes, hasNextPage, errors.New("Prometheus query error")
}
if len(warnings) > 0 {
log.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)
log.Debugf("LoadNodeListData of %v nodes took %s", len(data), t1)
return data, totalNodes, hasNextPage, nil
}

@ -36,7 +36,17 @@ func (tmdr *TestMetricDataRepository) LoadData(
func (tmdr *TestMetricDataRepository) LoadStats(
job *schema.Job,
metrics []string, ctx context.Context) (map[string]map[string]schema.MetricStatistics, error) {
metrics []string,
ctx context.Context) (map[string]map[string]schema.MetricStatistics, error) {
panic("TODO")
}
func (tmdr *TestMetricDataRepository) LoadScopedStats(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
ctx context.Context) (schema.ScopedJobStats, error) {
panic("TODO")
}

@ -59,17 +59,15 @@ func Connect(driver string, db string) {
} else {
dbHandle, err = sqlx.Open("sqlite3", opts.URL)
}
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:
log.Fatalf("unsupported database driver: %s", driver)
log.Abortf("DB Connection: Unsupported database driver '%s'.\n", driver)
}
if err != nil {
log.Abortf("DB Connection: Could not connect to '%s' database with sqlx.Open().\nError: %s\n", driver, err.Error())
}
dbHandle.SetMaxOpenConns(opts.MaxOpenConnections)
@ -80,7 +78,7 @@ func Connect(driver string, db string) {
dbConnInstance = &DBConnection{DB: dbHandle, Driver: driver}
err = checkDBVersion(driver, dbHandle.DB)
if err != nil {
log.Fatal(err)
log.Abortf("DB Connection: Failed DB version check.\nError: %s\n", err.Error())
}
})
}

@ -54,7 +54,7 @@ func checkDBVersion(backend string, db *sql.DB) error {
return err
}
default:
log.Fatalf("unsupported database backend: %s", backend)
log.Abortf("Migration: Unsupported database backend '%s'.\n", backend)
}
v, dirty, err := m.Version()
@ -102,7 +102,7 @@ func getMigrateInstance(backend string, db string) (m *migrate.Migrate, err erro
return m, err
}
default:
log.Fatalf("unsupported database backend: %s", backend)
log.Abortf("Migration: Unsupported database backend '%s'.\n", backend)
}
return m, nil

@ -35,7 +35,7 @@ func GetUserCfgRepo() *UserCfgRepo {
lookupConfigStmt, err := db.DB.Preparex(`SELECT confkey, value FROM configuration WHERE configuration.username = ?`)
if err != nil {
log.Fatalf("db.DB.Preparex() error: %v", err)
log.Fatalf("User Config: Call 'db.DB.Preparex()' failed.\nError: %s\n", err.Error())
}
userCfgRepoInstance = &UserCfgRepo{

@ -40,7 +40,7 @@ func Start() {
jobRepo = repository.GetJobRepository()
s, err = gocron.NewScheduler()
if err != nil {
log.Fatalf("Error while creating gocron scheduler: %s", err.Error())
log.Abortf("Taskmanager Start: Could not create gocron scheduler.\nError: %s\n", err.Error())
}
if config.Keys.StopJobsExceedingWalltime > 0 {

@ -27,6 +27,8 @@ type ArchiveBackend interface {
LoadJobData(job *schema.Job) (schema.JobData, error)
LoadJobStats(job *schema.Job) (schema.ScopedJobStats, error)
LoadClusterCfg(name string) (*schema.Cluster, error)
StoreJobMeta(jobMeta *schema.JobMeta) error
@ -87,7 +89,7 @@ func Init(rawConfig json.RawMessage, disableArchive bool) error {
var version uint64
version, err = ar.Init(rawConfig)
if err != nil {
log.Error("Error while initializing archiveBackend")
log.Errorf("Error while initializing archiveBackend: %s", err.Error())
return
}
log.Infof("Load archive version %d", version)
@ -110,7 +112,7 @@ func LoadAveragesFromArchive(
) error {
metaFile, err := ar.LoadJobMeta(job)
if err != nil {
log.Warn("Error while loading job metadata from archiveBackend")
log.Errorf("Error while loading job metadata from archiveBackend: %s", err.Error())
return err
}
@ -125,7 +127,7 @@ func LoadAveragesFromArchive(
return nil
}
// Helper to metricdataloader.LoadStatData().
// Helper to metricdataloader.LoadJobStats().
func LoadStatsFromArchive(
job *schema.Job,
metrics []string,
@ -133,7 +135,7 @@ func LoadStatsFromArchive(
data := make(map[string]schema.MetricStatistics, len(metrics))
metaFile, err := ar.LoadJobMeta(job)
if err != nil {
log.Warn("Error while loading job metadata from archiveBackend")
log.Errorf("Error while loading job metadata from archiveBackend: %s", err.Error())
return data, err
}
@ -154,10 +156,26 @@ func LoadStatsFromArchive(
return data, nil
}
// Helper to metricdataloader.LoadScopedJobStats().
func LoadScopedStatsFromArchive(
job *schema.Job,
metrics []string,
scopes []schema.MetricScope,
) (schema.ScopedJobStats, error) {
data, err := ar.LoadJobStats(job)
if err != nil {
log.Errorf("Error while loading job stats from archiveBackend: %s", err.Error())
return nil, err
}
return data, nil
}
func GetStatistics(job *schema.Job) (map[string]schema.JobStatistics, error) {
metaFile, err := ar.LoadJobMeta(job)
if err != nil {
log.Warn("Error while loading job metadata from archiveBackend")
log.Errorf("Error while loading job metadata from archiveBackend: %s", err.Error())
return nil, err
}
@ -173,7 +191,7 @@ func UpdateMetadata(job *schema.Job, metadata map[string]string) error {
jobMeta, err := ar.LoadJobMeta(job)
if err != nil {
log.Warn("Error while loading job metadata from archiveBackend")
log.Errorf("Error while loading job metadata from archiveBackend: %s", err.Error())
return err
}
@ -193,7 +211,7 @@ func UpdateTags(job *schema.Job, tags []*schema.Tag) error {
jobMeta, err := ar.LoadJobMeta(job)
if err != nil {
log.Warn("Error while loading job metadata from archiveBackend")
log.Errorf("Error while loading job metadata from archiveBackend: %s", err.Error())
return err
}

