Bump github.com/aws/aws-sdk-go-v2/config from 1.32.6 to 1.32.7

Bumps [github.com/aws/aws-sdk-go-v2/config](https://github.com/aws/aws-sdk-go-v2) from 1.32.6 to 1.32.7.
- [Release notes](https://github.com/aws/aws-sdk-go-v2/releases)
- [Changelog](https://github.com/aws/aws-sdk-go-v2/blob/main/changelog-template.json)
- [Commits](https://github.com/aws/aws-sdk-go-v2/compare/v1.32.6...v1.32.7)

---
updated-dependencies:
- dependency-name: github.com/aws/aws-sdk-go-v2/config
  dependency-version: 1.32.7
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

.github/dependabot.yml

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

.github/workflows/Release.yml

@@ -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 }}

.github/workflows/test.yml

@@ -7,7 +7,7 @@ jobs:
       - name: Install Go
         uses: actions/setup-go@v4
         with:
-          go-version: 1.19.x
+          go-version: 1.25.x
       - name: Checkout code
         uses: actions/checkout@v3
       - name: Build, Vet & Test

.gitignore

@@ -1,19 +1,32 @@
 /cc-backend
-
-/var/job-archive
-/var/*.db
-/var/machine-state
-
 /.env
 /config.json
+/uiConfig.json
+
+/var/job-archive
+/var/machine-state
+/var/*.db-shm
+/var/*.db-wal
+/var/*.db
+/var/*.txt
+
+/var/checkpoints*
+
+migrateTimestamps.pl
+test_ccms_write_api*
 
 /web/frontend/public/build
 /web/frontend/node_modules
-/.vscode/*
+
 /archive-migration
 /archive-manager
-var/job.db-shm
-var/job.db-wal
 
+/internal/repository/testdata/job.db-shm
+/internal/repository/testdata/job.db-wal
+
+/.vscode/*
 dist/
 *.db
+.idea
+tools/archive-migration/archive-migration
+tools/archive-manager/archive-manager

.goreleaser.yaml

@@ -34,19 +34,6 @@ builds:
     main: ./tools/archive-manager
     tags:
       - static_build
-  - env:
-      - CGO_ENABLED=0
-    goos:
-      - linux
-    goarch:
-      - amd64
-    goamd64:
-      - v3
-    id: "archive-migration"
-    binary: archive-migration
-    main: ./tools/archive-migration
-    tags:
-      - static_build
   - env:
       - CGO_ENABLED=0
     goos:
@@ -70,7 +57,7 @@ archives:
       {{- else }}{{ .Arch }}{{ end }}
       {{- if .Arm }}v{{ .Arm }}{{ end }}
 checksum:
-  name_template: 'checksums.txt'
+  name_template: "checksums.txt"
 snapshot:
   name_template: "{{ incpatch .Version }}-next"
 changelog:
@@ -100,6 +87,7 @@ changelog:
 release:
   draft: false
   footer: |
-    Supports job archive version 1 and database version 6.
+    Supports job archive version 2 and database version 8.
+    Please check out the [Release Notes](https://github.com/ClusterCockpit/cc-backend/blob/master/ReleaseNotes.md) for further details on breaking changes.
 
 # vim: set ts=2 sw=2 tw=0 fo=cnqoj

AGENTS.md

@@ -0,0 +1,26 @@
+# ClusterCockpit Backend - Agent Guidelines
+
+## Build/Test Commands
+
+- Build: `make` or `go build ./cmd/cc-backend`
+- Run all tests: `make test` (runs: `go clean -testcache && go build ./... && go vet ./... && go test ./...`)
+- Run single test: `go test -run TestName ./path/to/package`
+- Run single test file: `go test ./path/to/package -run TestName`
+- Frontend build: `cd web/frontend && npm install && npm run build`
+- Generate GraphQL: `make graphql` (uses gqlgen)
+- Generate Swagger: `make swagger` (uses swaggo/swag)
+
+## Code Style
+
+- **Formatting**: Use `gofumpt` for all Go files (strict requirement)
+- **Copyright header**: All files must include copyright header (see existing files)
+- **Package docs**: Document packages with comprehensive package-level comments explaining purpose, usage, configuration
+- **Imports**: Standard library first, then external packages, then internal packages (grouped with blank lines)
+- **Naming**: Use camelCase for private, PascalCase for exported; descriptive names (e.g., `JobRepository`, `handleError`)
+- **Error handling**: Return errors, don't panic; use custom error types where appropriate; log with cclog package
+- **Logging**: Use `cclog` package (e.g., `cclog.Errorf()`, `cclog.Warnf()`, `cclog.Debugf()`)
+- **Testing**: Use standard `testing` package; use `testify/assert` for assertions; name tests `TestFunctionName`
+- **Comments**: Document all exported functions/types with godoc-style comments
+- **Structs**: Document fields with inline comments, especially for complex configurations
+- **HTTP handlers**: Return proper status codes; use `handleError()` helper for consistent error responses
+- **JSON**: Use struct tags for JSON marshaling; `DisallowUnknownFields()` for strict decoding

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2022 NHR@FAU, University Erlangen-Nuremberg
+Copyright (c) NHR@FAU, University Erlangen-Nuremberg
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

Makefile

@@ -1,8 +1,6 @@
 TARGET = ./cc-backend
-VAR = ./var
-CFG = config.json .env
 FRONTEND = ./web/frontend
-VERSION = 1.0.0
+VERSION = 1.4.4
 GIT_HASH := $(shell git rev-parse --short HEAD || echo 'development')
 CURRENT_TIME = $(shell date +"%Y-%m-%d:T%H:%M:%S")
 LD_FLAGS = '-s -X main.date=${CURRENT_TIME} -X main.version=${VERSION} -X main.commit=${GIT_HASH}'
@@ -22,17 +20,27 @@ SVELTE_COMPONENTS = status   \
 					header
 
 SVELTE_TARGETS = $(addprefix $(FRONTEND)/public/build/,$(addsuffix .js, $(SVELTE_COMPONENTS)))
-SVELTE_SRC = $(wildcard $(FRONTEND)/src/*.svelte)         \
-			 $(wildcard $(FRONTEND)/src/*.js)             \
-			 $(wildcard $(FRONTEND)/src/filters/*.svelte) \
-			 $(wildcard $(FRONTEND)/src/plots/*.svelte)   \
-			 $(wildcard $(FRONTEND)/src/joblist/*.svelte)
+SVELTE_SRC = $(wildcard $(FRONTEND)/src/*.svelte)                 \
+			 $(wildcard $(FRONTEND)/src/*.js)                     \
+			 $(wildcard $(FRONTEND)/src/analysis/*.svelte)        \
+			 $(wildcard $(FRONTEND)/src/config/*.svelte)          \
+			 $(wildcard $(FRONTEND)/src/config/admin/*.svelte)    \
+			 $(wildcard $(FRONTEND)/src/config/user/*.svelte)     \
+			 $(wildcard $(FRONTEND)/src/generic/*.js)             \
+			 $(wildcard $(FRONTEND)/src/generic/*.svelte)         \
+			 $(wildcard $(FRONTEND)/src/generic/filters/*.svelte) \
+			 $(wildcard $(FRONTEND)/src/generic/plots/*.svelte)   \
+			 $(wildcard $(FRONTEND)/src/generic/joblist/*.svelte) \
+			 $(wildcard $(FRONTEND)/src/generic/helper/*.svelte)  \
+			 $(wildcard $(FRONTEND)/src/generic/select/*.svelte)  \
+			 $(wildcard $(FRONTEND)/src/header/*.svelte)          \
+			 $(wildcard $(FRONTEND)/src/job/*.svelte)
 
-.PHONY: clean test tags frontend $(TARGET)
+.PHONY: clean distclean test tags frontend swagger graphql $(TARGET)
 
 .NOTPARALLEL:
 
-$(TARGET): $(VAR) $(CFG) $(SVELTE_TARGETS)
+$(TARGET): $(SVELTE_TARGETS)
 	$(info ===>  BUILD cc-backend)
 	@go build -ldflags=${LD_FLAGS} ./cmd/cc-backend
 
@@ -40,6 +48,15 @@ frontend:
 	$(info ===>  BUILD frontend)
 	cd web/frontend && npm install && npm run build
 
+swagger:
+	$(info ===>  GENERATE swagger)
+	@go tool github.com/swaggo/swag/cmd/swag init  --parseDependency -d ./internal/api -g rest.go -o ./api
+	@mv ./api/docs.go ./internal/api/docs.go
+
+graphql:
+	$(info ===>  GENERATE graphql)
+	@go tool github.com/99designs/gqlgen
+
 clean:
 	$(info ===>  CLEAN)
 	@go clean
@@ -49,7 +66,7 @@ distclean:
 	@$(MAKE) clean
 	$(info ===>  DISTCLEAN)
 	@rm -rf $(FRONTEND)/node_modules
-	@rm -rf $(VAR)
+	@rm -rf ./var
 
 test:
 	$(info ===>  TESTING)
@@ -63,103 +80,8 @@ tags:
 	@ctags -R
 
 $(VAR):
-	@mkdir $(VAR)
-
-config.json:
-	$(info ===>  Initialize config.json file)
-	@cp configs/config.json config.json
-
-.env:
-	$(info ===>  Initialize .env file)
-	@cp configs/env-template.txt .env
+	@mkdir -p $(VAR)
 
 $(SVELTE_TARGETS): $(SVELTE_SRC)
 	$(info ===>  BUILD frontend)
 	cd web/frontend && npm install && npm run build
-
-install: $(TARGET)
-	@WORKSPACE=$(PREFIX)
-	@if [ -z "$${WORKSPACE}" ]; then exit 1; fi
-	@mkdir --parents --verbose $${WORKSPACE}/usr/$(BINDIR)
-	@install -Dpm 755 $(TARGET) $${WORKSPACE}/usr/$(BINDIR)/$(TARGET)
-	@install -Dpm 600 configs/config.json $${WORKSPACE}/etc/$(TARGET)/$(TARGET).json
-
-.ONESHELL:
-.PHONY: RPM
-RPM: build/package/cc-backend.spec
-	@WORKSPACE="$${PWD}"
-	@SPECFILE="$${WORKSPACE}/build/package/cc-backend.spec"
-	# Setup RPM build tree
-	@eval $$(rpm --eval "ARCH='%{_arch}' RPMDIR='%{_rpmdir}' SOURCEDIR='%{_sourcedir}' SPECDIR='%{_specdir}' SRPMDIR='%{_srcrpmdir}' BUILDDIR='%{_builddir}'")
-	@mkdir --parents --verbose "$${RPMDIR}" "$${SOURCEDIR}" "$${SPECDIR}" "$${SRPMDIR}" "$${BUILDDIR}"
-	# Create source tarball
-	@COMMITISH="HEAD"
-	@VERS=$$(git describe --tags $${COMMITISH})
-	@VERS=$${VERS#v}
-	@VERS=$$(echo $$VERS | sed -e s+'-'+'_'+g)
-	@if [ "$${VERS}" = "" ]; then VERS="$(VERSION)"; fi
-	@eval $$(rpmspec --query --queryformat "NAME='%{name}' VERSION='%{version}' RELEASE='%{release}' NVR='%{NVR}' NVRA='%{NVRA}'" --define="VERS $${VERS}" "$${SPECFILE}")
-	@PREFIX="$${NAME}-$${VERSION}"
-	@FORMAT="tar.gz"
-	@SRCFILE="$${SOURCEDIR}/$${PREFIX}.$${FORMAT}"
-	@git archive --verbose --format "$${FORMAT}" --prefix="$${PREFIX}/" --output="$${SRCFILE}" $${COMMITISH}
-	# Build RPM and SRPM
-	@rpmbuild -ba --define="VERS $${VERS}" --rmsource --clean "$${SPECFILE}"
-	# Report RPMs and SRPMs when in GitHub Workflow
-	@if [ "$${GITHUB_ACTIONS}" = true ]; then
-	@     RPMFILE="$${RPMDIR}/$${ARCH}/$${NVRA}.rpm"
-	@     SRPMFILE="$${SRPMDIR}/$${NVR}.src.rpm"
-	@     echo "RPM: $${RPMFILE}"
-	@     echo "SRPM: $${SRPMFILE}"
-	@     echo "::set-output name=SRPM::$${SRPMFILE}"
-	@     echo "::set-output name=RPM::$${RPMFILE}"
-	@fi
-
-.ONESHELL:
-.PHONY: DEB
-DEB: build/package/cc-backend.deb.control
-	@BASEDIR=$${PWD}
-	@WORKSPACE=$${PWD}/.dpkgbuild
-	@DEBIANDIR=$${WORKSPACE}/debian
-	@DEBIANBINDIR=$${WORKSPACE}/DEBIAN
-	@mkdir --parents --verbose $$WORKSPACE $$DEBIANBINDIR
-	#@mkdir --parents --verbose $$DEBIANDIR
-	@CONTROLFILE="$${BASEDIR}/build/package/cc-backend.deb.control"
-	@COMMITISH="HEAD"
-	@VERS=$$(git describe --tags --abbrev=0 $${COMMITISH})
-	@VERS=$${VERS#v}
-	@VERS=$$(echo $$VERS | sed -e s+'-'+'_'+g)
-	@if [ "$${VERS}" = "" ]; then VERS="$(VERSION)"; fi
-	@ARCH=$$(uname -m)
-	@ARCH=$$(echo $$ARCH | sed -e s+'_'+'-'+g)
-	@if [ "$${ARCH}" = "x86-64" ]; then ARCH=amd64; fi
-	@PREFIX="$${NAME}-$${VERSION}_$${ARCH}"
-	@SIZE_BYTES=$$(du -bcs --exclude=.dpkgbuild "$${WORKSPACE}"/ | awk '{print $$1}' | head -1 | sed -e 's/^0\+//')
-	@SIZE="$$(awk -v size="$$SIZE_BYTES" 'BEGIN {print (size/1024)+1}' | awk '{print int($$0)}')"
-	#@sed -e s+"{VERSION}"+"$$VERS"+g -e s+"{INSTALLED_SIZE}"+"$$SIZE"+g -e s+"{ARCH}"+"$$ARCH"+g $$CONTROLFILE > $${DEBIANDIR}/control
-	@sed -e s+"{VERSION}"+"$$VERS"+g -e s+"{INSTALLED_SIZE}"+"$$SIZE"+g -e s+"{ARCH}"+"$$ARCH"+g $$CONTROLFILE > $${DEBIANBINDIR}/control
-	@mkdir --parents --verbose "$${WORKSPACE}"/$(VAR)
-	@touch "$${WORKSPACE}"/$(VAR)/job.db
-	@cd web/frontend && yarn install && yarn build && cd -
-	@go build -ldflags=${LD_FLAGS} ./cmd/cc-backend
-	@mkdir --parents --verbose $${WORKSPACE}/usr/$(BINDIR)
-	@cp $(TARGET) $${WORKSPACE}/usr/$(BINDIR)/$(TARGET)
-	@chmod 0755 $${WORKSPACE}/usr/$(BINDIR)/$(TARGET)
-	@mkdir --parents --verbose $${WORKSPACE}/etc/$(TARGET)
-	@cp configs/config.json $${WORKSPACE}/etc/$(TARGET)/$(TARGET).json
-	@chmod 0600 $${WORKSPACE}/etc/$(TARGET)/$(TARGET).json
-	@mkdir --parents --verbose $${WORKSPACE}/usr/lib/systemd/system
-	@cp build/package/$(TARGET).service $${WORKSPACE}/usr/lib/systemd/system/$(TARGET).service
-	@chmod 0644 $${WORKSPACE}/usr/lib/systemd/system/$(TARGET).service
-	@mkdir --parents --verbose $${WORKSPACE}/etc/default
-	@cp build/package/$(TARGET).config $${WORKSPACE}/etc/default/$(TARGET)
-	@chmod 0600 $${WORKSPACE}/etc/default/$(TARGET)
-	@mkdir --parents --verbose $${WORKSPACE}/usr/lib/sysusers.d
-	@cp build/package/$(TARGET).sysusers $${WORKSPACE}/usr/lib/sysusers.d/$(TARGET).conf
-	@chmod 0644 $${WORKSPACE}/usr/lib/sysusers.d/$(TARGET).conf
-	@DEB_FILE="cc-metric-store_$${VERS}_$${ARCH}.deb"
-	@dpkg-deb -b $${WORKSPACE} "$$DEB_FILE"
-	@rm -r "$${WORKSPACE}"
-	@if [ "$${GITHUB_ACTIONS}" = "true" ]; then
-	@     echo "::set-output name=DEB::$${DEB_FILE}"
-	@fi

README.md

@@ -1,9 +1,19 @@
+# NOTE
+
+While we do our best to keep the master branch in a usable state, there is no guarantee the master branch works.
+Please do not use it for production!
+
+Please have a look at the [Release
+Notes](https://github.com/ClusterCockpit/cc-backend/blob/master/ReleaseNotes.md)
+for breaking changes!
+
 # ClusterCockpit REST and GraphQL API backend
 
 [![Build](https://github.com/ClusterCockpit/cc-backend/actions/workflows/test.yml/badge.svg)](https://github.com/ClusterCockpit/cc-backend/actions/workflows/test.yml)
 
 This is a Golang backend implementation for a REST and GraphQL API according to
-the [ClusterCockpit specifications](https://github.com/ClusterCockpit/cc-specifications). It also
+the [ClusterCockpit
+specifications](https://github.com/ClusterCockpit/cc-specifications). It also
 includes a web interface for ClusterCockpit. This implementation replaces the
 previous PHP Symfony based ClusterCockpit web interface. The reasons for
 switching from PHP Symfony to a Golang based solution are explained
@@ -11,31 +21,31 @@ switching from PHP Symfony to a Golang based solution are explained
 
 ## Overview
 
+This is a Golang web backend for the ClusterCockpit job-specific performance
+monitoring framework. It provides a REST API for integrating ClusterCockpit with
+an HPC cluster batch system and external analysis scripts. Data exchange between
+the web front-end and the back-end is based on a GraphQL API. The web frontend
+is also served by the backend using [Svelte](https://svelte.dev/) components.
+Layout and styling are based on [Bootstrap 5](https://getbootstrap.com/) using
+[Bootstrap Icons](https://icons.getbootstrap.com/).
 