@ -115,6 +115,40 @@ func loadJobData(filename string, isCompressed bool) (schema.JobData, error) {
}
}
func loadJobStats(filename string, isCompressed bool) (schema.ScopedJobStats, error) {
f, err := os.Open(filename)
if err != nil {
log.Errorf("fsBackend LoadJobStats()- %v", err)
return nil, err
}
defer f.Close()
if isCompressed {
r, err := gzip.NewReader(f)
if err != nil {
log.Errorf(" %v", err)
return nil, err
}
defer r.Close()
if config.Keys.Validate {
if err := schema.Validate(schema.Data, r); err != nil {
return nil, fmt.Errorf("validate job data: %v", err)
}
}
return DecodeJobStats(r, filename)
} else {
if config.Keys.Validate {
if err := schema.Validate(schema.Data, bufio.NewReader(f)); err != nil {
return nil, fmt.Errorf("validate job data: %v", err)
}
}
return DecodeJobStats(bufio.NewReader(f), filename)
}
}
func (fsa *FsArchive) Init(rawConfig json.RawMessage) (uint64, error) {
var config FsArchiveConfig
@ -389,6 +423,18 @@ func (fsa *FsArchive) LoadJobData(job *schema.Job) (schema.JobData, error) {
return loadJobData(filename, isCompressed)
}
func (fsa *FsArchive) LoadJobStats(job *schema.Job) (schema.ScopedJobStats, error) {
var isCompressed bool = true
filename := getPath(job, fsa.path, "data.json.gz")
if !util.CheckFileExists(filename) {
filename = getPath(job, fsa.path, "data.json")
isCompressed = false
}
return loadJobStats(filename, isCompressed)
}
func (fsa *FsArchive) LoadJobMeta(job *schema.Job) (*schema.JobMeta, error) {
filename := getPath(job, fsa.path, "meta.json")
return loadJobMeta(filename)

@ -32,6 +32,43 @@ func DecodeJobData(r io.Reader, k string) (schema.JobData, error) {
return data.(schema.JobData), nil
}
func DecodeJobStats(r io.Reader, k string) (schema.ScopedJobStats, error) {
jobData, err := DecodeJobData(r, k)
// Convert schema.JobData to schema.ScopedJobStats
if jobData != nil {
scopedJobStats := make(schema.ScopedJobStats)
for metric, metricData := range jobData {
if _, ok := scopedJobStats[metric]; !ok {
scopedJobStats[metric] = make(map[schema.MetricScope][]*schema.ScopedStats)
}
for scope, jobMetric := range metricData {
if _, ok := scopedJobStats[metric][scope]; !ok {
scopedJobStats[metric][scope] = make([]*schema.ScopedStats, 0)
}
for _, series := range jobMetric.Series {
scopedJobStats[metric][scope] = append(scopedJobStats[metric][scope], &schema.ScopedStats{
Hostname: series.Hostname,
Id: series.Id,
Data: &series.Statistics,
})
}
// 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)
}
}
}
}
return scopedJobStats, nil
}
return nil, err
}
func DecodeJobMeta(r io.Reader) (*schema.JobMeta, error) {
var d schema.JobMeta
if err := json.NewDecoder(r).Decode(&d); err != nil {

@ -46,12 +46,12 @@ var loglevel string = "info"
/* CONFIG */
func Init(lvl string, logdate bool) {
// Discard I/O for all writers below selected loglevel; <CRITICAL> is always written.
switch lvl {
case "crit":
ErrWriter = io.Discard
fallthrough
case "err", "fatal":
case "err":
WarnWriter = io.Discard
fallthrough
case "warn":
@ -63,8 +63,7 @@ func Init(lvl string, logdate bool) {
// Nothing to do...
break
default:
fmt.Printf("pkg/log: Flag 'loglevel' has invalid value %#v\npkg/log: Will use default loglevel 'debug'\n", lvl)
//SetLogLevel("debug")
fmt.Printf("pkg/log: Flag 'loglevel' has invalid value %#v\npkg/log: Will use default loglevel '%s'\n", lvl, loglevel)
}
if !logdate {
@ -84,109 +83,138 @@ func Init(lvl string, logdate bool) {
loglevel = lvl
}
/* PRINT */
// Private helper
func printStr(v ...interface{}) string {
return fmt.Sprint(v...)
}
// Uses Info() -> If errorpath required at some point:
// Will need own writer with 'Output(2, out)' to correctly render path
func Print(v ...interface{}) {
Info(v...)
}
func Debug(v ...interface{}) {
DebugLog.Output(2, printStr(v...))
}
func Info(v ...interface{}) {
InfoLog.Output(2, printStr(v...))
}
func Warn(v ...interface{}) {
WarnLog.Output(2, printStr(v...))
}
func Error(v ...interface{}) {
ErrLog.Output(2, printStr(v...))
}
// Writes panic stacktrace, but keeps application alive
func Panic(v ...interface{}) {
ErrLog.Output(2, printStr(v...))
panic("Panic triggered ...")
}
func Crit(v ...interface{}) {
CritLog.Output(2, printStr(v...))
}
// Writes critical log, stops application
func Fatal(v ...interface{}) {
CritLog.Output(2, printStr(v...))
os.Exit(1)
}
/* PRINT FORMAT*/
// Private helper
func printfStr(format string, v ...interface{}) string {
return fmt.Sprintf(format, v...)
}
// Uses Infof() -> If errorpath required at some point:
// Will need own writer with 'Output(2, out)' to correctly render path
func Printf(format string, v ...interface{}) {
Infof(format, v...)
}
func Debugf(format string, v ...interface{}) {
DebugLog.Output(2, printfStr(format, v...))
}
func Infof(format string, v ...interface{}) {
InfoLog.Output(2, printfStr(format, v...))
}
func Warnf(format string, v ...interface{}) {
WarnLog.Output(2, printfStr(format, v...))
}
func Errorf(format string, v ...interface{}) {
ErrLog.Output(2, printfStr(format, v...))
}
// Writes panic stacktrace, but keeps application alive
func Panicf(format string, v ...interface{}) {
ErrLog.Output(2, printfStr(format, v...))
panic("Panic triggered ...")
}
func Critf(format string, v ...interface{}) {
CritLog.Output(2, printfStr(format, v...))
}
// Writes crit log, stops application
func Fatalf(format string, v ...interface{}) {
CritLog.Output(2, printfStr(format, v...))
os.Exit(1)
}
/* HELPER */
func Loglevel() string {
return loglevel
}
/* SPECIAL */
/* PRIVATE HELPER */
// func Finfof(w io.Writer, format string, v ...interface{}) {
// if w != io.Discard {
// if logDateTime {
// currentTime := time.Now()
// fmt.Fprintf(InfoWriter, currentTime.String()+InfoPrefix+format+"\n", v...)
// } else {
// fmt.Fprintf(InfoWriter, InfoPrefix+format+"\n", v...)
// }
// }
// }
// Return unformatted string
func printStr(v ...interface{}) string {
return fmt.Sprint(v...)
}
// Return formatted string
func printfStr(format string, v ...interface{}) string {
return fmt.Sprintf(format, v...)
}
/* PRINT */
// Prints to STDOUT without string formatting; application continues.
// Used for special cases not requiring log information like date or location.
func Print(v ...interface{}) {
fmt.Fprintln(os.Stdout, v...)
}
// Prints to STDOUT without string formatting; application exits with error code 0.
// Used for exiting succesfully with message after expected outcome, e.g. successful single-call application runs.
func Exit(v ...interface{}) {
fmt.Fprintln(os.Stdout, v...)
os.Exit(0)
}
// Prints to STDOUT without string formatting; application exits with error code 1.
// Used for terminating with message after to be expected errors, e.g. wrong arguments or during init().
func Abort(v ...interface{}) {
fmt.Fprintln(os.Stdout, v...)
os.Exit(1)
}
// Prints to DEBUG writer without string formatting; application continues.
// Used for logging additional information, primarily for development.
func Debug(v ...interface{}) {
DebugLog.Output(2, printStr(v...))
}
// Prints to INFO writer without string formatting; application continues.
// Used for logging additional information, e.g. notable returns or common fail-cases.
func Info(v ...interface{}) {
InfoLog.Output(2, printStr(v...))
}
// Prints to WARNING writer without string formatting; application continues.
// Used for logging important information, e.g. uncommon edge-cases or administration related information.
func Warn(v ...interface{}) {
WarnLog.Output(2, printStr(v...))
}
// Prints to ERROR writer without string formatting; application continues.
// Used for logging errors, but code still can return default(s) or nil.
func Error(v ...interface{}) {
ErrLog.Output(2, printStr(v...))
}
// Prints to CRITICAL writer without string formatting; application exits with error code 1.
// Used for terminating on unexpected errors with date and code location.
func Fatal(v ...interface{}) {
CritLog.Output(2, printStr(v...))
os.Exit(1)
}
// Prints to PANIC function without string formatting; application exits with panic.
// Used for terminating on unexpected errors with stacktrace.
func Panic(v ...interface{}) {
panic(printStr(v...))
}
/* PRINT FORMAT*/
// Prints to STDOUT with string formatting; application continues.
// Used for special cases not requiring log information like date or location.
func Printf(format string, v ...interface{}) {
fmt.Fprintf(os.Stdout, format, v...)
}
// Prints to STDOUT with string formatting; application exits with error code 0.
// Used for exiting succesfully with message after expected outcome, e.g. successful single-call application runs.
func Exitf(format string, v ...interface{}) {
fmt.Fprintf(os.Stdout, format, v...)
os.Exit(0)
}
// Prints to STDOUT with string formatting; application exits with error code 1.
// Used for terminating with message after to be expected errors, e.g. wrong arguments or during init().
func Abortf(format string, v ...interface{}) {
fmt.Fprintf(os.Stdout, format, v...)
os.Exit(1)
}
// Prints to DEBUG writer with string formatting; application continues.
// Used for logging additional information, primarily for development.
func Debugf(format string, v ...interface{}) {
DebugLog.Output(2, printfStr(format, v...))
}
// Prints to INFO writer with string formatting; application continues.
// Used for logging additional information, e.g. notable returns or common fail-cases.
func Infof(format string, v ...interface{}) {
InfoLog.Output(2, printfStr(format, v...))
}
// Prints to WARNING writer with string formatting; application continues.
// Used for logging important information, e.g. uncommon edge-cases or administration related information.
func Warnf(format string, v ...interface{}) {
WarnLog.Output(2, printfStr(format, v...))
}
// Prints to ERROR writer with string formatting; application continues.
// Used for logging errors, but code still can return default(s) or nil.
func Errorf(format string, v ...interface{}) {
ErrLog.Output(2, printfStr(format, v...))
}
// Prints to CRITICAL writer with string formatting; application exits with error code 1.
// Used for terminating on unexpected errors with date and code location.
func Fatalf(format string, v ...interface{}) {
CritLog.Output(2, printfStr(format, v...))
os.Exit(1)
}
// Prints to PANIC function with string formatting; application exits with panic.
// Used for terminating on unexpected errors with stacktrace.
func Panicf(format string, v ...interface{}) {
panic(printfStr(format, v...))
}