-This is a Golang web backend for the ClusterCockpit job-specific performance monitoring framework.
-It provides a REST API for integrating ClusterCockpit with an HPC cluster batch system and external analysis scripts.
-Data exchange between the web front-end and the back-end is based on a GraphQL API.
-The web frontend is also served by the backend using [Svelte](https://svelte.dev/) components.
-Layout and styling are based on [Bootstrap 5](https://getbootstrap.com/) using [Bootstrap Icons](https://icons.getbootstrap.com/).
-
-The backend uses [SQLite 3](https://sqlite.org/) as a relational SQL database by default.
-Optionally it can use a MySQL/MariaDB database server.
-While there are metric data  backends for the InfluxDB and Prometheus time series databases, the only tested and supported setup is to use cc-metric-store as the metric data backend.
-Documentation on how to integrate ClusterCockpit with other time series databases will be added in the future. 
+The backend uses [SQLite 3](https://sqlite.org/) as a relational SQL database by
+default. Optionally it can use a MySQL/MariaDB database server. While there are
+metric data backends for the InfluxDB and Prometheus time series databases, the
+only tested and supported setup is to use cc-metric-store as the metric data
+backend. Documentation on how to integrate ClusterCockpit with other time series
+databases will be added in the future.
 
 Completed batch jobs are stored in a file-based job archive according to
-[this specification] (https://github.com/ClusterCockpit/cc-specifications/tree/master/job-archive).
+[this specification](https://github.com/ClusterCockpit/cc-specifications/tree/master/job-archive).
 The backend supports authentication via local accounts, an external LDAP
 directory, and JWT tokens. Authorization for APIs is implemented with
 [JWT](https://jwt.io/) tokens created with public/private key encryption.
 
-You find more detailed information here:
-* `./configs/README.md`: Infos about configuration and setup of cc-backend.
-* `./init/README.md`: Infos on how to setup cc-backend as systemd service on Linux.
-* `./tools/README.md`: Infos on the JWT authorizatin token workflows in ClusterCockpit.
-* `./docs`: You can find further documentation here. There is also a Hands-on tutorial that is recommended to get familiar with the ClusterCockpit setup.
+You find a detailed documentation on the [ClusterCockpit
+Webpage](https://clustercockpit.org).
 
-**NOTE**
+## Build requirements
 
 ClusterCockpit requires a current version of the golang toolchain and node.js.
 You can check `go.mod` to see what is the current minimal golang version needed.
@@ -46,7 +56,7 @@ on the Go standard library, it is crucial for security and performance to use a
 current version of golang. In addition, an old golang toolchain may limit the supported
 versions of third-party packages.
 
-## How to try ClusterCockpit with a demo setup.
+## How to try ClusterCockpit with a demo setup
 
 We provide a shell script that downloads demo data and automatically starts the
 cc-backend. You will need `wget`, `go`, `node`, `npm` in your path to
@@ -58,31 +68,37 @@ cd ./cc-backend
 ./startDemo.sh
 ```
 
-You can also try the demo using the lates release binary.
+You can also try the demo using the latest release binary.
 Create a folder and put the release binary `cc-backend` into this folder.
 Execute the following steps:
-```
-$ ./cc-backend -init
-$ vim config.json (Add a second cluster entry and name the clusters alex and fritz)
-$ wget https://hpc-mover.rrze.uni-erlangen.de/HPC-Data/0x7b58aefb/eig7ahyo6fo2bais0ephuf2aitohv1ai/job-archive-demo.tar
-$ tar xf job-archive-demo.tar
-$ ./cc-backend -init-db -add-user demo:admin:demo -loglevel info
-$ ./cc-backend -server -dev -loglevel info
+
+```shell
+./cc-backend -init
+vim config.json (Add a second cluster entry and name the clusters alex and fritz)
+wget https://hpc-mover.rrze.uni-erlangen.de/HPC-Data/0x7b58aefb/eig7ahyo6fo2bais0ephuf2aitohv1ai/job-archive-demo.tar
+tar xf job-archive-demo.tar
+./cc-backend -init-db -add-user demo:admin:demo -loglevel info
+./cc-backend -server -dev -loglevel info
 ```
 
-You can access the web interface at http://localhost:8080.
+You can access the web interface at [http://localhost:8080](http://localhost:8080).
 Credentials for login are `demo:demo`.
 Please note that some views do not work without a metric backend (e.g., the
 Analysis, Systems and Status views).
 
-## Howto build and run
+## How to build and run
 
-There is a Makefile to automate the build of cc-backend. The Makefile supports the following targets:
-* `$ make`: Initialize `var` directory and build svelte frontend and backend binary. Note that there is no proper prerequesite handling. Any change of frontend source files will result in a complete rebuild.
-* `$ make clean`: Clean go build cache and remove binary.
-* `$ make test`: Run the tests that are also run in the GitHub workflow setup.
+There is a Makefile to automate the build of cc-backend. The Makefile supports
+the following targets:
+
+- `make`: Initialize `var` directory and build svelte frontend and backend
+  binary. Note that there is no proper prerequisite handling. Any change of
+  frontend source files will result in a complete rebuild.
+- `make clean`: Clean go build cache and remove binary.
+- `make test`: Run the tests that are also run in the GitHub workflow setup.
 
 A common workflow for setting up cc-backend from scratch is:
+
 ```sh
 git clone https://github.com/ClusterCockpit/cc-backend.git
 
@@ -113,89 +129,43 @@ ln -s <your-existing-job-archive> ./var/job-archive
 ./cc-backend -help
 ```
 
-### Run as systemd daemon
-
-To run this program as a daemon, cc-backend comes with a [example systemd setup](./init/README.md).
-
-## Configuration and setup
-
-cc-backend can be used as a local web interface for an existing job archive or
-as a server for the ClusterCockpit monitoring framework.
-
-Create your job archive according to [this specification] (https://github.com/ClusterCockpit/cc-specifications/tree/master/job-archive).
-At least one cluster directory with a valid `cluster.json` file is required. If
-you configure the job archive from scratch, you must also create the job
-archive version file that contains the job archive version as an integer.
-You can retrieve the currently supported version by running the following
-command:
-```
-$ ./cc-backend -version
-```
-It is ok to have no jobs in the job archive.
-
-### Configuration
-
-A configuration file in JSON format must be specified with `-config` to override the default settings.
-By default, a `config.json` file located in the current directory of the `cc-backend` process will be loaded even without the `-config` flag.
-Documentation of all supported configuration and command line options can be found [here](./configs/README.md).
-
-## Database initialization and migration
-
-Each `cc-backend` version supports a specific database version.
-At startup, the version of the sqlite database is checked and `cc-backend` terminates if the version does not match.
-`cc-backend` supports the migration of the database schema to the required version with the command line option `-migrate-db`.
-If the database file does not exist yet, it will be created and initialized with the command line option `-migrate-db`.
-If you want to use a newer database version with an older version of cc-backend, you can downgrade a database with the external tool [migrate](https://github.com/golang-migrate/migrate).
-In this case, you must specify the path to the migration files in a current source tree: `./internal/repository/migrations/`.
-
-## Development and testing
-When making changes to the REST or GraphQL API, the appropriate code generators must be used.
-You must always rebuild `cc-backend` after updating the API files.
-
-### Update GraphQL schema
-
-This project uses [gqlgen](https://github.com/99designs/gqlgen) for the GraphQL API.
-The schema can be found in `./api/schema.graphqls`.
-After changing it, you need to run `go run github.com/99designs/gqlgen`, which will update `./internal/graph/model`.
-If new resolvers are needed, they will be added to `./internal/graph/schema.resolvers.go`, where you will then need to implement them.
-If you start `cc-backend` with the `-dev` flag, the GraphQL Playground UI is available at http://localhost:8080/playground.
-
-### Update Swagger UI
-
-This project integrates [swagger ui] (https://swagger.io/tools/swagger-ui/) to document and test its REST API.
-The swagger documentation files can be found in `./api/`.
-You can generate the swagger-ui configuration by running `go run github.com/swaggo/swag/cmd/swag init -d ./internal/api,./pkg/schema -g rest.go -o ./api `.
-You need to move the created `./api/doc.go` to `./internal/api/doc.go`.
-If you start cc-backend with the `-dev` flag, the Swagger interface is available
-at http://localhost:8080/swagger/.
-You must enter a JWT key for a user with the API role.
-
-**NOTE**
-
-The user who owns the JWT key must not be logged into the same browser (have a
-running session), or the Swagger requests will not work. It is recommended to
-create a separate user that has only the API role.
-
-## Development and testing
-In case the REST or GraphQL API is changed the according code generators have to be used.
-
 ## Project file structure
 
-- [`api/`](https://github.com/ClusterCockpit/cc-backend/tree/master/api) contains the API schema files for the REST and GraphQL APIs. The REST API is documented in the OpenAPI 3.0 format in [./api/openapi.yaml](./api/openapi.yaml).
-- [`cmd/cc-backend`](https://github.com/ClusterCockpit/cc-backend/tree/master/cmd/cc-backend) contains `main.go` for the main application.
-- [`configs/`](https://github.com/ClusterCockpit/cc-backend/tree/master/configs) contains documentation about configuration and command line options and required environment variables. A sample configuration file is provided.
-- [`docs/`](https://github.com/ClusterCockpit/cc-backend/tree/master/docs) contains more in-depth documentation.
-- [`init/`](https://github.com/ClusterCockpit/cc-backend/tree/master/init) contains an example of setting up systemd for production use.
-- [`internal/`](https://github.com/ClusterCockpit/cc-backend/tree/master/internal) contains library source code that is not intended for use by others.
-- [`pkg/`](https://github.com/ClusterCockpit/cc-backend/tree/master/pkg) contains Go packages that can be used by other projects.
-- [`tools/`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools) Additional command line helper tools.
-   - [`archive-manager`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/archive-manager) Commands for getting infos about and existing job archive.
-   - [`archive-migration`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/archive-migration) Tool to migrate from previous to current job archive version.
-   - [`convert-pem-pubkey`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/convert-pem-pubkey) Tool to convert external pubkey for use in `cc-backend`.
-   - [`gen-keypair`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/gen-keypair) contains a small application to generate a compatible JWT keypair. You find documentation on how to use it [here](https://github.com/ClusterCockpit/cc-backend/blob/master/docs/JWT-Handling.md).
-- [`web/`](https://github.com/ClusterCockpit/cc-backend/tree/master/web) Server-side templates and frontend-related files:
-   - [`frontend`](https://github.com/ClusterCockpit/cc-backend/tree/master/web/frontend) Svelte components and static assets for the frontend UI
-   - [`templates`](https://github.com/ClusterCockpit/cc-backend/tree/master/web/templates) Server-side Go templates
-- [`gqlgen.yml`](https://github.com/ClusterCockpit/cc-backend/blob/master/gqlgen.yml) Configures the behaviour and generation of [gqlgen](https://github.com/99designs/gqlgen).
-- [`startDemo.sh`](https://github.com/ClusterCockpit/cc-backend/blob/master/startDemo.sh) is a shell script that sets up demo data, and builds and starts `cc-backend`.
-
+- [`api/`](https://github.com/ClusterCockpit/cc-backend/tree/master/api)
+  contains the API schema files for the REST and GraphQL APIs. The REST API is
+  documented in the OpenAPI 3.0 format in
+  [./api/openapi.yaml](./api/openapi.yaml).
+- [`cmd/cc-backend`](https://github.com/ClusterCockpit/cc-backend/tree/master/cmd/cc-backend)
+  contains `main.go` for the main application.
+- [`configs/`](https://github.com/ClusterCockpit/cc-backend/tree/master/configs)
+  contains documentation about configuration and command line options and required
+  environment variables. A sample configuration file is provided.
+- [`docs/`](https://github.com/ClusterCockpit/cc-backend/tree/master/docs)
+  contains more in-depth documentation.
+- [`init/`](https://github.com/ClusterCockpit/cc-backend/tree/master/init)
+  contains an example of setting up systemd for production use.
+- [`internal/`](https://github.com/ClusterCockpit/cc-backend/tree/master/internal)
+  contains library source code that is not intended for use by others.
+- [`pkg/`](https://github.com/ClusterCockpit/cc-backend/tree/master/pkg)
+  contains Go packages that can be used by other projects.
+- [`tools/`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools)
+  Additional command line helper tools.
+  - [`archive-manager`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/archive-manager)
+    Commands for getting infos about and existing job archive.
+  - [`convert-pem-pubkey`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/convert-pem-pubkey)
+    Tool to convert external pubkey for use in `cc-backend`.
+  - [`gen-keypair`](https://github.com/ClusterCockpit/cc-backend/tree/master/tools/gen-keypair)
+    contains a small application to generate a compatible JWT keypair. You find
+    documentation on how to use it
+    [here](https://github.com/ClusterCockpit/cc-backend/blob/master/docs/JWT-Handling.md).
+- [`web/`](https://github.com/ClusterCockpit/cc-backend/tree/master/web)
+  Server-side templates and frontend-related files:
+  - [`frontend`](https://github.com/ClusterCockpit/cc-backend/tree/master/web/frontend)
+    Svelte components and static assets for the frontend UI
+  - [`templates`](https://github.com/ClusterCockpit/cc-backend/tree/master/web/templates)
+    Server-side Go templates
+- [`gqlgen.yml`](https://github.com/ClusterCockpit/cc-backend/blob/master/gqlgen.yml)
+  Configures the behaviour and generation of
+  [gqlgen](https://github.com/99designs/gqlgen).
+- [`startDemo.sh`](https://github.com/ClusterCockpit/cc-backend/blob/master/startDemo.sh)
+  is a shell script that sets up demo data, and builds and starts `cc-backend`.

ReleaseNotes.md

@@ -1,36 +1,47 @@
-# `cc-backend` version 1.0.0
+# `cc-backend` version 1.4.4
 
-Supports job archive version 1 and database version 6.
+Supports job archive version 2 and database version 8.
 
-This is the initial release of `cc-backend`, the API backend and frontend
+This is a bug fix release of `cc-backend`, the API backend and frontend
 implementation of ClusterCockpit.
+For release specific notes visit the [ClusterCockpit Documentation](https://clusterockpit.org/docs/release/).
 
-** Breaking changes **
+## Breaking changes
 
-The aggregate job statistic core hours is now computed using the job table
-column `num_hwthreads`. In a future release this column will be renamed to
-`num_cores`. For correct display of core hours `num_hwthreads` must be correctly
-filled on job start. If your existing jobs do not provide the correct value in
-this column then you can set this with one SQL INSERT statement. This only applies
-if you have exclusive jobs, only. Please be aware that we treat this column as
-it is the number of cores. In case you have SMT enabled and `num_hwthreads`
-is not the number of cores the core hours will be too high by a factor!
+The option `apiAllowedIPs` is now a required configuration attribute in
+`config.json`. This option restricts access to the admin API.
 
-** NOTE **
-If you are using the sqlite3 backend the `PRAGMA` option `foreign_keys` must be
-explicitly set to ON. If using the sqlite3 console it is per default set to
-OFF!  On every console session you must set:
+To retain the previous behavior that the API is per default accessible from
+everywhere set:
+
+```json
+  "apiAllowedIPs": [
+    "*"
+  ]
 ```
-sqlite> PRAGMA foreign_keys = ON;
 
-```
-Otherwise if you delete jobs the jobtag relation table will not be updated accordingly!
+## Breaking changes for minor release 1.4.x
 
-**Notable changes**
-* Supports user roles admin, support, manager, user, and api.
-* Unified search bar supports job id, job name, project id, user name, and name
-* Performance improvements for sqlite db backend
-* Extended REST api supports to query job metrics
-* Better support for shared jobs
-* More flexible metric list configuration
-* Versioning and migration for database and job archive
+- You need to perform a database migration. Depending on your database size the
+  migration might require several hours!
+- You need to adapt the `cluster.json` configuration files in the job-archive,
+  add new required attributes to the metric list and after that edit
+  `./job-archive/version.txt` to version 2. Only metrics that have the footprint
+  attribute set can be filtered and show up in the footprint UI and polar plot.
+- Continuous scrolling is default now in all job lists. You can change this back
+  to paging globally, also every user can configure to use paging or continuous
+  scrolling individually.
+- Tags have a scope now. Existing tags will get global scope in the database
+  migration.
+
+## New features
+
+- Enable to delete tags from the web interface
+
+## Known issues
+
+- Currently energy footprint metrics of type energy are ignored for calculating
+  total energy.
+- Resampling for running jobs only works with cc-metric-store
+- With energy footprint metrics of type power the unit is ignored and it is
+  assumed the metric has the unit Watt.

api/schema.graphqls

@@ -4,133 +4,222 @@ scalar Any
 scalar NullableFloat
 scalar MetricScope
 scalar JobState
+scalar SchedulerState
+scalar MonitoringState
+
+type Node {
+  id: ID!
+  hostname: String!
+  cluster: String!
+  subCluster: String!
+  jobsRunning: Int!
+  cpusAllocated: Int
+  memoryAllocated: Int
+  gpusAllocated: Int
+  schedulerState: SchedulerState!
+  healthState: MonitoringState!
+  metaData: Any
+}
+
+type NodeStates {
+  state: String!
+  count: Int!
+}
+
+type NodeStatesTimed {
+  state: String!
+  counts: [Int!]!
+  times: [Int!]!
+}
 
 type Job {
-  id:               ID!
-  jobId:            Int!
-  user:             String!
-  project:          String!
-  cluster:          String!
-  subCluster:       String!
-  startTime:        Time!
-  duration:         Int!
-  walltime:         Int!
-  numNodes:         Int!
-  numHWThreads:     Int!
-  numAcc:           Int!
-  SMT:              Int!
-  exclusive:        Int!
-  partition:        String!
-  arrayJobId:       Int!
+  id: ID!
+  jobId: Int!
+  user: String!
+  project: String!
+  cluster: String!
+  subCluster: String!
+  startTime: Time!
+  duration: Int!
+  walltime: Int!
+  numNodes: Int!
+  numHWThreads: Int!
+  numAcc: Int!
+  energy: Float!
+  SMT: Int!
+  shared: String!
+  partition: String!
+  arrayJobId: Int!
   monitoringStatus: Int!
-  state:            JobState!
-  tags:             [Tag!]!
-  resources:        [Resource!]!
-  concurrentJobs:   JobLinkResultList
-
-  metaData:         Any
-  userData:         User
+  state: JobState!
+  tags: [Tag!]!
+  resources: [Resource!]!
+  concurrentJobs: JobLinkResultList
+  footprint: [FootprintValue]
+  energyFootprint: [EnergyFootprintValue]
+  metaData: Any
+  userData: User
 }
 
 type JobLink {
-  id:               ID!
-  jobId:            Int!
+  id: ID!
+  jobId: Int!
 }
 
 type Cluster {
-  name:         String!
-  partitions:   [String!]!        # Slurm partitions
-  metricConfig: [MetricConfig!]!
-  subClusters:  [SubCluster!]!    # Hardware partitions/subclusters
+  name: String!
+  partitions: [String!]! # Slurm partitions
+  subClusters: [SubCluster!]! # Hardware partitions/subclusters
 }
 
 type SubCluster {
-  name:            String!
-  nodes:           String!
-  numberOfNodes:   Int!
-  processorType:   String!
-  socketsPerNode:  Int!
-  coresPerSocket:  Int!
-  threadsPerCore:  Int!
-  flopRateScalar:  MetricValue!
-  flopRateSimd:    MetricValue!
+  name: String!
+  nodes: String!
+  numberOfNodes: Int!
+  processorType: String!
+  socketsPerNode: Int!
+  coresPerSocket: Int!
+  threadsPerCore: Int!
+  flopRateScalar: MetricValue!
+  flopRateSimd: MetricValue!
   memoryBandwidth: MetricValue!
-  topology:        Topology!
+  topology: Topology!
+  metricConfig: [MetricConfig!]!
+  footprint: [String!]!
+}
+
+type FootprintValue {
+  name: String!
+  stat: String!
+  value: Float!
+}
+
+type EnergyFootprintValue {
+  hardware: String!
+  metric: String!
+  value: Float!
 }
 
 type MetricValue {
+  name: String
   unit: Unit!
   value: Float!
 }
 
 type Topology {
-  node:         [Int!]
-  socket:       [[Int!]!]
+  node: [Int!]
+  socket: [[Int!]!]
   memoryDomain: [[Int!]!]
-  die:          [[Int!]!]
-  core:         [[Int!]!]
+  die: [[Int!]!]
+  core: [[Int!]!]
   accelerators: [Accelerator!]
 }
 
 type Accelerator {
-  id:    String!
-  type:  String!
+  id: String!
+  type: String!
   model: String!
 }
 
 type SubClusterConfig {
-  name:    String!
-  peak:    Float
-  normal:  Float
+  name: String!
+  peak: Float
+  normal: Float
   caution: Float
-  alert:   Float
-  remove:  Boolean
+  alert: Float
+  remove: Boolean
 }
 
 type MetricConfig {
-  name:        String!
-  unit:        Unit!
-  scope:       MetricScope!
+  name: String!
+  unit: Unit!
+  scope: MetricScope!
   aggregation: String!
-  timestep:    Int!
-  peak:    Float!
-  normal:  Float
+  timestep: Int!
+  peak: Float!
+  normal: Float
   caution: Float!
-  alert:   Float!
+  alert: Float!
+  lowerIsBetter: Boolean
   subClusters: [SubClusterConfig!]!
 }
 
 type Tag {
-  id:   ID!
+  id: ID!
   type: String!
   name: String!
+  scope: String!
 }
 
 type Resource {
-  hostname:      String!
-  hwthreads:     [Int!]
-  accelerators:  [String!]
+  hostname: String!
+  hwthreads: [Int!]
+  accelerators: [String!]
   configuration: String
 }
 
 type JobMetricWithName {
-  name:   String!
-  scope:  MetricScope!
+  name: String!
+  scope: MetricScope!
   metric: JobMetric!
 }
 
+type ClusterMetricWithName {
+  name: String!
+  unit: Unit
+  timestep: Int!
+  data: [NullableFloat!]!
+}
+
 type JobMetric {
-  unit:             Unit
-  timestep:         Int!
-  series:           [Series!]
+  unit: Unit
+  timestep: Int!
+  series: [Series!]
   statisticsSeries: StatsSeries
 }
 
 type Series {
-  hostname:   String!
-  id:         String
+  hostname: String!
+  id: String
   statistics: MetricStatistics
-  data:       [NullableFloat!]!
+  data: [NullableFloat!]!
+}
+
+type StatsSeries {
+  mean: [NullableFloat!]!
+  median: [NullableFloat!]!
+  min: [NullableFloat!]!
+  max: [NullableFloat!]!
+}
+
+type NamedStatsWithScope {
+  name: String!
+  scope: MetricScope!
+  stats: [ScopedStats!]!
+}
+
+type ScopedStats {
+  hostname: String!
+  id: String
+  data: MetricStatistics!
+}
+
+type JobStats {
+  id: Int!
+  jobId: String!
+  startTime: Int!
+  duration: Int!
+  cluster: String!
+  subCluster: String!
+  numNodes: Int!
+  numHWThreads: Int
+  numAccelerators: Int
+  stats: [NamedStats!]!
+}
+
+type NamedStats {
+  name: String!
+  data: MetricStatistics!
 }
 
 type Unit {
@@ -144,104 +233,237 @@ type MetricStatistics {
   max: Float!
 }
 
-type StatsSeries {
-  mean: [NullableFloat!]!
-  min:  [NullableFloat!]!
-  max:  [NullableFloat!]!
-}
-
 type MetricFootprints {
   metric: String!
-  data:   [NullableFloat!]!
+  data: [NullableFloat!]!
 }
 
 type Footprints {
-  nodehours: [NullableFloat!]!
-  metrics:   [MetricFootprints!]!
+  timeWeights: TimeWeights!
+  metrics: [MetricFootprints!]!
 }
 
-enum Aggregate { USER, PROJECT, CLUSTER }
-enum Weights { NODE_COUNT, NODE_HOURS }
+type TimeWeights {
+  nodeHours: [NullableFloat!]!
+  accHours: [NullableFloat!]!
+  coreHours: [NullableFloat!]!
+}
+
+enum Aggregate {
+  USER
+  PROJECT
+  CLUSTER
+  SUBCLUSTER
+}
+enum SortByAggregate {
+  TOTALWALLTIME
+  TOTALJOBS
+  TOTALUSERS
+  TOTALNODES
+  TOTALNODEHOURS
+  TOTALCORES
+  TOTALCOREHOURS
+  TOTALACCS
+  TOTALACCHOURS
+}
 
 type NodeMetrics {
-  host:       String!
+  host: String!
+  state: String!
   subCluster: String!
-  metrics:    [JobMetricWithName!]!
+  metrics: [JobMetricWithName!]!
+}
+
+type ClusterMetrics {
+  nodeCount: Int!
+  metrics: [ClusterMetricWithName!]!
+}
+
+type NodesResultList {
+  items: [NodeMetrics!]!
+  offset: Int
+  limit: Int
+  count: Int
+  totalNodes: Int
+  hasNextPage: Boolean
+}
+
+type ClusterSupport {
+  cluster: String!
+  subClusters: [String!]!
+}
+
+type GlobalMetricListItem {
+  name: String!
+  unit: Unit!
+  scope: MetricScope!
+  footprint: String
+  availability: [ClusterSupport!]!
 }
 
 type Count {
-  name:  String!
+  name: String!
   count: Int!
 }
 
 type User {
   username: String!
-  name:     String!
-  email:    String!
+  name: String!
+  email: String!
+}
+
+input MetricStatItem {
+  metricName: String!
+  range: FloatRange!
 }
 
 type Query {
-  clusters:     [Cluster!]!   # List of all clusters
-  tags:         [Tag!]!       # List of all tags
+  clusters: [Cluster!]! # List of all clusters
+  tags: [Tag!]! # List of all tags
+  globalMetrics: [GlobalMetricListItem!]!
 
   user(username: String!): User
   allocatedNodes(cluster: String!): [Count!]!
 
+  ## Node Queries New
+  node(id: ID!): Node
+  nodes(filter: [NodeFilter!], order: OrderByInput): NodeStateResultList!
+  nodeStates(filter: [NodeFilter!]): [NodeStates!]!
+  nodeStatesTimed(filter: [NodeFilter!], type: String!): [NodeStatesTimed!]!
+
   job(id: ID!): Job
-  jobMetrics(id: ID!, metrics: [String!], scopes: [MetricScope!]): [JobMetricWithName!]!
+  jobMetrics(
+    id: ID!
+    metrics: [String!]
+    scopes: [MetricScope!]
+    resolution: Int
+  ): [JobMetricWithName!]!
+
+  jobStats(id: ID!, metrics: [String!]): [NamedStats!]!
+
+  scopedJobStats(
+    id: ID!
+    metrics: [String!]
+    scopes: [MetricScope!]
+  ): [NamedStatsWithScope!]!
+
+  jobs(
+    filter: [JobFilter!]
+    page: PageRequest
+    order: OrderByInput
+  ): JobResultList!
+
+  jobsStatistics(
+    filter: [JobFilter!]
+    metrics: [String!]
+    page: PageRequest
+    sortBy: SortByAggregate
+    groupBy: Aggregate
+    numDurationBins: String
+    numMetricBins: Int
+  ): [JobsStatistics!]!
+
+  jobsMetricStats(filter: [JobFilter!], metrics: [String!]): [JobStats!]!
   jobsFootprints(filter: [JobFilter!], metrics: [String!]!): Footprints
 
-  jobs(filter: [JobFilter!], page: PageRequest, order: OrderByInput): JobResultList!
-  jobsStatistics(filter: [JobFilter!], groupBy: Aggregate): [JobsStatistics!]!
-  jobsCount(filter: [JobFilter]!, groupBy: Aggregate!, weight: Weights, limit: Int): [Count!]!
+  rooflineHeatmap(
+    filter: [JobFilter!]!
+    rows: Int!
+    cols: Int!
+    minX: Float!
+    minY: Float!
+    maxX: Float!
+    maxY: Float!
+  ): [[Float!]!]!
 
-  rooflineHeatmap(filter: [JobFilter!]!, rows: Int!, cols: Int!, minX: Float!, minY: Float!, maxX: Float!, maxY: Float!): [[Float!]!]!
+  nodeMetrics(
+    cluster: String!
+    nodes: [String!]
+    scopes: [MetricScope!]
+    metrics: [String!]
+    from: Time!
+    to: Time!
+  ): [NodeMetrics!]!
 
-  nodeMetrics(cluster: String!, nodes: [String!], scopes: [MetricScope!], metrics: [String!], from: Time!, to: Time!): [NodeMetrics!]!
+  nodeMetricsList(
+    cluster: String!
+    subCluster: String!
+    stateFilter: String!
+    nodeFilter: String!
+    scopes: [MetricScope!]
+    metrics: [String!]
+    from: Time!
+    to: Time!
+    page: PageRequest
+    resolution: Int
+  ): NodesResultList!
+
+  clusterMetrics(
+    cluster: String!
+    metrics: [String!]
+    from: Time!
+    to: Time!
+  ): ClusterMetrics!
 }
 
 type Mutation {
-  createTag(type: String!, name: String!): Tag!
+  createTag(type: String!, name: String!, scope: String!): Tag!
   deleteTag(id: ID!): ID!
   addTagsToJob(job: ID!, tagIds: [ID!]!): [Tag!]!
   removeTagsFromJob(job: ID!, tagIds: [ID!]!): [Tag!]!
+  removeTagFromList(tagIds: [ID!]!): [Int!]!
 
   updateConfiguration(name: String!, value: String!): String
 }
 
-type IntRangeOutput { from: Int!, to: Int! }
-type TimeRangeOutput { from: Time!, to: Time! }
+type IntRangeOutput {
+  from: Int!
+  to: Int!
+}
+type TimeRangeOutput {
+  range: String
+  from: Time!
+  to: Time!
+}
+
+input NodeFilter {
+  hostname: StringInput
+  cluster: StringInput
+  subcluster: StringInput
+  schedulerState: SchedulerState
+  healthState: MonitoringState
+  timeStart: Int
+}
 
 input JobFilter {
-  tags:        [ID!]
-  jobId:       StringInput
-  arrayJobId:  Int
-  user:        StringInput
-  project:     StringInput
-  jobName:     StringInput
-  cluster:     StringInput
-  partition:   StringInput
-  duration:    IntRange
+  tags: [ID!]
+  dbId: [ID!]
+  jobId: StringInput
+  arrayJobId: Int
+  user: StringInput
+  project: StringInput
+  jobName: StringInput
+  cluster: StringInput
+  partition: StringInput
+  duration: IntRange
+  energy: FloatRange
 
   minRunningFor: Int
 
-  numNodes:        IntRange
+  numNodes: IntRange
   numAccelerators: IntRange
-  numHWThreads:    IntRange
+  numHWThreads: IntRange
 
-  startTime:   TimeRange
-  state:       [JobState!]
-  flopsAnyAvg: FloatRange
-  memBwAvg:    FloatRange
-  loadAvg:     FloatRange
-  memUsedMax:  FloatRange
-
-  exclusive:     Int
-  node:    StringInput
+  startTime: TimeRange
+  state: [JobState!]
+  metricStats: [MetricStatItem!]
+  shared: String
+  node: StringInput
 }
 
 input OrderByInput {
   field: String!
+  type: String!
   order: SortDirectionEnum! = ASC
 }
 
@@ -251,29 +473,46 @@ enum SortDirectionEnum {
 }
 
 input StringInput {
-  eq:         String
-  neq:        String
-  contains:   String
+  eq: String
+  neq: String
+  contains: String
   startsWith: String
-  endsWith:   String
-  in:         [String!]
+  endsWith: String
+  in: [String!]
 }
 
-input IntRange   { from: Int!,   to: Int! }
-input FloatRange { from: Float!, to: Float! }
-input TimeRange  { from: Time,   to: Time }
+input IntRange {
+  from: Int!
+  to: Int!
+}
+input TimeRange {
+  range: String
+  from: Time
+  to: Time
+}
+
+input FloatRange {
+  from: Float!
+  to: Float!
+}
+
+type NodeStateResultList {
+  items: [Node!]!
+  count: Int
+}
 
 type JobResultList {
-  items:  [Job!]!
+  items: [Job!]!
   offset: Int
-  limit:  Int
-  count:  Int
+  limit: Int
+  count: Int
+  hasNextPage: Boolean
 }
 
 type JobLinkResultList {
   listQuery: String
-  items:  [JobLink!]!
-  count:  Int
+  items: [JobLink!]!
+  count: Int
 }
 
 type HistoPoint {
@@ -281,21 +520,42 @@ type HistoPoint {
   value: Int!
 }
 
-type JobsStatistics  {
-  id:             ID!            # If `groupBy` was used, ID of the user/project/cluster
-  name:           String!        # if User-Statistics: Given Name of Account (ID) Owner
-  totalJobs:      Int!           # Number of jobs
-  runningJobs:    Int!           # Number of running jobs
-  shortJobs:      Int!           # Number of jobs with a duration of less than duration