@ -15,6 +15,7 @@ import (
)
type JobData map[string]map[MetricScope]*JobMetric
type ScopedJobStats map[string]map[MetricScope][]*ScopedStats
type JobMetric struct {
StatisticsSeries *StatsSeries `json:"statisticsSeries,omitempty"`
@ -30,6 +31,12 @@ type Series struct {
Statistics MetricStatistics `json:"statistics"`
}
type ScopedStats struct {
Hostname string `json:"hostname"`
Id *string `json:"id,omitempty"`
Data *MetricStatistics `json:"data"`
}
type MetricStatistics struct {
Avg float64 `json:"avg"`
Min float64 `json:"min"`

@ -22,8 +22,7 @@ func parseDate(in string) int64 {
if in != "" {
t, err := time.ParseInLocation(shortForm, in, loc)
if err != nil {
fmt.Printf("date parse error %v", err)
os.Exit(0)
log.Abortf("Archive Manager Main: Date parse failed with input: '%s'\nError: %s\n", in, err.Error())
}
return t.Unix()
}

@ -59,9 +59,9 @@
}
},
"node_modules/@babel/runtime": {
"version": "7.26.0",
"resolved": "https://registry.npmjs.org/@babel/runtime/-/runtime-7.26.0.tgz",
"integrity": "sha512-FDSOghenHTiToteC/QRlv2q3DhPZ/oOXTBoirfWNx1Cx3TMVcGWQtMMmQcSvb/JjpNeGzx8Pq/b4fKEJuWm1sw==",
"version": "7.27.0",
"resolved": "https://registry.npmjs.org/@babel/runtime/-/runtime-7.27.0.tgz",
"integrity": "sha512-VtPOkrdPHZsKc/clNqyi9WUA8TINkZ4cGk63UUE3u4pmB2k+ZMQRDuIOagv8UVd6j7k0T3+RRIb7beKTebNbcw==",
"license": "MIT",
"dependencies": {
"regenerator-runtime": "^0.14.0"