-  totalWalltime:  Int!           # Sum of the duration of all matched jobs in hours
-  totalNodeHours: Int!           # Sum of the node hours of all matched jobs
-  totalCoreHours: Int!           # Sum of the core hours of all matched jobs
-  totalAccHours:  Int!           # Sum of the gpu hours of all matched jobs
-  histDuration:   [HistoPoint!]! # value: hour, count: number of jobs with a rounded duration of value
-  histNumNodes:   [HistoPoint!]! # value: number of nodes, count: number of jobs with that number of nodes
+type MetricHistoPoints {
+  metric: String!
+  unit: String!
+  stat: String
+  data: [MetricHistoPoint!]
+}
+
+type MetricHistoPoint {
+  bin: Int
+  count: Int!
+  min: Int
+  max: Int
+}
+
+type JobsStatistics {
+  id: ID! # If `groupBy` was used, ID of the user/project/cluster/subcluster
+  name: String! # if User-Statistics: Given Name of Account (ID) Owner
+  totalUsers: Int! # if *not* User-Statistics: Number of active users (based on running jobs)
+  totalJobs: Int! # Number of jobs
+  runningJobs: Int! # Number of running jobs
+  shortJobs: Int! # Number of jobs with a duration of less than config'd ShortRunningJobsDuration
+  totalWalltime: Int! # Sum of the duration of all matched jobs in hours
+  totalNodes: Int! # Sum of the nodes of all matched jobs
+  totalNodeHours: Int! # Sum of the node hours of all matched jobs
+  totalCores: Int! # Sum of the cores of all matched jobs
+  totalCoreHours: Int! # Sum of the core hours of all matched jobs
+  totalAccs: Int! # Sum of the accs of all matched jobs
+  totalAccHours: Int! # Sum of the gpu hours of all matched jobs
+  histDuration: [HistoPoint!]! # value: hour, count: number of jobs with a rounded duration of value
+  histNumNodes: [HistoPoint!]! # value: number of nodes, count: number of jobs with that number of nodes
+  histNumCores: [HistoPoint!]! # value: number of cores, count: number of jobs with that number of cores
+  histNumAccs: [HistoPoint!]! # value: number of accs, count: number of jobs with that number of accs
+  histMetrics: [MetricHistoPoints!]! # metric: metricname, data array of histopoints: value: metric average bin, count: number of jobs with that metric average
 }
 
 input PageRequest {
   itemsPerPage: Int!
-  page:         Int!
+  page: Int!
 }

build/package/cc-backend.config

@@ -1,17 +0,0 @@
-CC_USER=clustercockpit
-
-CC_GROUP=clustercockpit
-
-CC_HOME=/tmp
-
-LOG_DIR=/var/log
-
-DATA_DIR=/var/run/cc-backend
-
-MAX_OPEN_FILES=10000
-
-CONF_DIR=/etc/cc-backend
-
-CONF_FILE=/etc/cc-backend/cc-backend.json
-
-RESTART_ON_UPGRADE=true

build/package/cc-backend.deb.control

@@ -1,12 +0,0 @@
-Package: cc-backend
-Version: {VERSION}
-Installed-Size: {INSTALLED_SIZE}
-Architecture: {ARCH}
-Maintainer: thomas.gruber@fau.de
-Depends: libc6 (>= 2.2.1)
-Build-Depends: debhelper-compat (= 13), git, golang-go, npm, yarn
-Description: ClusterCockpit backend and web frontend
-Homepage: https://github.com/ClusterCockpit/cc-backend
-Source: cc-backend
-Rules-Requires-Root: no
-

build/package/cc-backend.service

@@ -1,18 +0,0 @@
-[Unit]
-Description=ClusterCockpit backend and web frontend (cc-backend)
-Documentation=https://github.com/ClusterCockpit/cc-backend
-Wants=network-online.target
-After=network-online.target
-
-[Service]
-EnvironmentFile=/etc/default/cc-backend
-Type=simple
-User=clustercockpit
-Group=clustercockpit
-Restart=on-failure
-TimeoutStopSec=100
-LimitNOFILE=infinity
-ExecStart=/usr/bin/cc-backend --config ${CONF_FILE}
-
-[Install]
-WantedBy=multi-user.target

build/package/cc-backend.spec

@@ -1,70 +0,0 @@
-Name:           cc-backend
-Version:        %{VERS}
-Release:        1%{?dist}
-Summary:        ClusterCockpit backend and web frontend
-
-License:        MIT
-Source0:        %{name}-%{version}.tar.gz
-
-#BuildRequires:  go-toolset
-#BuildRequires:  systemd-rpm-macros
-#BuildRequires:  npm
-
-Provides:       %{name} = %{version}
-
-%description
-ClusterCockpit backend and web frontend
-
-%global debug_package %{nil}
-
-%prep
-%autosetup
-
-
-%build
-#CURRENT_TIME=$(date +%Y-%m-%d:T%H:%M:\%S)
-#LD_FLAGS="-s -X main.buildTime=${CURRENT_TIME} -X main.version=%{VERS}"
-mkdir ./var
-touch ./var/job.db
-cd web/frontend && yarn install && yarn build && cd -
-go build -ldflags="-s -X main.version=%{VERS}" ./cmd/cc-backend
-
-
-%install
-# Install cc-backend
-#make PREFIX=%{buildroot} install
-install -Dpm 755 cc-backend %{buildroot}/%{_bindir}/%{name}
-install -Dpm 0600 configs/config.json %{buildroot}%{_sysconfdir}/%{name}/%{name}.json
-# Integrate into system
-install -Dpm 0644 build/package/%{name}.service %{buildroot}%{_unitdir}/%{name}.service
-install -Dpm 0600 build/package/%{name}.config %{buildroot}%{_sysconfdir}/default/%{name}
-install -Dpm 0644 build/package/%{name}.sysusers %{buildroot}%{_sysusersdir}/%{name}.conf
-
-
-%check
-# go test should be here... :)
-
-%pre
-%sysusers_create_package scripts/%{name}.sysusers
-
-%post
-%systemd_post %{name}.service
-
-%preun
-%systemd_preun %{name}.service
-
-%files
-# Binary
-%attr(-,clustercockpit,clustercockpit) %{_bindir}/%{name}
-# Config
-%dir %{_sysconfdir}/%{name}
-%attr(0600,clustercockpit,clustercockpit) %config(noreplace) %{_sysconfdir}/%{name}/%{name}.json
-# Systemd
-%{_unitdir}/%{name}.service
-%{_sysconfdir}/default/%{name}
-%{_sysusersdir}/%{name}.conf
-
-%changelog
-* Mon Mar 07 2022 Thomas Gruber - 0.1
-- Initial metric store implementation
-

build/package/cc-backend.sysusers

@@ -1,2 +0,0 @@
-#Type Name           ID GECOS                     Home directory           Shell
-u     clustercockpit -  "User for ClusterCockpit" /run/cc-backend /sbin/nologin

cmd/cc-backend/cli.go

@@ -0,0 +1,38 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+// Package main provides the entry point for the ClusterCockpit backend server.
+// This file defines all command-line flags and their default values.
+package main
+
+import "flag"
+
+var (
+	flagReinitDB, flagInit, flagServer, flagSyncLDAP, flagGops, flagMigrateDB, flagRevertDB,
+	flagForceDB, flagDev, flagVersion, flagLogDateTime, flagApplyTags bool
+	flagNewUser, flagDelUser, flagGenJWT, flagConfigFile, flagImportJob, flagLogLevel string
+)
+
+func cliInit() {
+	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")
+	flag.BoolVar(&flagGops, "gops", false, "Listen via github.com/google/gops/agent (for debugging)")
+	flag.BoolVar(&flagDev, "dev", false, "Enable development components: GraphQL Playground and Swagger UI")
+	flag.BoolVar(&flagVersion, "version", false, "Show version information and exit")
+	flag.BoolVar(&flagMigrateDB, "migrate-db", false, "Migrate database to supported version and exit")
+	flag.BoolVar(&flagRevertDB, "revert-db", false, "Migrate database to previous version and exit")
+	flag.BoolVar(&flagApplyTags, "apply-tags", false, "Run taggers on all completed jobs and exit")
+	flag.BoolVar(&flagForceDB, "force-db", false, "Force database version, clear dirty flag and exit")
+	flag.BoolVar(&flagLogDateTime, "logdate", false, "Set this flag to add date and time to log messages")
+	flag.StringVar(&flagConfigFile, "config", "./config.json", "Specify alternative path to `config.json`")
+	flag.StringVar(&flagNewUser, "add-user", "", "Add a new user. Argument format: <username>:[admin,support,manager,api,user]:<password>")
+	flag.StringVar(&flagDelUser, "del-user", "", "Remove a existing user. Argument format: <username>")
+	flag.StringVar(&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, crit]`")
+	flag.Parse()
+}

cmd/cc-backend/init.go

@@ -0,0 +1,119 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+// Package main provides the entry point for the ClusterCockpit backend server.
+// This file contains bootstrap logic for initializing the environment,
+// creating default configuration files, and setting up the database.
+package main
+
+import (
+	"encoding/json"
+	"os"
+
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	"github.com/ClusterCockpit/cc-backend/pkg/archive"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/util"
+)
+
+const envString = `
+# Base64 encoded Ed25519 keys (DO NOT USE THESE TWO IN PRODUCTION!)
+# You can generate your own keypair using the gen-keypair tool
+JWT_PUBLIC_KEY="kzfYrYy+TzpanWZHJ5qSdMj5uKUWgq74BWhQG6copP0="
+JWT_PRIVATE_KEY="dtPC/6dWJFKZK7KZ78CvWuynylOmjBFyMsUWArwmodOTN9itjL5POlqdZkcnmpJ0yPm4pRaCrvgFaFAbpyik/Q=="
+
+# Some random bytes used as secret for cookie-based sessions (DO NOT USE THIS ONE IN PRODUCTION)
+SESSION_KEY="67d829bf61dc5f87a73fd814e2c9f629"
+`
+
+const configString = `
+{
+  "main": {
+    "addr": "127.0.0.1:8080",
+    "short-running-jobs-duration": 300,
+    "resampling": {
+      "minimumPoints": 600,
+      "trigger": 180,
+      "resolutions": [
+        240,
+        60
+      ]
+    },
+    "apiAllowedIPs": [
+      "*"
+    ],
+    "emission-constant": 317
+  },
+  "cron": {
+    "commit-job-worker": "2m",
+    "duration-worker": "5m",
+    "footprint-worker": "10m"
+  },
+  "archive": {
+    "kind": "file",
+    "path": "./var/job-archive"
+  },
+  "auth": {
+    "jwts": {
+      "max-age": "2000h"
+    }
+  },
+ "clusters": [
+     {
+         "name": "name",
+         "metricDataRepository": {
+             "kind": "cc-metric-store",
+             "url": "http://localhost:8082",
+             "token": ""
+         },
+         "filterRanges": {
+             "numNodes": {
+                 "from": 1,
+                 "to": 64
+             },
+             "duration": {
+                 "from": 0,
+                 "to": 86400
+             },
+             "startTime": {
+                 "from": "2023-01-01T00:00:00Z",
+                 "to": null
+             }
+         }
+     }
+ ]
+}
+`
+
+func initEnv() {
+	if util.CheckFileExists("var") {
+		cclog.Exit("Directory ./var already exists. Cautiously exiting application initialization.")
+	}
+
+	if err := os.WriteFile("config.json", []byte(configString), 0o666); err != nil {
+		cclog.Abortf("Could not write default ./config.json with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
+	}
+
+	if err := os.WriteFile(".env", []byte(envString), 0o666); err != nil {
+		cclog.Abortf("Could not write default ./.env file with permissions '0o666'. Application initialization failed, exited.\nError: %s\n", err.Error())
+	}
+
+	if err := os.Mkdir("var", 0o777); err != nil {
+		cclog.Abortf("Could not create default ./var folder with permissions '0o777'. Application initialization failed, exited.\nError: %s\n", err.Error())
+	}
+
+	err := repository.MigrateDB("sqlite3", "./var/job.db")
+	if err != nil {
+		cclog.Abortf("Could not initialize default sqlite3 database as './var/job.db'. Application initialization failed, exited.\nError: %s\n", err.Error())
+	}
+	if err := os.Mkdir("var/job-archive", 0o777); err != nil {
+		cclog.Abortf("Could not create default ./var/job-archive folder with permissions '0o777'. Application initialization failed, exited.\nError: %s\n", err.Error())
+	}
+	archiveCfg := "{\"kind\": \"file\",\"path\": \"./var/job-archive\"}"
+	if err := archive.Init(json.RawMessage(archiveCfg), config.Keys.DisableArchive); err != nil {
+		cclog.Abortf("Could not initialize job-archive, exited.\nError: %s\n", err.Error())
+	}
+}

cmd/cc-backend/main.go

@@ -1,107 +1,67 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
+// Package main provides the entry point for the ClusterCockpit backend server.
+// It orchestrates initialization of all subsystems including configuration,
+// database, authentication, and the HTTP server.
 package main
 
 import (
 	"context"
-	"crypto/tls"
 	"encoding/json"
-	"errors"
-	"flag"
 	"fmt"
-	"io"
-	"net"
-	"net/http"
 	"os"
 	"os/signal"
-	"runtime"
 	"runtime/debug"
 	"strings"
 	"sync"
 	"syscall"
 	"time"
 
-	"github.com/99designs/gqlgen/graphql/handler"
-	"github.com/99designs/gqlgen/graphql/playground"
-	"github.com/ClusterCockpit/cc-backend/internal/api"
+	"github.com/ClusterCockpit/cc-backend/internal/archiver"
 	"github.com/ClusterCockpit/cc-backend/internal/auth"
 	"github.com/ClusterCockpit/cc-backend/internal/config"
-	"github.com/ClusterCockpit/cc-backend/internal/graph"
-	"github.com/ClusterCockpit/cc-backend/internal/graph/generated"
 	"github.com/ClusterCockpit/cc-backend/internal/importer"
+	"github.com/ClusterCockpit/cc-backend/internal/memorystore"
 	"github.com/ClusterCockpit/cc-backend/internal/metricdata"
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
-	"github.com/ClusterCockpit/cc-backend/internal/routerConfig"
-	"github.com/ClusterCockpit/cc-backend/internal/runtimeEnv"
-	"github.com/ClusterCockpit/cc-backend/internal/util"
+	"github.com/ClusterCockpit/cc-backend/internal/tagger"
+	"github.com/ClusterCockpit/cc-backend/internal/taskmanager"
 	"github.com/ClusterCockpit/cc-backend/pkg/archive"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	"github.com/ClusterCockpit/cc-backend/pkg/nats"
 	"github.com/ClusterCockpit/cc-backend/web"
-	"github.com/go-co-op/gocron"
+	ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/runtimeEnv"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/ClusterCockpit/cc-lib/util"
 	"github.com/google/gops/agent"
-	"github.com/gorilla/handlers"
-	"github.com/gorilla/mux"
-	httpSwagger "github.com/swaggo/http-swagger"
+	"github.com/joho/godotenv"
 
 	_ "github.com/go-sql-driver/mysql"
 	_ "github.com/mattn/go-sqlite3"
 )
 
 const logoString = `
- ____ _           _             ____           _          _ _
-/ ___| |_   _ ___| |_ ___ _ __ / ___|___   ___| | ___ __ (_) |_
+ _____ _           _             ____           _          _ _
+/  ___| |_   _ ___| |_ ___ _ __ / ___|___   ___| | ___ __ (_) |_
 | |   | | | | / __| __/ _ \ '__| |   / _ \ / __| |/ / '_ \| | __|
 | |___| | |_| \__ \ ||  __/ |  | |__| (_) | (__|   <| |_) | | |_
-\____|_|\__,_|___/\__\___|_|   \____\___/ \___|_|\_\ .__/|_|\__|
+\_____|_|\__,_|___/\__\___|_|   \____\___/ \___|_|\_\ .__/|_|\__|
                                                     |_|
 `
 
-const envString = `
-# Base64 encoded Ed25519 keys (DO NOT USE THESE TWO IN PRODUCTION!)
-# You can generate your own keypair using the gen-keypair tool
-JWT_PUBLIC_KEY="kzfYrYy+TzpanWZHJ5qSdMj5uKUWgq74BWhQG6copP0="
-JWT_PRIVATE_KEY="dtPC/6dWJFKZK7KZ78CvWuynylOmjBFyMsUWArwmodOTN9itjL5POlqdZkcnmpJ0yPm4pRaCrvgFaFAbpyik/Q=="
+// Environment variable names
+const (
+	envGOGC = "GOGC"
+)
 
-# Some random bytes used as secret for cookie-based sessions (DO NOT USE THIS ONE IN PRODUCTION)
-SESSION_KEY="67d829bf61dc5f87a73fd814e2c9f629"
-`
-
-const configString = `
-{
-    "addr": "127.0.0.1:8080",
-    "archive": {
-        "kind": "file",
-        "path": "./var/job-archive"
-    },
-    "clusters": [
-        {
-            "name": "name",
-            "metricDataRepository": {
-                "kind": "cc-metric-store",
-                "url": "http://localhost:8082",
-                "token": ""
-            },
-            "filterRanges": {
-                "numNodes": {
-                    "from": 1,
-                    "to": 64
-                },
-                "duration": {
-                    "from": 0,
-                    "to": 86400
-                },
-                "startTime": {
-                    "from": "2023-01-01T00:00:00Z",
-                    "to": null
-                }
-            }
-        }
-    ]
-}
-`
+// Default configurations
+const (