@ -40,7 +40,7 @@
import JobRoofline from "./job/JobRoofline.svelte";
import EnergySummary from "./job/EnergySummary.svelte";
import PlotGrid from "./generic/PlotGrid.svelte";
import StatsTable from "./job/StatsTable.svelte";
import StatsTab from "./job/StatsTab.svelte";
export let dbid;
export let username;
@ -53,10 +53,8 @@
let isMetricsSelectionOpen = false,
selectedMetrics = [],
selectedScopes = [];
let plots = {},
statsTable
selectedScopes = [],
plots = {};
let availableMetrics = new Set(),
missingMetrics = [],
@ -127,28 +125,17 @@
let job = $initq.data.job;
if (!job) return;
const pendingMetrics = [
...(
(
ccconfig[`job_view_selectedMetrics:${job.cluster}:${job.subCluster}`] ||
ccconfig[`job_view_selectedMetrics:${job.cluster}`]
) ||
$initq.data.globalMetrics
.reduce((names, gm) => {
if (gm.availability.find((av) => av.cluster === job.cluster && av.subClusters.includes(job.subCluster))) {
names.push(gm.name);
}
return names;
}, [])
),
...(
(
ccconfig[`job_view_nodestats_selectedMetrics:${job.cluster}:${job.subCluster}`] ||
ccconfig[`job_view_nodestats_selectedMetrics:${job.cluster}`]
) ||
ccconfig[`job_view_nodestats_selectedMetrics`]
),
];
const pendingMetrics = (
ccconfig[`job_view_selectedMetrics:${job.cluster}:${job.subCluster}`] ||
ccconfig[`job_view_selectedMetrics:${job.cluster}`]
) ||
$initq.data.globalMetrics
.reduce((names, gm) => {
if (gm.availability.find((av) => av.cluster === job.cluster && av.subClusters.includes(job.subCluster))) {
names.push(gm.name);
}
return names;
}, [])
// Select default Scopes to load: Check before if any metric has accelerator scope by default
const accScopeDefault = [...pendingMetrics].some(function (m) {
@ -231,7 +218,7 @@
<Col xs={12} md={6} xl={3} class="mb-3 mb-xxl-0">
{#if $initq.error}
<Card body color="danger">{$initq.error.message}</Card>
{:else if $initq.data}
{:else if $initq?.data}
<Card class="overflow-auto" style="height: 400px;">
<TabContent> <!-- on:tab={(e) => (status = e.detail)} -->
{#if $initq.data?.job?.metaData?.message}
@ -305,7 +292,7 @@
<Card class="mb-3">
<CardBody>
<Row class="mb-2">
{#if $initq.data}
{#if $initq?.data}
<Col xs="auto">
<Button outline on:click={() => (isMetricsSelectionOpen = true)} color="primary">
Select Metrics (Selected {selectedMetrics.length} of {availableMetrics.size} available)
@ -318,7 +305,7 @@
{#if $jobMetrics.error}
<Row class="mt-2">
<Col>
{#if $initq.data.job.monitoringStatus == 0 || $initq.data.job.monitoringStatus == 2}
{#if $initq?.data && ($initq.data.job?.monitoringStatus == 0 || $initq.data.job?.monitoringStatus == 2)}
<Card body color="warning">Not monitored or archiving failed</Card>
<br />
{/if}
@ -343,7 +330,6 @@
{#if item.data}
<Metric
bind:this={plots[item.metric]}
on:more-loaded={({ detail }) => statsTable.moreLoaded(detail)}
job={$initq.data.job}
metricName={item.metric}
metricUnit={$initq.data.globalMetrics.find((gm) => gm.name == item.metric)?.unit}
@ -365,7 +351,7 @@
<!-- Statistcics Table -->
<Row class="mb-3">
<Col>
{#if $initq.data}
{#if $initq?.data}
<Card>
<TabContent>
{#if somethingMissing}
@ -398,22 +384,8 @@
</div>
</TabPane>
{/if}
<TabPane
tabId="stats"
tab="Statistics Table"
class="overflow-x-auto"
active={!somethingMissing}
>
{#if $jobMetrics?.data?.jobMetrics}
{#key $jobMetrics.data.jobMetrics}
<StatsTable
bind:this={statsTable}
job={$initq.data.job}
jobMetrics={$jobMetrics.data.jobMetrics}
/>
{/key}
{/if}
</TabPane>
<!-- Includes <TabPane> Statistics Table with Independent GQL Query -->
<StatsTab job={$initq.data.job} clusters={$initq.data.clusters} tabActive={!somethingMissing}/>
<TabPane tabId="job-script" tab="Job Script">
<div class="pre-wrapper">
{#if $initq.data.job.metaData?.jobScript}
@ -440,7 +412,7 @@
</Col>
</Row>
{#if $initq.data}
{#if $initq?.data}
<MetricSelection
cluster={$initq.data.job.cluster}
subCluster={$initq.data.job.subCluster}

@ -150,11 +150,6 @@
// On additional scope request
if (selectedScope == "load-all") {
// Push scope to statsTable (Needs to be in this case, else newly selected 'Metric.svelte' renders cause statsTable race condition)
const statsTableData = $metricData.data.singleUpdate.filter((x) => x.scope !== "node")
if (statsTableData.length > 0) {
dispatch("more-loaded", statsTableData);
}
// Set selected scope to min of returned scopes
selectedScope = minScope(scopes)
nodeOnly = (selectedScope == "node") // "node" still only scope after load-all

@ -0,0 +1,145 @@
<!--
@component Job-View subcomponent; Wraps the statsTable in a TabPane and contains GQL query for scoped statsData
Properties:
- `job Object`: The job object
- `clusters Object`: The clusters object
- `tabActive bool`: Boolean if StatsTabe Tab is Active on Creation
-->
<script>
import {
queryStore,
gql,
getContextClient
} from "@urql/svelte";
import { getContext } from "svelte";
import {
Card,
Button,
Row,
Col,
TabPane,
Spinner,
Icon
} from "@sveltestrap/sveltestrap";
import MetricSelection from "../generic/select/MetricSelection.svelte";
import StatsTable from "./statstab/StatsTable.svelte";
export let job;
export let clusters;
export let tabActive;
let loadScopes = false;
let selectedScopes = [];
let selectedMetrics = [];
let availableMetrics = new Set(); // For Info Only, filled by MetricSelection Component
let isMetricSelectionOpen = false;
const client = getContextClient();
const query = gql`
query ($dbid: ID!, $selectedMetrics: [String!]!, $selectedScopes: [MetricScope!]!) {
scopedJobStats(id: $dbid, metrics: $selectedMetrics, scopes: $selectedScopes) {
name
scope
stats {
hostname
id
data {
min
avg
max
}
}
}
}
`;
$: scopedStats = queryStore({
client: client,
query: query,
variables: { dbid: job.id, selectedMetrics, selectedScopes },
});
$: if (loadScopes) {
selectedScopes = ["node", "socket", "core", "hwthread", "accelerator"];
}
// Handle Job Query on Init -> is not executed anymore
getContext("on-init")(() => {
if (!job) return;
const pendingMetrics = (
getContext("cc-config")[`job_view_nodestats_selectedMetrics:${job.cluster}:${job.subCluster}`] ||
getContext("cc-config")[`job_view_nodestats_selectedMetrics:${job.cluster}`]
) || getContext("cc-config")["job_view_nodestats_selectedMetrics"];
// Select default Scopes to load: Check before if any metric has accelerator scope by default
const accScopeDefault = [...pendingMetrics].some(function (m) {
const cluster = clusters.find((c) => c.name == job.cluster);
const subCluster = cluster.subClusters.find((sc) => sc.name == job.subCluster);
return subCluster.metricConfig.find((smc) => smc.name == m)?.scope === "accelerator";
});
const pendingScopes = ["node"]
if (job.numNodes === 1) {
pendingScopes.push("socket")
pendingScopes.push("core")
pendingScopes.push("hwthread")
if (accScopeDefault) { pendingScopes.push("accelerator") }
}
selectedMetrics = [...pendingMetrics];
selectedScopes = [...pendingScopes];
});
</script>
<TabPane tabId="stats" tab="Statistics Table" class="overflow-x-auto" active={tabActive}>
<Row>
<Col class="m-2">
<Button outline on:click={() => (isMetricSelectionOpen = true)} class="px-2" color="primary" style="margin-right:0.5rem">
Select Metrics (Selected {selectedMetrics.length} of {availableMetrics.size} available)
</Button>
{#if job.numNodes > 1}
<Button class="px-2 ml-auto" color="success" outline on:click={() => (loadScopes = !loadScopes)} disabled={loadScopes}>
{#if !loadScopes}
<Icon name="plus-square-fill" style="margin-right:0.25rem"/> Add More Scopes
{:else}
<Icon name="check-square-fill" style="margin-right:0.25rem"/> OK: Scopes Added
{/if}
</Button>
{/if}
</Col>
</Row>
<hr class="mb-1 mt-1"/>
<!-- ROW1: Status-->
{#if $scopedStats.fetching}
<Row>
<Col class="m-3" style="text-align: center;">
<Spinner secondary/>
</Col>
</Row>
{:else if $scopedStats.error}
<Row>
<Col class="m-2">
<Card body color="danger">{$scopedStats.error.message}</Card>
</Col>
</Row>
{:else}
<StatsTable
hosts={job.resources.map((r) => r.hostname).sort()}
data={$scopedStats?.data?.scopedJobStats}
{selectedMetrics}
/>
{/if}
</TabPane>
<MetricSelection
cluster={job.cluster}
subCluster={job.subCluster}
configName="job_view_nodestats_selectedMetrics"
bind:allMetrics={availableMetrics}
bind:metrics={selectedMetrics}
bind:isOpen={isMetricSelectionOpen}
/>