+	defaultArchiveConfig = `{"kind":"file","path":"./var/job-archive"}`
+)
 
 var (
 	date    string
@@ -109,507 +69,420 @@ var (
 	version string
 )
 
-func initEnv() {
-	if util.CheckFileExists("var") {
-		fmt.Print("Directory ./var already exists. Exiting!\n")
-		os.Exit(0)
-	}
-
-	if err := os.WriteFile("config.json", []byte(configString), 0666); err != nil {
-		log.Fatalf("Writing config.json failed: %s", err.Error())
-	}
-
-	if err := os.WriteFile(".env", []byte(envString), 0666); err != nil {
-		log.Fatalf("Writing .env failed: %s", err.Error())
-	}
-
-	if err := os.Mkdir("var", 0777); err != nil {
-		log.Fatalf("Mkdir var failed: %s", err.Error())
-	}
-
-	err := repository.MigrateDB("sqlite3", "./var/job.db")
-	if err != nil {
-		log.Fatalf("Initialize job.db failed: %s", err.Error())
-	}
+func printVersion() {
+	fmt.Print(logoString)
+	fmt.Printf("Version:\t%s\n", version)
+	fmt.Printf("Git hash:\t%s\n", commit)
+	fmt.Printf("Build time:\t%s\n", date)
+	fmt.Printf("SQL db version:\t%d\n", repository.Version)
+	fmt.Printf("Job archive version:\t%d\n", archive.Version)
 }
 
-func main() {
-	var flagReinitDB, flagInit, flagServer, flagSyncLDAP, flagGops, flagMigrateDB, flagDev, flagVersion, flagLogDateTime bool
-	var flagNewUser, flagDelUser, flagGenJWT, flagConfigFile, flagImportJob, flagLogLevel string
-	flag.BoolVar(&flagInit, "init", false, "Setup var directory, initialize swlite database file, config.json and .env")
-	flag.BoolVar(&flagReinitDB, "init-db", false, "Go through job-archive and re-initialize the 'job', 'tag', and 'jobtag' tables (all running jobs will be lost!)")
-	flag.BoolVar(&flagSyncLDAP, "sync-ldap", false, "Sync the 'user' table with ldap")
-	flag.BoolVar(&flagServer, "server", false, "Start a server, continues listening on port after initialization and argument handling")
-	flag.BoolVar(&flagGops, "gops", false, "Listen via github.com/google/gops/agent (for debugging)")
-	flag.BoolVar(&flagDev, "dev", false, "Enable development components: GraphQL Playground and Swagger UI")
-	flag.BoolVar(&flagVersion, "version", false, "Show version information and exit")
-	flag.BoolVar(&flagMigrateDB, "migrate-db", false, "Migrate database to supported version and exit")
-	flag.BoolVar(&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(&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.Parse()
-
-	if flagVersion {
-		fmt.Print(logoString)
-		fmt.Printf("Version:\t%s\n", version)
-		fmt.Printf("Git hash:\t%s\n", commit)
-		fmt.Printf("Build time:\t%s\n", date)
-		fmt.Printf("SQL db version:\t%d\n", repository.Version)
-		fmt.Printf("Job archive version:\t%d\n", archive.Version)
-		os.Exit(0)
+func initGops() error {
+	if !flagGops {
+		return nil
 	}
 
-	// Apply config flags for pkg/log
-	log.Init(flagLogLevel, flagLogDateTime)
+	if err := agent.Listen(agent.Options{}); err != nil {
+		return fmt.Errorf("starting gops agent: %w", err)
+	}
+	return nil
+}
 
-	if flagInit {
-		initEnv()
-		fmt.Print("Succesfully 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)
+func loadEnvironment() error {
+	if err := godotenv.Load(); err != nil {
+		return fmt.Errorf("loading .env file: %w", err)
+	}
+	return nil
+}
+
+func initConfiguration() error {
+	ccconf.Init(flagConfigFile)
+
+	cfg := ccconf.GetPackageConfig("main")
+	if cfg == nil {
+		return fmt.Errorf("main configuration must be present")
 	}
 
-	// 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())
-		}
+	clustercfg := ccconf.GetPackageConfig("clusters")
+	if clustercfg == nil {
+		return fmt.Errorf("cluster configuration must be present")
 	}
 
-	if err := runtimeEnv.LoadEnv("./.env"); err != nil && !os.IsNotExist(err) {
-		log.Fatalf("parsing './.env' file failed: %s", err.Error())
-	}
+	config.Init(cfg, clustercfg)
+	return nil
+}
 
-	// Initialize sub-modules and handle command line flags.
-	// The order here is important!
-	config.Init(flagConfigFile)
-
-	// As a special case for `db`, allow using an environment variable instead of the value
-	// stored in the config. This can be done for people having security concerns about storing
-	// the password for their mysql database in config.json.
-	if strings.HasPrefix(config.Keys.DB, "env:") {
-		envvar := strings.TrimPrefix(config.Keys.DB, "env:")
-		config.Keys.DB = os.Getenv(envvar)
-	}
+func initDatabase() error {
+	repository.Connect(config.Keys.DBDriver, config.Keys.DB)
+	return nil
+}
 
+func handleDatabaseCommands() error {
 	if flagMigrateDB {
 		err := repository.MigrateDB(config.Keys.DBDriver, config.Keys.DB)
 		if err != nil {
-			log.Fatal(err)
+			return fmt.Errorf("migrating database to version %d: %w", repository.Version, err)
 		}
-		os.Exit(0)
+		cclog.Exitf("MigrateDB Success: Migrated '%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)
-	db := repository.GetConnection()
+	if flagRevertDB {
+		err := repository.RevertDB(config.Keys.DBDriver, config.Keys.DB)
+		if err != nil {
+			return fmt.Errorf("reverting database to version %d: %w", repository.Version-1, err)
+		}
+		cclog.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 {
+			return fmt.Errorf("forcing database to version %d: %w", repository.Version, err)
+		}
+		cclog.Exitf("ForceDB Success: Forced '%s' database at location '%s' to version %d.\n",
+			config.Keys.DBDriver, config.Keys.DB, repository.Version)
+	}
+
+	return nil
+}
+
+func handleUserCommands() error {
+	if config.Keys.DisableAuthentication && (flagNewUser != "" || flagDelUser != "") {
+		return fmt.Errorf("--add-user and --del-user can only be used if authentication is enabled")
+	}
 
-	var authentication *auth.Authentication
 	if !config.Keys.DisableAuthentication {
-		var err error
-		if authentication, err = auth.Init(db.DB, map[string]interface{}{
-			"ldap": config.Keys.LdapConfig,
-			"jwt":  config.Keys.JwtConfig,
-		}); err != nil {
-			log.Fatalf("auth initialization failed: %v", err)
+		if cfg := ccconf.GetPackageConfig("auth"); cfg != nil {
+			auth.Init(&cfg)
+		} else {
+			cclog.Warn("Authentication disabled due to missing configuration")
+			auth.Init(nil)
 		}
 
-		if d, err := time.ParseDuration(config.Keys.SessionMaxAge); err != nil {
-			authentication.SessionMaxAge = d
-		}
+		// Check for default security keys
+		checkDefaultSecurityKeys()
 
 		if flagNewUser != "" {
-			parts := strings.SplitN(flagNewUser, ":", 3)
-			if len(parts) != 3 || len(parts[0]) == 0 {
-				log.Fatal("invalid argument format for user creation")
+			if err := addUser(flagNewUser); err != nil {
+				return err
 			}
+		}
 
-			if err := authentication.AddUser(&auth.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)
-			}
-		}
 		if flagDelUser != "" {
-			if err := authentication.DelUser(flagDelUser); err != nil {
-				log.Fatalf("deleting user failed: %v", err)
+			if err := delUser(flagDelUser); err != nil {
+				return err
 			}
 		}
 
+		authHandle := auth.GetAuthInstance()
+
 		if flagSyncLDAP {
-			if authentication.LdapAuth == nil {
-				log.Fatal("cannot sync: LDAP authentication is not configured")
+			if err := syncLDAP(authHandle); err != nil {
+				return err
 			}
-
-			if err := authentication.LdapAuth.Sync(); err != nil {
-				log.Fatalf("LDAP sync failed: %v", err)
-			}
-			log.Info("LDAP sync successfull")
 		}
 
 		if flagGenJWT != "" {
-			user, err := authentication.GetUser(flagGenJWT)
-			if err != nil {
-				log.Fatalf("could not get user from JWT: %v", err)
+			if err := generateJWT(authHandle, flagGenJWT); err != nil {
+				return err
 			}
-
-			if !user.HasRole(auth.RoleApi) {
-				log.Warnf("user '%s' does not have the API role", user.Username)
-			}
-
-			jwt, err := authentication.JwtAuth.ProvideJWT(user)
-			if err != nil {
-				log.Fatalf("failed to provide JWT to user '%s': %v", user.Username, err)
-			}
-
-			fmt.Printf("MAIN > JWT for '%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")
 	}
 
-	if err := archive.Init(config.Keys.Archive, config.Keys.DisableArchive); err != nil {
-		log.Fatalf("failed to initialize archive: %s", err.Error())
+	return nil
+}
+
+// checkDefaultSecurityKeys warns if default JWT keys are detected
+func checkDefaultSecurityKeys() {
+	// Default JWT public key from init.go
+	defaultJWTPublic := "kzfYrYy+TzpanWZHJ5qSdMj5uKUWgq74BWhQG6copP0="
+
+	if os.Getenv("JWT_PUBLIC_KEY") == defaultJWTPublic {
+		cclog.Warn("Using default JWT keys - not recommended for production environments")
+	}
+}
+
+func addUser(userSpec string) error {
+	parts := strings.SplitN(userSpec, ":", 3)
+	if len(parts) != 3 || len(parts[0]) == 0 {
+		return fmt.Errorf("invalid user format, want: <username>:[admin,support,manager,api,user]:<password>, have: %s", userSpec)
 	}
 
-	if err := metricdata.Init(config.Keys.DisableArchive); err != nil {
-		log.Fatalf("failed to initialize metricdata repository: %s", err.Error())
+	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 {
+		return fmt.Errorf("adding user '%s' with roles '%s': %w", parts[0], parts[1], err)
 	}
 
+	cclog.Infof("Add User: Added new user '%s' with roles '%s'", parts[0], parts[1])
+	return nil
+}
+
+func delUser(username string) error {
+	ur := repository.GetUserRepository()
+	if err := ur.DelUser(username); err != nil {
+		return fmt.Errorf("deleting user '%s': %w", username, err)
+	}
+	cclog.Infof("Delete User: Deleted user '%s' from DB", username)
+	return nil
+}
+
+func syncLDAP(authHandle *auth.Authentication) error {
+	if authHandle.LdapAuth == nil {
+		return fmt.Errorf("LDAP authentication is not configured")
+	}
+
+	if err := authHandle.LdapAuth.Sync(); err != nil {
+		return fmt.Errorf("synchronizing LDAP: %w", err)
+	}
+
+	cclog.Print("Sync LDAP: LDAP synchronization successfull.")
+	return nil
+}
+
+func generateJWT(authHandle *auth.Authentication, username string) error {
+	ur := repository.GetUserRepository()
+	user, err := ur.GetUser(username)
+	if err != nil {
+		return fmt.Errorf("getting user '%s': %w", username, err)
+	}
+
+	if !user.HasRole(schema.RoleApi) {
+		cclog.Warnf("JWT: User '%s' does not have the role 'api'. REST API endpoints will return error!\n", user.Username)
+	}
+
+	jwt, err := authHandle.JwtAuth.ProvideJWT(user)
+	if err != nil {
+		return fmt.Errorf("generating JWT for user '%s': %w", user.Username, err)
+	}
+
+	cclog.Printf("JWT: Successfully generated JWT for user '%s': %s\n", user.Username, jwt)
+	return nil
+}
+
+func initSubsystems() error {
+	// Initialize nats client
+	natsConfig := ccconf.GetPackageConfig("nats")
+	if err := nats.Init(natsConfig); err != nil {
+		cclog.Warnf("initializing (optional) nats client: %s", err.Error())
+	}
+	nats.Connect()
+
+	// Initialize job archive
+	archiveCfg := ccconf.GetPackageConfig("archive")
+	if archiveCfg == nil {
+		archiveCfg = json.RawMessage(defaultArchiveConfig)
+	}
+	if err := archive.Init(archiveCfg, config.Keys.DisableArchive); err != nil {
+		return fmt.Errorf("initializing archive: %w", err)
+	}
+
+	// Initialize metricdata
+	if err := metricdata.Init(); err != nil {
+		return fmt.Errorf("initializing metricdata repository: %w", err)
+	}
+
+	// Handle database re-initialization
 	if flagReinitDB {
 		if err := importer.InitDB(); err != nil {
-			log.Fatalf("failed to re-initialize repository DB: %s", err.Error())
+			return fmt.Errorf("re-initializing repository DB: %w", err)
 		}
+		cclog.Print("Init DB: Successfully re-initialized repository DB.")
 	}
 
+	// Handle job import
 	if flagImportJob != "" {
 		if err := importer.HandleImportFlag(flagImportJob); err != nil {
-			log.Fatalf("job import failed: %s", err.Error())
+			return fmt.Errorf("importing job: %w", err)
+		}
+		cclog.Infof("Import Job: Imported Job '%s' into DB", flagImportJob)
+	}
+
+	// Initialize taggers
+	if config.Keys.EnableJobTaggers {
+		tagger.Init()
+	}
+
+	// Apply tags if requested
+	if flagApplyTags {
+		if err := tagger.RunTaggers(); err != nil {
+			return fmt.Errorf("running job taggers: %w", err)
 		}
 	}
 
-	if !flagServer {
-		return
-	}
-
-	// Setup the http.Handler/Router used by the server
-	jobRepo := repository.GetJobRepository()
-	resolver := &graph.Resolver{DB: db.DB, Repo: jobRepo}
-	graphQLEndpoint := handler.NewDefaultServer(generated.NewExecutableSchema(generated.Config{Resolvers: resolver}))
-	if os.Getenv("DEBUG") != "1" {
-		// Having this handler means that a error message is returned via GraphQL instead of the connection simply beeing closed.
-		// The problem with this is that then, no more stacktrace is printed to stderr.
-		graphQLEndpoint.SetRecoverFunc(func(ctx context.Context, err interface{}) error {
-			switch e := err.(type) {
-			case string:
-				return fmt.Errorf("MAIN > Panic: %s", e)
-			case error:
-				return fmt.Errorf("MAIN > Panic caused by: %w", e)
-			}
-
-			return errors.New("MAIN > Internal server error (panic)")
-		})
-	}
-
-	api := &api.RestApi{
-		JobRepository:   jobRepo,
-		Resolver:        resolver,
-		MachineStateDir: config.Keys.MachineStateDir,
-		Authentication:  authentication,
-	}
-
-	r := mux.NewRouter()
-	buildInfo := web.Build{Version: version, Hash: commit, Buildtime: date}
-
-	r.HandleFunc("/login", func(rw http.ResponseWriter, r *http.Request) {
-		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
-		web.RenderTemplate(rw, r, "login.tmpl", &web.Page{Title: "Login", Build: buildInfo})
-	}).Methods(http.MethodGet)
-	r.HandleFunc("/imprint", func(rw http.ResponseWriter, r *http.Request) {
-		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
-		web.RenderTemplate(rw, r, "imprint.tmpl", &web.Page{Title: "Imprint", Build: buildInfo})
-	})
-	r.HandleFunc("/privacy", func(rw http.ResponseWriter, r *http.Request) {
-		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
-		web.RenderTemplate(rw, r, "privacy.tmpl", &web.Page{Title: "Privacy", Build: buildInfo})
-	})
-
-	// Some routes, such as /login or /query, should only be accessible to a user that is logged in.
-	// Those should be mounted to this subrouter. If authentication is enabled, a middleware will prevent
-	// any unauthenticated accesses.
-	secured := r.PathPrefix("/").Subrouter()
-	if !config.Keys.DisableAuthentication {
-		r.Handle("/login", authentication.Login(
-			// On success:
-			http.RedirectHandler("/", http.StatusTemporaryRedirect),
-
-			// On failure:
-			func(rw http.ResponseWriter, r *http.Request, err error) {
-				rw.Header().Add("Content-Type", "text/html; charset=utf-8")
-				rw.WriteHeader(http.StatusUnauthorized)
-				web.RenderTemplate(rw, r, "login.tmpl", &web.Page{
-					Title:   "Login failed - ClusterCockpit",
-					MsgType: "alert-warning",
-					Message: err.Error(),
-					Build:   buildInfo,
-				})
-			})).Methods(http.MethodPost)
-
-		r.Handle("/logout", authentication.Logout(http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
-			rw.Header().Add("Content-Type", "text/html; charset=utf-8")
-			rw.WriteHeader(http.StatusOK)
-			web.RenderTemplate(rw, r, "login.tmpl", &web.Page{
-				Title:   "Bye - ClusterCockpit",
-				MsgType: "alert-info",
-				Message: "Logout successful",
-				Build:   buildInfo,
-			})
-		}))).Methods(http.MethodPost)
-
-		secured.Use(func(next http.Handler) http.Handler {
-			return authentication.Auth(
-				// On success;
-				next,
-
-				// On failure:
-				func(rw http.ResponseWriter, r *http.Request, err error) {
-					rw.WriteHeader(http.StatusUnauthorized)
-					web.RenderTemplate(rw, r, "login.tmpl", &web.Page{
-						Title:   "Authentication failed - ClusterCockpit",
-						MsgType: "alert-danger",
-						Message: err.Error(),
-						Build:   buildInfo,
-					})
-				})
-		})
-	}
-
-	if flagDev {
-		r.Handle("/playground", playground.Handler("GraphQL playground", "/query"))
-		r.PathPrefix("/swagger/").Handler(httpSwagger.Handler(
-			httpSwagger.URL("http://" + config.Keys.Addr + "/swagger/doc.json"))).Methods(http.MethodGet)
-	}
-	secured.Handle("/query", graphQLEndpoint)
-
-	// Send a searchId and then reply with a redirect to a user, or directly send query to job table for jobid and project.
-	secured.HandleFunc("/search", func(rw http.ResponseWriter, r *http.Request) {
-		routerConfig.HandleSearchBar(rw, r, buildInfo)
-	})
-
-	// Mount all /monitoring/... and /api/... routes.
-	routerConfig.SetupRoutes(secured, buildInfo)
-	api.MountRoutes(secured)
-
-	if config.Keys.EmbedStaticFiles {
-		if i, err := os.Stat("./var/img"); err == nil {
-			if i.IsDir() {
-				log.Info("Use local directory for static images")
-				r.PathPrefix("/img/").Handler(http.StripPrefix("/img/", http.FileServer(http.Dir("./var/img"))))
-			}
-		}
-		r.PathPrefix("/").Handler(web.ServeFiles())
-	} else {
-		r.PathPrefix("/").Handler(http.FileServer(http.Dir(config.Keys.StaticFiles)))
-	}
-
-	r.Use(handlers.CompressHandler)
-	r.Use(handlers.RecoveryHandler(handlers.PrintRecoveryStack(true)))
-	r.Use(handlers.CORS(
-		handlers.AllowCredentials(),
-		handlers.AllowedHeaders([]string{"X-Requested-With", "Content-Type", "Authorization", "Origin"}),
-		handlers.AllowedMethods([]string{"GET", "POST", "HEAD", "OPTIONS"}),
-		handlers.AllowedOrigins([]string{"*"})))
-	handler := handlers.CustomLoggingHandler(io.Discard, r, func(_ io.Writer, params handlers.LogFormatterParams) {
-		if strings.HasPrefix(params.Request.RequestURI, "/api/") {
-			log.Debugf("%s %s (%d, %.02fkb, %dms)",
-				params.Request.Method, params.URL.RequestURI(),
-				params.StatusCode, float32(params.Size)/1024,
-				time.Since(params.TimeStamp).Milliseconds())
-		} else {
-			log.Debugf("%s %s (%d, %.02fkb, %dms)",
-				params.Request.Method, params.URL.RequestURI(),
-				params.StatusCode, float32(params.Size)/1024,
-				time.Since(params.TimeStamp).Milliseconds())
-		}
-	})
+	return nil
+}
 
+func runServer(ctx context.Context) error {
 	var wg sync.WaitGroup
-	server := http.Server{
-		ReadTimeout:  10 * time.Second,
-		WriteTimeout: 10 * time.Second,
-		Handler:      handler,
-		Addr:         config.Keys.Addr,
-	}
 
-	// Start http or https server
-	listener, err := net.Listen("tcp", config.Keys.Addr)
-	if err != nil {
-		log.Fatalf("starting http listener failed: %v", err)
-	}
-
-	if !strings.HasSuffix(config.Keys.Addr, ":80") && config.Keys.RedirectHttpTo != "" {
-		go func() {
-			http.ListenAndServe(":80", http.RedirectHandler(config.Keys.RedirectHttpTo, http.StatusMovedPermanently))
-		}()
-	}
-
-	if config.Keys.HttpsCertFile != "" && config.Keys.HttpsKeyFile != "" {
-		cert, err := tls.LoadX509KeyPair(config.Keys.HttpsCertFile, config.Keys.HttpsKeyFile)
-		if err != nil {
-			log.Fatalf("loading X509 keypair failed: %v", err)
+	// Start metric store if enabled
+	if memorystore.InternalCCMSFlag {
+		mscfg := ccconf.GetPackageConfig("metric-store")
+		if mscfg == nil {
+			return fmt.Errorf("metric store configuration must be present")
 		}
-		listener = tls.NewListener(listener, &tls.Config{
-			Certificates: []tls.Certificate{cert},
-			CipherSuites: []uint16{
-				tls.TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256,
-				tls.TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256,
-			},
-			MinVersion:               tls.VersionTLS12,
-			PreferServerCipherSuites: true,
-		})
-		fmt.Printf("HTTPS server listening at %s...", config.Keys.Addr)
-	} else {
-		fmt.Printf("HTTP server listening at %s...", config.Keys.Addr)
+		memorystore.Init(mscfg, &wg)
 	}
 
-	// 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())
+	// Start archiver and task manager
+	archiver.Start(repository.GetJobRepository(), ctx)
+	taskmanager.Start(ccconf.GetPackageConfig("cron"), ccconf.GetPackageConfig("archive"))
+
+	// Initialize web UI
+	cfg := ccconf.GetPackageConfig("ui")
+	if err := web.Init(cfg); err != nil {
+		return fmt.Errorf("initializing web UI: %w", err)
 	}
 
+	// Initialize HTTP server
+	srv, err := NewServer(version, commit, date)
+	if err != nil {
+		return fmt.Errorf("creating server: %w", err)
+	}
+
+	// Channel to collect errors from server
+	errChan := make(chan error, 1)
+
+	// Start HTTP server
 	wg.Add(1)
 	go func() {
 		defer wg.Done()
-		if err = server.Serve(listener); err != nil && err != http.ErrServerClosed {
-			log.Fatalf("starting server failed: %v", err)
+		if err := srv.Start(ctx); err != nil {
+			errChan <- err
 		}
 	}()
 
+	// Handle shutdown signals
 	wg.Add(1)
 	sigs := make(chan os.Signal, 1)
 	signal.Notify(sigs, syscall.SIGINT, syscall.SIGTERM)
 	go func() {
 		defer wg.Done()
-		<-sigs
+		select {
+		case <-sigs:
+			cclog.Info("Shutdown signal received")
+		case <-ctx.Done():
+		}
+
 		runtimeEnv.SystemdNotifiy(false, "Shutting down ...")
-
-		// First shut down the server gracefully (waiting for all ongoing requests)
-		server.Shutdown(context.Background())
-
-		// Then, wait for any async archivings still pending...
-		api.JobRepository.WaitForArchiving()
+		srv.Shutdown(ctx)
+		util.FsWatcherShutdown()
+		taskmanager.Shutdown()
 	}()
 
-	s := gocron.NewScheduler(time.Local)
-
-	if config.Keys.StopJobsExceedingWalltime > 0 {
-		log.Info("Register undead jobs service")
-
-		s.Every(1).Day().At("3:00").Do(func() {
-			err = jobRepo.StopJobsExceedingWalltimeBy(config.Keys.StopJobsExceedingWalltime)
-			if err != nil {
-				log.Warnf("Error while looking for jobs exceeding their walltime: %s", err.Error())
-			}
-			runtime.GC()
-		})
-	}
-
-	var cfg struct {
-		Compression int              `json:"compression"`
-		Retention   schema.Retention `json:"retention"`
-	}
-
-	cfg.Retention.IncludeDB = true
-
-	if err = json.Unmarshal(config.Keys.Archive, &cfg); err != nil {
-		log.Warn("Error while unmarshaling raw config json")
-	}
-
-	switch cfg.Retention.Policy {
-	case "delete":
-		log.Info("Register retention delete service")
-
-		s.Every(1).Day().At("4:00").Do(func() {
-			startTime := time.Now().Unix() - int64(cfg.Retention.Age*24*3600)
-			jobs, err := jobRepo.FindJobsBetween(0, startTime)
-			if err != nil {
-				log.Warnf("Error while looking for retention jobs: %s", err.Error())
-			}
-			archive.GetHandle().CleanUp(jobs)
-
-			if cfg.Retention.IncludeDB {
-				cnt, err := jobRepo.DeleteJobsBefore(startTime)
-				if err != nil {
-					log.Errorf("Error while deleting retention jobs from db: %s", err.Error())
-				} else {
-					log.Infof("Retention: Removed %d jobs from db", cnt)
-				}
-				if err = jobRepo.Optimize(); err != nil {
-					log.Errorf("Error occured in db optimization: %s", err.Error())
-				}
-			}
-		})
-	case "move":
-		log.Info("Register retention move service")
-
-		s.Every(1).Day().At("4:00").Do(func() {
-			startTime := time.Now().Unix() - int64(cfg.Retention.Age*24*3600)
-			jobs, err := jobRepo.FindJobsBetween(0, startTime)
-			if err != nil {
-				log.Warnf("Error while looking for retention jobs: %s", err.Error())
-			}
-			archive.GetHandle().Move(jobs, cfg.Retention.Location)
-
-			if cfg.Retention.IncludeDB {
-				cnt, err := jobRepo.DeleteJobsBefore(startTime)
-				if err != nil {
-					log.Errorf("Error while deleting retention jobs from db: %v", err)
-				} else {
-					log.Infof("Retention: Removed %d jobs from db", cnt)
-				}
-				if err = jobRepo.Optimize(); err != nil {
-					log.Errorf("Error occured in db optimization: %v", err)
-				}
-			}
-		})
-	}
-
-	if cfg.Compression > 0 {
-		log.Info("Register compression service")
-
-		s.Every(1).Day().At("5:00").Do(func() {
-			var jobs []*schema.Job
-
-			ar := archive.GetHandle()
-			startTime := time.Now().Unix() - int64(cfg.Compression*24*3600)
-			lastTime := ar.CompressLast(startTime)
-			if startTime == lastTime {
-				log.Info("Compression Service - Complete archive run")
-				jobs, err = jobRepo.FindJobsBetween(0, startTime)
-
-			} else {
-				jobs, err = jobRepo.FindJobsBetween(lastTime, startTime)
-			}
-
-			if err != nil {
-				log.Warnf("Error while looking for compression jobs: %v", err)
-			}
-			ar.Compress(jobs)
-		})
-	}
-
-	s.StartAsync()
-
-	if os.Getenv("GOGC") == "" {
+	// Set GC percent if not configured
+	if os.Getenv(envGOGC) == "" {
 		debug.SetGCPercent(25)
 	}
 	runtimeEnv.SystemdNotifiy(true, "running")
-	wg.Wait()
-	log.Print("Gracefull shutdown completed!")
+
+	// Wait for completion or error
+	go func() {
+		wg.Wait()
+		close(errChan)
+	}()
+
+	// Check for server startup errors
+	select {
+	case err := <-errChan:
+		if err != nil {
+			return err
+		}
+	case <-time.After(100 * time.Millisecond):
+		// Server started successfully, wait for completion
+		if err := <-errChan; err != nil {
+			return err
+		}
+	}
+
+	cclog.Print("Graceful shutdown completed!")
+	return nil
+}
+
+func run() error {
+	cliInit()
+
+	if flagVersion {
+		printVersion()
+		return nil
+	}
+
+	// Initialize logger
+	cclog.Init(flagLogLevel, flagLogDateTime)
+
+	// Handle init flag
+	if flagInit {
+		initEnv()
+		cclog.Exit("Successfully setup environment!\n" +
+			"Please review config.json and .env and adjust it to your needs.\n" +
+			"Add your job-archive at ./var/job-archive.")
+	}
+
+	// Initialize gops agent
+	if err := initGops(); err != nil {
+		return err
+	}
+
+	// Initialize subsystems in dependency order:
+	// 1. Load environment variables from .env file (contains sensitive configuration)
+	// 2. Load configuration from config.json (may reference environment variables)
+	// 3. Handle database migration commands if requested
+	// 4. Initialize database connection (requires config for connection string)
+	// 5. Handle user commands if requested (requires database and authentication config)
+	// 6. Initialize subsystems like archive and metrics (require config and database)
+
+	// Load environment and configuration
+	if err := loadEnvironment(); err != nil {
+		return err
+	}
+
+	if err := initConfiguration(); err != nil {
+		return err
+	}
+
+	// Handle database migration (migrate, revert, force)
+	if err := handleDatabaseCommands(); err != nil {
+		return err
+	}
+
+	// Initialize database
+	if err := initDatabase(); err != nil {
+		return err
+	}
+
+	// Handle user commands (add, delete, sync, JWT)
+	if err := handleUserCommands(); err != nil {
+		return err
+	}
+
+	// Initialize subsystems (archive, metrics, taggers)
+	if err := initSubsystems(); err != nil {
+		return err
+	}
+
+	// Exit if start server is not requested
+	if !flagServer {
+		cclog.Exit("No errors, server flag not set. Exiting cc-backend.")
+	}
+
+	// Run server with context
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	return runServer(ctx)
+}
+
+func main() {
+	if err := run(); err != nil {
+		cclog.Error(err.Error())
+		os.Exit(1)
+	}
 }

cmd/cc-backend/server.go

@@ -0,0 +1,386 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+// Package main provides the entry point for the ClusterCockpit backend server.
+// This file contains HTTP server setup, routing configuration, and
+// authentication middleware integration.
+package main
+
+import (
+	"context"
+	"crypto/tls"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net"
+	"net/http"
+	"os"
+	"strings"
+	"time"
+
+	"github.com/99designs/gqlgen/graphql/handler"
+	"github.com/99designs/gqlgen/graphql/handler/transport"
+	"github.com/99designs/gqlgen/graphql/playground"
+	"github.com/ClusterCockpit/cc-backend/internal/api"
+	"github.com/ClusterCockpit/cc-backend/internal/archiver"
+	"github.com/ClusterCockpit/cc-backend/internal/auth"
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-backend/internal/graph"
+	"github.com/ClusterCockpit/cc-backend/internal/graph/generated"
+	"github.com/ClusterCockpit/cc-backend/internal/memorystore"
+	"github.com/ClusterCockpit/cc-backend/internal/routerConfig"
+	"github.com/ClusterCockpit/cc-backend/pkg/nats"
+	"github.com/ClusterCockpit/cc-backend/web"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/runtimeEnv"
+	"github.com/gorilla/handlers"
+	"github.com/gorilla/mux"
+	httpSwagger "github.com/swaggo/http-swagger"
+)
+
+var buildInfo web.Build
+
+// Environment variable names
+const (
+	envDebug = "DEBUG"
+)
+
+// Server encapsulates the HTTP server state and dependencies
+type Server struct {
+	router    *mux.Router
+	server    *http.Server
+	apiHandle *api.RestAPI
+}
+
+func onFailureResponse(rw http.ResponseWriter, r *http.Request, err error) {
+	rw.Header().Add("Content-Type", "application/json")
+	rw.WriteHeader(http.StatusUnauthorized)
+	json.NewEncoder(rw).Encode(map[string]string{
+		"status": http.StatusText(http.StatusUnauthorized),
+		"error":  err.Error(),
+	})
+}
+
+// NewServer creates and initializes a new Server instance
+func NewServer(version, commit, buildDate string) (*Server, error) {
+	buildInfo = web.Build{Version: version, Hash: commit, Buildtime: buildDate}
+
+	s := &Server{
+		router: mux.NewRouter(),
+	}
+
+	if err := s.init(); err != nil {
+		return nil, err
+	}
+
+	return s, nil
+}
+
+func (s *Server) init() error {
+	// Setup the http.Handler/Router used by the server
+	graph.Init()
+	resolver := graph.GetResolverInstance()
+	graphQLServer := handler.New(
+		generated.NewExecutableSchema(generated.Config{Resolvers: resolver}))
+
+	graphQLServer.AddTransport(transport.POST{})
+
+	if os.Getenv(envDebug) != "1" {
+		// Having this handler means that a error message is returned via GraphQL instead of the connection simply beeing closed.
+		// The problem with this is that then, no more stacktrace is printed to stderr.
+		graphQLServer.SetRecoverFunc(func(ctx context.Context, err any) error {
+			switch e := err.(type) {
+			case string:
+				return fmt.Errorf("MAIN > Panic: %s", e)
+			case error:
+				return fmt.Errorf("MAIN > Panic caused by: %s", e.Error())
+			}
+
+			return errors.New("MAIN > Internal server error (panic)")
+		})
+	}
+
+	authHandle := auth.GetAuthInstance()
+
+	s.apiHandle = api.New()
+
+	info := map[string]any{}
+	info["hasOpenIDConnect"] = false
+
+	if auth.Keys.OpenIDConfig != nil {
+		openIDConnect := auth.NewOIDC(authHandle)
+		openIDConnect.RegisterEndpoints(s.router)
+		info["hasOpenIDConnect"] = true
+	}
+
+	s.router.HandleFunc("/login", func(rw http.ResponseWriter, r *http.Request) {
+		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
+		cclog.Debugf("##%v##", info)
+		web.RenderTemplate(rw, "login.tmpl", &web.Page{Title: "Login", Build: buildInfo, Infos: info})
+	}).Methods(http.MethodGet)
+	s.router.HandleFunc("/imprint", func(rw http.ResponseWriter, r *http.Request) {
+		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
+		web.RenderTemplate(rw, "imprint.tmpl", &web.Page{Title: "Imprint", Build: buildInfo})
+	})
+	s.router.HandleFunc("/privacy", func(rw http.ResponseWriter, r *http.Request) {
+		rw.Header().Add("Content-Type", "text/html; charset=utf-8")
+		web.RenderTemplate(rw, "privacy.tmpl", &web.Page{Title: "Privacy", Build: buildInfo})
+	})
+
+	secured := s.router.PathPrefix("/").Subrouter()
+	securedapi := s.router.PathPrefix("/api").Subrouter()
+	userapi := s.router.PathPrefix("/userapi").Subrouter()
+	configapi := s.router.PathPrefix("/config").Subrouter()
+	frontendapi := s.router.PathPrefix("/frontend").Subrouter()