@ -1,178 +0,0 @@
<!--
@component Job-View subcomponent; display table of metric data statistics with selectable scopes
Properties:
- `job Object`: The job object
- `jobMetrics [Object]`: The jobs metricdata
Exported:
- `moreLoaded`: Adds additional scopes requested from Metric.svelte in Job-View
-->
<script>
import { getContext } from "svelte";
import {
Button,
Table,
Input,
InputGroup,
InputGroupText,
Icon,
Row,
Col
} from "@sveltestrap/sveltestrap";
import { maxScope } from "../generic/utils.js";
import StatsTableEntry from "./StatsTableEntry.svelte";
import MetricSelection from "../generic/select/MetricSelection.svelte";
export let job;
export let jobMetrics;
const sortedJobMetrics = [...new Set(jobMetrics.map((m) => m.name))].sort()
const scopesForMetric = (metric) =>
jobMetrics.filter((jm) => jm.name == metric).map((jm) => jm.scope);
let hosts = job.resources.map((r) => r.hostname).sort(),
selectedScopes = {},
sorting = {},
isMetricSelectionOpen = false,
availableMetrics = new Set(),
selectedMetrics = (
getContext("cc-config")[`job_view_nodestats_selectedMetrics:${job.cluster}:${job.subCluster}`] ||
getContext("cc-config")[`job_view_nodestats_selectedMetrics:${job.cluster}`]
) || getContext("cc-config")["job_view_nodestats_selectedMetrics"];
for (let metric of sortedJobMetrics) {
// Not Exclusive or Multi-Node: get maxScope directly (mostly: node)
// -> Else: Load smallest available granularity as default as per availability
const availableScopes = scopesForMetric(metric);
if (job.exclusive != 1 || job.numNodes == 1) {
if (availableScopes.includes("accelerator")) {
selectedScopes[metric] = "accelerator";
} else if (availableScopes.includes("core")) {
selectedScopes[metric] = "core";
} else if (availableScopes.includes("socket")) {
selectedScopes[metric] = "socket";
} else {
selectedScopes[metric] = "node";
}
} else {
selectedScopes[metric] = maxScope(availableScopes);
}
sorting[metric] = {
min: { dir: "up", active: false },
avg: { dir: "up", active: false },
max: { dir: "up", active: false },
};
}
function sortBy(metric, stat) {
let s = sorting[metric][stat];
if (s.active) {
s.dir = s.dir == "up" ? "down" : "up";
} else {
for (let metric in sorting)
for (let stat in sorting[metric]) sorting[metric][stat].active = false;
s.active = true;
}
let series = jobMetrics.find(
(jm) => jm.name == metric && jm.scope == "node",
)?.metric.series;
sorting = { ...sorting };
hosts = hosts.sort((h1, h2) => {
let s1 = series.find((s) => s.hostname == h1)?.statistics;
let s2 = series.find((s) => s.hostname == h2)?.statistics;
if (s1 == null || s2 == null) return -1;
return s.dir != "up" ? s1[stat] - s2[stat] : s2[stat] - s1[stat];
});
}
export function moreLoaded(moreJobMetrics) {
moreJobMetrics.forEach(function (newMetric) {
if (!jobMetrics.some((m) => m.scope == newMetric.scope)) {
jobMetrics = [...jobMetrics, newMetric]
}
});
};
</script>
<Row>
<Col class="m-2">
<Button outline on:click={() => (isMetricSelectionOpen = true)} class="w-auto px-2" color="primary">
Select Metrics (Selected {selectedMetrics.length} of {availableMetrics.size} available)
</Button>
</Col>
</Row>
<hr class="mb-1 mt-1"/>
<Table class="mb-0">
<thead>
<!-- Header Row 1: Selectors -->
<tr>
<th/>
{#each selectedMetrics as metric}
<!-- To Match Row-2 Header Field Count-->
<th colspan={selectedScopes[metric] == "node" ? 3 : 4}>
<InputGroup>
<InputGroupText>
{metric}
</InputGroupText>
<Input type="select" bind:value={selectedScopes[metric]}>
{#each scopesForMetric(metric, jobMetrics) as scope}
<option value={scope}>{scope}</option>
{/each}
</Input>
</InputGroup>
</th>
{/each}
</tr>
<!-- Header Row 2: Fields -->
<tr>
<th>Node</th>
{#each selectedMetrics as metric}
{#if selectedScopes[metric] != "node"}
<th>Id</th>
{/if}
{#each ["min", "avg", "max"] as stat}
<th on:click={() => sortBy(metric, stat)}>
{stat}
{#if selectedScopes[metric] == "node"}
<Icon
name="caret-{sorting[metric][stat].dir}{sorting[metric][stat]
.active
? '-fill'
: ''}"
/>
{/if}
</th>
{/each}
{/each}
</tr>
</thead>
<tbody>
{#each hosts as host (host)}
<tr>
<th scope="col">{host}</th>
{#each selectedMetrics as metric (metric)}
<StatsTableEntry
{host}
{metric}
scope={selectedScopes[metric]}
{jobMetrics}
/>
{/each}
</tr>
{/each}
</tbody>
</Table>
<MetricSelection
cluster={job.cluster}
subCluster={job.subCluster}
configName="job_view_nodestats_selectedMetrics"
bind:allMetrics={availableMetrics}
bind:metrics={selectedMetrics}
bind:isOpen={isMetricSelectionOpen}
/>