+	metricstoreapi := s.router.PathPrefix("/metricstore").Subrouter()
+
+	if !config.Keys.DisableAuthentication {
+		// Create login failure handler (used by both /login and /jwt-login)
+		loginFailureHandler := func(rw http.ResponseWriter, r *http.Request, err error) {
+			rw.Header().Add("Content-Type", "text/html; charset=utf-8")
+			rw.WriteHeader(http.StatusUnauthorized)
+			web.RenderTemplate(rw, "login.tmpl", &web.Page{
+				Title:   "Login failed - ClusterCockpit",
+				MsgType: "alert-warning",
+				Message: err.Error(),
+				Build:   buildInfo,
+				Infos:   info,
+			})
+		}
+
+		s.router.Handle("/login", authHandle.Login(loginFailureHandler)).Methods(http.MethodPost)
+		s.router.Handle("/jwt-login", authHandle.Login(loginFailureHandler))
+
+		s.router.Handle("/logout", authHandle.Logout(
+			http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+				rw.Header().Add("Content-Type", "text/html; charset=utf-8")
+				rw.WriteHeader(http.StatusOK)
+				web.RenderTemplate(rw, "login.tmpl", &web.Page{
+					Title:   "Bye - ClusterCockpit",
+					MsgType: "alert-info",
+					Message: "Logout successful",
+					Build:   buildInfo,
+					Infos:   info,
+				})
+			}))).Methods(http.MethodPost)
+
+		secured.Use(func(next http.Handler) http.Handler {
+			return authHandle.Auth(
+				// On success;
+				next,
+
+				// On failure:
+				func(rw http.ResponseWriter, r *http.Request, err error) {
+					rw.WriteHeader(http.StatusUnauthorized)
+					web.RenderTemplate(rw, "login.tmpl", &web.Page{
+						Title:    "Authentication failed - ClusterCockpit",
+						MsgType:  "alert-danger",
+						Message:  err.Error(),
+						Build:    buildInfo,
+						Infos:    info,
+						Redirect: r.RequestURI,
+					})
+				})
+		})
+
+		securedapi.Use(func(next http.Handler) http.Handler {
+			return authHandle.AuthAPI(
+				// On success;
+				next,
+				// On failure: JSON Response
+				onFailureResponse)
+		})
+
+		userapi.Use(func(next http.Handler) http.Handler {
+			return authHandle.AuthUserAPI(
+				// On success;
+				next,
+				// On failure: JSON Response
+				onFailureResponse)
+		})
+
+		metricstoreapi.Use(func(next http.Handler) http.Handler {
+			return authHandle.AuthMetricStoreAPI(
+				// On success;
+				next,
+				// On failure: JSON Response
+				onFailureResponse)
+		})
+
+		configapi.Use(func(next http.Handler) http.Handler {
+			return authHandle.AuthConfigAPI(
+				// On success;
+				next,
+				// On failure: JSON Response
+				onFailureResponse)
+		})
+
+		frontendapi.Use(func(next http.Handler) http.Handler {
+			return authHandle.AuthFrontendAPI(
+				// On success;
+				next,
+				// On failure: JSON Response
+				onFailureResponse)
+		})
+	}
+
+	if flagDev {
+		s.router.Handle("/playground", playground.Handler("GraphQL playground", "/query"))
+		s.router.PathPrefix("/swagger/").Handler(httpSwagger.Handler(
+			httpSwagger.URL("http://" + config.Keys.Addr + "/swagger/doc.json"))).Methods(http.MethodGet)
+	}
+	secured.Handle("/query", graphQLServer)
+
+	// Send a searchId and then reply with a redirect to a user, or directly send query to job table for jobid and project.
+	secured.HandleFunc("/search", func(rw http.ResponseWriter, r *http.Request) {
+		routerConfig.HandleSearchBar(rw, r, buildInfo)
+	})
+
+	// Mount all /monitoring/... and /api/... routes.
+	routerConfig.SetupRoutes(secured, buildInfo)
+	s.apiHandle.MountAPIRoutes(securedapi)
+	s.apiHandle.MountUserAPIRoutes(userapi)
+	s.apiHandle.MountConfigAPIRoutes(configapi)
+	s.apiHandle.MountFrontendAPIRoutes(frontendapi)
+
+	if memorystore.InternalCCMSFlag {
+		s.apiHandle.MountMetricStoreAPIRoutes(metricstoreapi)
+	}
+
+	if config.Keys.EmbedStaticFiles {
+		if i, err := os.Stat("./var/img"); err == nil {
+			if i.IsDir() {
+				cclog.Info("Use local directory for static images")
+				s.router.PathPrefix("/img/").Handler(http.StripPrefix("/img/", http.FileServer(http.Dir("./var/img"))))
+			}
+		}
+		s.router.PathPrefix("/").Handler(http.StripPrefix("/", web.ServeFiles()))
+	} else {
+		s.router.PathPrefix("/").Handler(http.FileServer(http.Dir(config.Keys.StaticFiles)))
+	}
+
+	s.router.Use(handlers.CompressHandler)
+	s.router.Use(handlers.RecoveryHandler(handlers.PrintRecoveryStack(true)))
+	s.router.Use(handlers.CORS(
+		handlers.AllowCredentials(),
+		handlers.AllowedHeaders([]string{"X-Requested-With", "Content-Type", "Authorization", "Origin"}),
+		handlers.AllowedMethods([]string{"GET", "POST", "HEAD", "OPTIONS"}),
+		handlers.AllowedOrigins([]string{"*"})))
+
+	return nil
+}
+
+// Server timeout defaults (in seconds)
+const (
+	defaultReadTimeout  = 20
+	defaultWriteTimeout = 20
+)
+
+func (s *Server) Start(ctx context.Context) error {
+	handler := handlers.CustomLoggingHandler(io.Discard, s.router, func(_ io.Writer, params handlers.LogFormatterParams) {
+		if strings.HasPrefix(params.Request.RequestURI, "/api/") {
+			cclog.Debugf("%s %s (%d, %.02fkb, %dms)",
+				params.Request.Method, params.URL.RequestURI(),
+				params.StatusCode, float32(params.Size)/1024,
+				time.Since(params.TimeStamp).Milliseconds())
+		} else {
+			cclog.Debugf("%s %s (%d, %.02fkb, %dms)",
+				params.Request.Method, params.URL.RequestURI(),
+				params.StatusCode, float32(params.Size)/1024,
+				time.Since(params.TimeStamp).Milliseconds())
+		}
+	})
+
+	// Use configurable timeouts with defaults
+	readTimeout := time.Duration(defaultReadTimeout) * time.Second
+	writeTimeout := time.Duration(defaultWriteTimeout) * time.Second
+
+	s.server = &http.Server{
+		ReadTimeout:  readTimeout,
+		WriteTimeout: writeTimeout,
+		Handler:      handler,
+		Addr:         config.Keys.Addr,
+	}
+
+	// Start http or https server
+	listener, err := net.Listen("tcp", config.Keys.Addr)
+	if err != nil {
+		return fmt.Errorf("starting listener on '%s': %w", config.Keys.Addr, err)
+	}
+
+	if !strings.HasSuffix(config.Keys.Addr, ":80") && config.Keys.RedirectHTTPTo != "" {
+		go func() {
+			http.ListenAndServe(":80", http.RedirectHandler(config.Keys.RedirectHTTPTo, http.StatusMovedPermanently))
+		}()
+	}
+
+	if config.Keys.HTTPSCertFile != "" && config.Keys.HTTPSKeyFile != "" {
+		cert, err := tls.LoadX509KeyPair(
+			config.Keys.HTTPSCertFile, config.Keys.HTTPSKeyFile)
+		if err != nil {
+			return fmt.Errorf("loading X509 keypair (check 'https-cert-file' and 'https-key-file' in config.json): %w", err)
+		}
+		listener = tls.NewListener(listener, &tls.Config{
+			Certificates: []tls.Certificate{cert},
+			CipherSuites: []uint16{
+				tls.TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256,
+				tls.TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256,
+			},
+			MinVersion:               tls.VersionTLS12,
+			PreferServerCipherSuites: true,
+		})
+		cclog.Infof("HTTPS server listening at %s...", config.Keys.Addr)
+	} else {
+		cclog.Infof("HTTP server listening at %s...", config.Keys.Addr)
+	}
+	//
+	// Because this program will want to bind to a privileged port (like 80), the listener must
+	// be established first, then the user can be changed, and after that,
+	// the actual http server can be started.
+	if err := runtimeEnv.DropPrivileges(config.Keys.Group, config.Keys.User); err != nil {
+		return fmt.Errorf("dropping privileges: %w", err)
+	}
+
+	// Handle context cancellation for graceful shutdown
+	go func() {
+		<-ctx.Done()
+		shutdownCtx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+		defer cancel()
+		if err := s.server.Shutdown(shutdownCtx); err != nil {
+			cclog.Errorf("Server shutdown error: %v", err)
+		}
+	}()
+
+	if err = s.server.Serve(listener); err != nil && err != http.ErrServerClosed {
+		return fmt.Errorf("server failed: %w", err)
+	}
+	return nil
+}
+
+func (s *Server) Shutdown(ctx context.Context) {
+	// Create a shutdown context with timeout
+	shutdownCtx, cancel := context.WithTimeout(ctx, 30*time.Second)
+	defer cancel()
+
+	nc := nats.GetClient()
+	if nc != nil {
+		nc.Close()
+	}
+
+	// First shut down the server gracefully (waiting for all ongoing requests)
+	if err := s.server.Shutdown(shutdownCtx); err != nil {
+		cclog.Errorf("Server shutdown error: %v", err)
+	}
+
+	// Archive all the metric store data
+	if memorystore.InternalCCMSFlag {
+		memorystore.Shutdown()
+	}
+
+	// Shutdown archiver with 10 second timeout for fast shutdown
+	if err := archiver.Shutdown(10 * time.Second); err != nil {
+		cclog.Warnf("Archiver shutdown: %v", err)
+	}
+}

configs/README.md

@@ -1,76 +0,0 @@
-## Intro
-
-cc-backend requires a configuration file that specifies the cluster systems to be used.
-To override the default, specify the location of a json configuration file with the `-config <file path>` command line option.
-All security-related configurations, e.g. keys and passwords, are set using
-environment variables.
-It is supported to set these by means of a `.env` file in the project root.
-
-## Configuration Options
-
-* `addr`: Type string.  Address where the http (or https) server will listen on (for example: 'localhost:80'). Default `:8080`.
-* `user`: Type string. Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.
-* `group`: Type string.  Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.
-* `disable-authentication`: Type bool.  Disable authentication (for everything: API, Web-UI, ...). Default `false`.
-* `embed-static-files`: Type bool. If all files in `web/frontend/public` should be served from within the binary itself (they are embedded) or not. Default `true`.
-* `static-files`: Type string. Folder where static assets can be found, if `embed-static-files` is `false`. No default.
-* `db-driver`: Type string. 'sqlite3' or 'mysql' (mysql will work for mariadb as well). Default `sqlite3`.
-* `db`: Type string. For sqlite3 a filename, for mysql a DSN in this format: https://github.com/go-sql-driver/mysql#dsn-data-source-name (Without query parameters!). Default: `./var/job.db`.
-* `job-archive`: Type string. Path to the job-archive. Default: `./var/job-archive`.
-* `disable-archive`: Type bool. Keep all metric data in the metric data repositories, do not write to the job-archive. Default `false`.
-* `validate`: Type bool. Validate all input json documents against json schema.
-* `session-max-age`: Type string. Specifies for how long a session shall be valid  as a string parsable by time.ParseDuration(). If 0 or empty, the session/token does not expire! Default `168h`.
-* `jwt-max-age`: Type string. Specifies for how long a JWT token shall be valid  as a string parsable by time.ParseDuration(). If 0 or empty, the session/token does not expire! Default `0`.
-* `https-cert-file` and `https-key-file`: Type string. If both those options are not empty, use HTTPS using those certificates.
-* `redirect-http-to`: Type string. If not the empty string and `addr` does not end in ":80", redirect every request incoming at port 80 to that url.
-* `machine-state-dir`: Type string. Where to store MachineState files. TODO: Explain in more detail!
-* `stop-jobs-exceeding-walltime`: Type int. If not zero, automatically mark jobs as stopped running X seconds longer than their walltime. Only applies if walltime is set for job. Default `0`.
-* `short-running-jobs-duration`: Type int. Do not show running jobs shorter than X seconds. Default `300`.
-* `ldap`: Type object. For LDAP Authentication and user synchronisation. Default `nil`.
-   - `url`: Type string.  URL of LDAP directory server.
-   - `user_base`: Type string. Base DN of user tree root.
-   - `search_dn`: Type string. DN for authenticating LDAP admin account with general read rights.
-   - `user_bind`: Type string. Expression used to authenticate users via LDAP bind. Must contain `uid={username}`.
-   - `user_filter`: Type string. Filter to extract users for syncing.
-   - `sync_interval`: Type string. Interval used for syncing local user table with LDAP directory. Parsed using time.ParseDuration.
-   - `sync_del_old_users`: Type bool. Delete obsolete users in database.
-* `clusters`: Type array of objects
-   - `name`: Type string. The name of the cluster.
-   - `metricDataRepository`: Type object with properties: `kind` (Type string, can be one of `cc-metric-store`, `influxdb` ), `url` (Type string), `token` (Type string)
-   - `filterRanges` Type object. This option controls the slider ranges for the UI controls of numNodes, duration, and startTime.  Example:
-   ```
-   "filterRanges": {
-                "numNodes": { "from": 1, "to": 64 },
-                "duration": { "from": 0, "to": 86400 },
-                "startTime": { "from": "2022-01-01T00:00:00Z", "to": null }
-            }
-   ```
-* `ui-defaults`: Type object. Default configuration for ui views. If overwritten, all options  must be provided! Most options can be overwritten by the user via the web interface.
-   - `analysis_view_histogramMetrics`: Type string array. Metrics to show as job count histograms in analysis view. Default `["flops_any", "mem_bw", "mem_used"]`.
-   - `analysis_view_scatterPlotMetrics`: Type array of string array. Initial
-   scatter plot configuration in analysis view. Default `[["flops_any", "mem_bw"], ["flops_any", "cpu_load"], ["cpu_load", "mem_bw"]]`.
-   - `job_view_nodestats_selectedMetrics`: Type string array. Initial metrics shown in node statistics table of single job view. Default `["flops_any", "mem_bw", "mem_used"]`.
-   - `job_view_polarPlotMetrics`: Type string array. Metrics shown in polar plot of single job view. Default `["flops_any", "mem_bw", "mem_used", "net_bw", "file_bw"]`.
-   - `job_view_selectedMetrics`: Type string array.  Default `["flops_any", "mem_bw", "mem_used"]`.
-   - `plot_general_colorBackground`: Type bool. Color plot background according to job average threshold limits. Default `true`.
-   - `plot_general_colorscheme`: Type string array. Initial color scheme. Default `"#00bfff", "#0000ff", "#ff00ff", "#ff0000", "#ff8000", "#ffff00", "#80ff00"`.
-   - `plot_general_lineWidth`: Type int. Initial linewidth. Default `3`.
-   - `plot_list_jobsPerPage`: Type int. Jobs shown per page in job lists. Default `50`.
-   - `plot_list_selectedMetrics`: Type string array. Initial metric plots shown in jobs lists. Default `"cpu_load", "ipc", "mem_used", "flops_any", "mem_bw"`.
-   - `plot_view_plotsPerRow`: Type int. Number of plots per row in single job view. Default `3`.
-   - `plot_view_showPolarplot`: Type bool. Option to toggle polar plot in single job view. Default `true`.
-   - `plot_view_showRoofline`: Type bool. Option to toggle roofline plot in single job view. Default `true`.
-   - `plot_view_showStatTable`: Type bool. Option to toggle the node statistic table in single job view. Default `true`.
-   - `system_view_selectedMetric`: Type string. Initial metric shown in system view. Default `cpu_load`.
-
-Some of the `ui-defaults` values can be appended by `:<clustername>` in order to have different settings depending on the current cluster. Those are notably `job_view_nodestats_selectedMetrics`, `job_view_polarPlotMetrics`, `job_view_selectedMetrics` and `plot_list_selectedMetrics`.
-
-## Environment Variables
-
-An example env file is found in this directory. Copy it to `.env` in the project root and adapt it for your needs.
-
-* `JWT_PUBLIC_KEY` and `JWT_PRIVATE_KEY`: Base64 encoded Ed25519 keys used for JSON Web Token (JWT) authentication. You can generate your own keypair using `go run ./cmd/gen-keypair/gen-keypair.go`. More information in [README_TOKENS.md](./README_TOKENS.md).