@ -40,14 +40,14 @@
const client = getContextClient();
const polarQuery = gql`
query ($dbid: ID!, $selectedMetrics: [String!]!) {
jobMetricStats(id: $dbid, metrics: $selectedMetrics) {
jobStats(id: $dbid, metrics: $selectedMetrics) {
name
stats {
min
avg
max
}
min
avg
max
}
}
}
`;
@ -66,7 +66,7 @@
{:else}
<Polar
{polarMetrics}
polarData={$polarData.data.jobMetricStats}
polarData={$polarData.data.jobStats}
/>
{/if}
</CardBody>

@ -0,0 +1,139 @@
<!--:
@component Job-View subcomponent; display table of metric data statistics with selectable scopes
Properties:
- `data Object`: The data object
- `selectedMetrics [String]`: The selected metrics
- `hosts [String]`: The list of hostnames of this job
-->
<script>
import {
Table,
Input,
InputGroup,
InputGroupText,
Icon,
} from "@sveltestrap/sveltestrap";
import StatsTableEntry from "./StatsTableEntry.svelte";
export let data = [];
export let selectedMetrics = [];
export let hosts = [];
let sorting = {};
let availableScopes = {};
let selectedScopes = {};
const scopesForMetric = (metric) =>
data?.filter((jm) => jm.name == metric)?.map((jm) => jm.scope) || [];
const setScopeForMetric = (metric, scope) =>
selectedScopes[metric] = scope
$: if (data && selectedMetrics) {
for (let metric of selectedMetrics) {
availableScopes[metric] = scopesForMetric(metric);
// Set Initial Selection, but do not use selectedScopes: Skips reactivity
if (availableScopes[metric].includes("accelerator")) {
setScopeForMetric(metric, "accelerator");
} else if (availableScopes[metric].includes("core")) {
setScopeForMetric(metric, "core");
} else if (availableScopes[metric].includes("socket")) {
setScopeForMetric(metric, "socket");
} else {
setScopeForMetric(metric, "node");
}
sorting[metric] = {
min: { dir: "up", active: false },
avg: { dir: "up", active: false },
max: { dir: "up", active: false },
};
}
}
function sortBy(metric, stat) {
let s = sorting[metric][stat];
if (s.active) {
s.dir = s.dir == "up" ? "down" : "up";
} else {
for (let metric in sorting)
for (let stat in sorting[metric]) sorting[metric][stat].active = false;
s.active = true;
}
let stats = data.find(
(d) => d.name == metric && d.scope == "node",
)?.stats || [];
sorting = { ...sorting };
hosts = hosts.sort((h1, h2) => {
let s1 = stats.find((s) => s.hostname == h1)?.data;
let s2 = stats.find((s) => s.hostname == h2)?.data;
if (s1 == null || s2 == null) return -1;
return s.dir != "up" ? s1[stat] - s2[stat] : s2[stat] - s1[stat];
});
}
</script>
<Table class="mb-0">
<thead>
<!-- Header Row 1: Selectors -->
<tr>
<th/>
{#each selectedMetrics as metric}
<!-- To Match Row-2 Header Field Count-->
<th colspan={selectedScopes[metric] == "node" ? 3 : 4}>
<InputGroup>
<InputGroupText>
{metric}
</InputGroupText>
<Input type="select" bind:value={selectedScopes[metric]} disabled={availableScopes[metric].length === 1}>
{#each (availableScopes[metric] || []) as scope}
<option value={scope}>{scope}</option>
{/each}
</Input>
</InputGroup>
</th>
{/each}
</tr>
<!-- Header Row 2: Fields -->
<tr>
<th>Node</th>
{#each selectedMetrics as metric}
{#if selectedScopes[metric] != "node"}
<th>Id</th>
{/if}
{#each ["min", "avg", "max"] as stat}
<th on:click={() => sortBy(metric, stat)}>
{stat}
{#if selectedScopes[metric] == "node"}
<Icon
name="caret-{sorting[metric][stat].dir}{sorting[metric][stat]
.active
? '-fill'
: ''}"
/>
{/if}
</th>
{/each}
{/each}
</tr>
</thead>
<tbody>
{#each hosts as host (host)}
<tr>
<th scope="col">{host}</th>
{#each selectedMetrics as metric (metric)}
<StatsTableEntry
{data}
{host}
{metric}
scope={selectedScopes[metric]}
/>
{/each}
</tr>
{/each}
</tbody>
</Table>

@ -1,11 +1,11 @@
<!--
@component Job-View subcomponent; Single Statistics entry component fpr statstable
@component Job-View subcomponent; Single Statistics entry component for statstable
Properties:
- `host String`: The hostname (== node)
- `metric String`: The metric name
- `scope String`: The selected scope
- `jobMetrics [Object]`: The jobs metricdata
- `data [Object]`: The jobs statsdata
-->
<script>
@ -14,59 +14,59 @@
export let host;
export let metric;
export let scope;
export let jobMetrics;
export let data;
function compareNumbers(a, b) {
return a.id - b.id;
}
function sortByField(field) {
let s = sorting[field];
if (s.active) {
s.dir = s.dir == "up" ? "down" : "up";
} else {
for (let field in sorting) sorting[field].active = false;
s.active = true;
}
sorting = { ...sorting };
series = series.sort((a, b) => {
if (a == null || b == null) return -1;
if (field === "id") {
return s.dir != "up" ? a[field] - b[field] : b[field] - a[field];
} else {
return s.dir != "up"
? a.statistics[field] - b.statistics[field]
: b.statistics[field] - a.statistics[field];
}
});
}
let sorting = {
let entrySorting = {
id: { dir: "down", active: true },
min: { dir: "up", active: false },
avg: { dir: "up", active: false },
max: { dir: "up", active: false },
};
$: series = jobMetrics
.find((jm) => jm.name == metric && jm.scope == scope)
?.metric.series.filter((s) => s.hostname == host && s.statistics != null)
?.sort(compareNumbers);
function compareNumbers(a, b) {
return a.id - b.id;
}
function sortByField(field) {
let s = entrySorting[field];
if (s.active) {
s.dir = s.dir == "up" ? "down" : "up";
} else {
for (let field in entrySorting) entrySorting[field].active = false;
s.active = true;
}
entrySorting = { ...entrySorting };
stats = stats.sort((a, b) => {
if (a == null || b == null) return -1;
if (field === "id") {
return s.dir != "up" ? a[field].localeCompare(b[field]) : b[field].localeCompare(a[field])
} else {
return s.dir != "up"
? a.data[field] - b.data[field]
: b.data[field] - a.data[field];
}
});
}
$: stats = data
?.find((d) => d.name == metric && d.scope == scope)
?.stats.filter((s) => s.hostname == host && s.data != null)
?.sort(compareNumbers) || [];
</script>
{#if series == null || series.length == 0}
{#if stats == null || stats.length == 0}
<td colspan={scope == "node" ? 3 : 4}><i>No data</i></td>
{:else if series.length == 1 && scope == "node"}
{:else if stats.length == 1 && scope == "node"}
<td>
{series[0].statistics.min}
{stats[0].data.min}
</td>
<td>
{series[0].statistics.avg}
{stats[0].data.avg}
</td>
<td>
{series[0].statistics.max}
{stats[0].data.max}
</td>
{:else}
<td colspan="4">
@ -76,19 +76,19 @@
<th on:click={() => sortByField(field)}>
Sort
<Icon
name="caret-{sorting[field].dir}{sorting[field].active
name="caret-{entrySorting[field].dir}{entrySorting[field].active
? '-fill'
: ''}"
/>
</th>
{/each}
</tr>
{#each series as s, i}
{#each stats as s, i}
<tr>
<th>{s.id ?? i}</th>
<td>{s.statistics.min}</td>
<td>{s.statistics.avg}</td>
<td>{s.statistics.max}</td>
<td>{s.data.min}</td>
<td>{s.data.avg}</td>
<td>{s.data.max}</td>
</tr>
{/each}
</table>

@ -26,8 +26,7 @@ var frontendFiles embed.FS
func ServeFiles() http.Handler {
publicFiles, err := fs.Sub(frontendFiles, "frontend/public")
if err != nil {
log.Fatalf("WEB/WEB > cannot find frontend public files")
panic(err)
log.Abortf("Serve Files: Could not find 'frontend/public' file directory.\nError: %s\n", err.Error())
}
return http.FileServer(http.FS(publicFiles))
}
@ -75,8 +74,7 @@ func init() {
templates[strings.TrimPrefix(path, "templates/")] = template.Must(template.Must(base.Clone()).ParseFS(templateFiles, path))
return nil
}); err != nil {
log.Fatalf("WEB/WEB > cannot find frontend template files")
panic(err)
log.Abortf("Web init(): Could not find frontend template files.\nError: %s\n", err.Error())
}
_ = base