-* `SESSION_KEY`: Some random bytes used as secret for cookie-based sessions.
-* `LDAP_ADMIN_PASSWORD`: The LDAP admin user password (optional).
-* `CROSS_LOGIN_JWT_HS512_KEY`: Used for token based logins via another authentication service.
-* `LOGLEVEL`: Can be `err`, `warn`, `info` or `debug` (optional, `warn` by default). Can be used to reduce logging.

configs/README_TOKENS.md

@@ -1,51 +0,0 @@
-## Introduction
-
-ClusterCockpit uses JSON Web Tokens (JWT) for authorization of its APIs.
-JSON Web Token (JWT) is an open standard (RFC 7519) that defines a compact and self-contained way for securely transmitting information between parties as a JSON object.
-This information can be verified and trusted because it is digitally signed.
-In ClusterCockpit JWTs are signed using a public/private key pair using ECDSA.
-Because tokens are signed using public/private key pairs, the signature also certifies that only the party holding the private key is the one that signed it.
-Expiration of the generated tokens as well as the max. length of a browser session can be configured in the `config.json` file described [here](./README.md).
-
-The [Ed25519](https://ed25519.cr.yp.to/) algorithm for signatures was used because it is compatible with other tools that require authentication, such as NATS.io, and because these elliptic-curve methods provide simillar security with smaller keys compared to something like RSA. They are sligthly more expensive to validate, but that effect is negligible.
-
-## JWT Payload
-
-You may view the payload of a JWT token at [https://jwt.io/#debugger-io](https://jwt.io/#debugger-io).
-Currently ClusterCockpit sets the following claims:
-* `iat`: Issued at claim. The “iat” claim is used to identify the the time at which the JWT was issued. This claim can be used to determine the age of the JWT.
-* `sub`: Subject claim. Identifies the subject of the JWT, in our case this is the username.
-* `roles`: An array of strings specifying the roles set for the subject.
-* `exp`: Expiration date of the token (only if explicitly configured)
-
-It is important to know that JWTs are not encrypted, only signed. This means that outsiders cannot create new JWTs or modify existing ones, but they are able to read out the username.
-
-## Workflow
-
-1. Create a new ECDSA Public/private keypair:
-```
-$ go build ./cmd/gen-keypair/
-$ ./gen-keypair
-```
-2. Add keypair in your `.env` file. A template can be found in `./configs`.
-
-When a user logs in via the `/login` page using a browser, a session cookie (secured using the random bytes in the `SESSION_KEY` env. variable you shoud change as well) is used for all requests after the successfull login. The JWTs make it easier to use the APIs of ClusterCockpit using scripts or other external programs. The token is specified n the `Authorization` HTTP header using the [Bearer schema](https://datatracker.ietf.org/doc/html/rfc6750) (there is an example below). Tokens can be issued to users from the configuration view in the Web-UI or the command line. In order to use the token for API endpoints such as `/api/jobs/start_job/`, the user that executes it needs to have the `api` role. Regular users can only perform read-only queries and only look at data connected to jobs they started themselves.
-
-## cc-metric-store
-
-The [cc-metric-store](https://github.com/ClusterCockpit/cc-metric-store) also uses JWTs for authentication. As it does not issue new tokens, it does not need to kown the private key. The public key of the keypair that is used to generate the JWTs that grant access to the `cc-metric-store` can be specified in its `config.json`. When configuring the `metricDataRepository` object in the `cluster.json` file, you can put a token issued by ClusterCockpit itself.
-
-## Setup user and JWT token for REST API authorization
-
-1. Create user:
-```
-$ ./cc-backend --add-user <username>:api:<password> --no-server
-```
-2. Issue token for user:
-```
-$ ./cc-backend --jwt <username> --no-server
-```
-3. Use issued token token on client side:
-```
-$ curl -X GET "<API ENDPOINT>" -H "accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer <JWT TOKEN>"
-```

configs/config-demo.json

@@ -1,53 +1,106 @@
 {
+  "main": {
     "addr": "127.0.0.1:8080",
-    "archive": {
-        "kind": "file",
-        "path": "./var/job-archive"
+    "short-running-jobs-duration": 300,
+    "resampling": {
+      "minimumPoints": 600,
+      "trigger": 180,
+      "resolutions": [
+        240,
+        60
+      ]
     },
-    "clusters": [
-        {
-            "name": "fritz",
-            "metricDataRepository": {
-                "kind": "cc-metric-store",
-                "url": "http://localhost:8082",
-                "token": ""
-            },
-            "filterRanges": {
-                "numNodes": {
-                    "from": 1,
-                    "to": 64
-                },
-                "duration": {
-                    "from": 0,
-                    "to": 86400
-                },
-                "startTime": {
-                    "from": "2022-01-01T00:00:00Z",
-                    "to": null
-                }
-            }
+    "apiAllowedIPs": [
+      "*"
+    ],
+    "emission-constant": 317
+  },
+  "cron": {
+    "commit-job-worker": "2m",
+    "duration-worker": "5m",
+    "footprint-worker": "10m"
+  },
+  "archive": {
+    "kind": "file",
+    "path": "./var/job-archive"
+  },
+  "auth": {
+    "jwts": {
+      "max-age": "2000h"
+    }
+  },
+  "nats": {
+    "address": "nats://0.0.0.0:4222",
+    "username": "root",
+    "password": "root"
+  },
+  "clusters": [
+    {
+      "name": "fritz",
+      "metricDataRepository": {
+        "kind": "cc-metric-store-internal",
+        "url": "http://localhost:8082",
+        "token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJFZERTQSJ9.eyJ1c2VyIjoiYWRtaW4iLCJyb2xlcyI6WyJST0xFX0FETUlOIiwiUk9MRV9BTkFMWVNUIiwiUk9MRV9VU0VSIl19.d-3_3FZTsadPjDEdsWrrQ7nS0edMAR4zjl-eK7rJU3HziNBfI9PDHDIpJVHTNN5E5SlLGLFXctWyKAkwhXL-Dw"
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
         },
-        {
-            "name": "alex",
-            "metricDataRepository": {
-                "kind": "cc-metric-store",
-                "url": "http://localhost:8082",
-                "token": ""
-            },
-            "filterRanges": {
-                "numNodes": {
-                    "from": 1,
-                    "to": 64
-                },
-                "duration": {
-                    "from": 0,
-                    "to": 86400
-                },
-                "startTime": {
-                    "from": "2022-01-01T00:00:00Z",
-                    "to": null
-                }
-            }
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
         }
+      }
+    },
+    {
+      "name": "alex",
+      "metricDataRepository": {
+        "kind": "cc-metric-store-internal",
+        "url": "http://localhost:8082",
+        "token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJFZERTQSJ9.eyJ1c2VyIjoiYWRtaW4iLCJyb2xlcyI6WyJST0xFX0FETUlOIiwiUk9MRV9BTkFMWVNUIiwiUk9MRV9VU0VSIl19.d-3_3FZTsadPjDEdsWrrQ7nS0edMAR4zjl-eK7rJU3HziNBfI9PDHDIpJVHTNN5E5SlLGLFXctWyKAkwhXL-Dw"
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
+        }
+      }
+    }
+  ],
+  "metric-store": {
+    "checkpoints": {
+      "file-format": "avro",
+      "interval": "1h",
+      "directory": "./var/checkpoints",
+      "restore": "48h"
+    },
+    "archive": {
+      "interval": "1h",
+      "directory": "./var/archive"
+    },
+    "retention-in-memory": "48h",
+    "subscriptions": [
+      {
+        "subscribe-to": "hpc-nats",
+        "cluster-tag": "fritz"
+      },
+      {
+        "subscribe-to": "hpc-nats",
+        "cluster-tag": "alex"
+      }
     ]
-}
+  }
+}

configs/config-mariadb.json

@@ -0,0 +1,64 @@
+{
+  "addr": "127.0.0.1:8080",
+  "short-running-jobs-duration": 300,
+  "archive": {
+    "kind": "file",
+    "path": "./var/job-archive"
+  },
+  "jwts": {
+    "max-age": "2000h"
+  },
+  "db-driver": "mysql",
+  "db": "clustercockpit:demo@tcp(127.0.0.1:3306)/clustercockpit",
+  "enable-resampling": {
+    "trigger": 30,
+    "resolutions": [600, 300, 120, 60]
+  },
+  "emission-constant": 317,
+  "clusters": [
+    {
+      "name": "fritz",
+      "metricDataRepository": {
+        "kind": "cc-metric-store",
+        "url": "http://localhost:8082",
+        "token": ""
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
+        }
+      }
+    },
+    {
+      "name": "alex",
+      "metricDataRepository": {
+        "kind": "cc-metric-store",
+        "url": "http://localhost:8082",
+        "token": ""
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
+        }
+      }
+    }
+  ]
+}

configs/config.json

@@ -1,50 +1,54 @@
 {
+  "main": {
     "addr": "0.0.0.0:443",
-    "ldap": {
-        "url": "ldaps://test",
-        "user_base": "ou=people,ou=hpc,dc=test,dc=de",
-        "search_dn": "cn=hpcmonitoring,ou=roadm,ou=profile,ou=hpc,dc=test,dc=de",
-        "user_bind": "uid={username},ou=people,ou=hpc,dc=test,dc=de",
-        "user_filter": "(&(objectclass=posixAccount)(uid=*))"
-    },
     "https-cert-file": "/etc/letsencrypt/live/url/fullchain.pem",
     "https-key-file": "/etc/letsencrypt/live/url/privkey.pem",
     "user": "clustercockpit",
     "group": "clustercockpit",
-    "archive": {
-        "kind": "file",
-        "path": "./var/job-archive"
-    },
-    "validate": true,
-    "clusters": [
-        {
-            "name": "test",
-            "metricDataRepository": {
-                "kind": "cc-metric-store",
-                "url": "http://localhost:8082",
-                "token": "eyJhbGciOiJF-E-pQBQ"
-            },
-            "filterRanges": {
-                "numNodes": {
-                    "from": 1,
-                    "to": 64
-                },
-                "duration": {
-                    "from": 0,
-                    "to": 86400
-                },
-                "startTime": {
-                    "from": "2022-01-01T00:00:00Z",
-                    "to": null
-                }
-            }
+    "validate": false,
+    "apiAllowedIPs": ["*"],
+    "short-running-jobs-duration": 300,
+    "resampling": {
+      "minimumPoints": 600,
+      "trigger": 180,
+      "resolutions": [
+        240,
+        60
+      ]
+    }
+  },
+  "cron": {
+    "commit-job-worker": "2m",
+    "duration-worker": "5m",
+    "footprint-worker": "10m"
+  },
+  "archive": {
+    "kind": "file",
+    "path": "./var/job-archive"
+  },
+  "clusters": [
+    {
+      "name": "test",
+      "metricDataRepository": {
+        "kind": "cc-metric-store",
+        "url": "http://localhost:8082",
+        "token": "eyJhbGciOiJF-E-pQBQ"
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
         }
-    ],
-    "jwts": {
-        "cookieName": "",
-        "forceJWTValidationViaDatabase": false,
-        "max-age": 0,
-        "trustedExternalIssuer": ""
-    },
-    "short-running-jobs-duration": 300
+      }
+    }
+  ]
 }
+

configs/generate-subcluster.pl

@@ -117,10 +117,12 @@ foreach my $ln (split("\n", $topo)) {
 
 my $node;
 my @sockets;
+my @nodeCores;
 foreach my $socket ( @{$DOMAINS{socket}} ) {
     push @sockets, "[".join(",", @{$socket})."]";
-    $node .=  join(",", @{$socket})
+    push @nodeCores, join(",", @{$socket});
 }
+$node =  join(",", @nodeCores);
 $INFO{sockets} = join(",\n", @sockets);
 
 my @memDomains;
@@ -212,9 +214,27 @@ print <<"END";
       "socketsPerNode": $INFO{socketsPerNode},
       "coresPerSocket": $INFO{coresPerSocket},
       "threadsPerCore": $INFO{threadsPerCore},
-      "flopRateScalar": $flopsScalar,
-      "flopRateSimd": $flopsSimd,
-      "memoryBandwidth": $memBw,
+      "flopRateScalar": {
+           "unit": {
+               "base": "F/s",
+               "prefix": "G"
+           },
+           "value": $flopsScalar
+      },
+      "flopRateSimd": {
+           "unit": {
+               "base": "F/s",
+               "prefix": "G"
+           },
+           "value": $flopsSimd
+      },
+      "memoryBandwidth": {
+           "unit": {
+               "base": "B/s",
+               "prefix": "G"
+           },
+           "value": $memBw
+      },
       "nodes": "<FILL IN NODE RANGES>",
       "topology": {
           "node": [$node],

configs/uiConfig.json

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

docs/ConfigurationManagement.md

@@ -1,38 +0,0 @@
-# Release versions
-
-Versions are marked according to [semantic versioning] (https://semver.org).
-Each version embeds the following static assets in the binary:
-* Web frontend with javascript files and all static assets.
-* Golang template files for server-side rendering.
-* JSON schema files for validation.
-* Database migration files.
-
-The remaining external assets are:
-* The SQL database used.
-* The job archive
-* The configuration files `config.json` and `.env`.
-
-The external assets are versioned with integer IDs.
-This means that each release binary is bound to specific versions of the SQL
-database and the job archive.
-The configuration file is checked against the current schema at startup.
-The `-migrate-db` command line switch can be used to upgrade the SQL database
-to migrate from a previous version to the latest one.
-We offer a separate tool `archive-migration` to migrate an existing job archive
-archive from the previous to the latest version.
-
-# Versioning of APIs
-
-cc-backend provides two API backends:
-* A REST API for querying jobs.
-* A GraphQL API for data exchange between web frontend and cc-backend.
-
-The REST API will also be versioned. We still have to decide whether we will also
-support older REST API versions by versioning the endpoint URLs.
-The GraphQL API is for internal use and will not be versioned.
-
-# How to build
-
-In general it is recommended to use the provided release binary.
-In case you want to build build `cc-backend` please always use the provided makefile. This will ensure
-that the frontend is also built correctly and that the version in the binary is encoded in the binary.

docs/Hands-on.md

@@ -1,234 +0,0 @@
-# Hands-on setup ClusterCockpit from scratch (w/o docker)
-
-## Prerequisites
-* perl
-* go
-* npm
-* Optional: curl
-* Script migrateTimestamp.pl
-
-## Documentation
-You find READMEs or api docs in
-* ./cc-backend/configs
-* ./cc-backend/init
-* ./cc-backend/api
-
-## ClusterCockpit configuration files
-### cc-backend
-* `./.env` Passwords and Tokens set in the environment
-* `./config.json` Configuration options for cc-backend
-
-### cc-metric-store
-* `./config.json` Optional to overwrite configuration options
-
-### cc-metric-collector
-Not yet included in the hands-on setup.
-
-## Setup Components
-Start by creating a base folder for all of the following steps.
-* `mkdir clustercockpit`
-* `cd clustercockpit`
-
-### Setup cc-backend
-* Clone Repository
-    - `git clone https://github.com/ClusterCockpit/cc-backend.git`
-    - `cd cc-backend`
-* Build
-    - `make`
-* Activate & configure environment for cc-backend
-    - `cp configs/env-template.txt  .env`
-    - Optional: Have a look via `vim .env`
-    - Copy the `config.json` file included in this tarball into the root directory of cc-backend: `cp ../../config.json  ./`
-* Back to toplevel `clustercockpit`
-    - `cd ..`
-* Prepare Datafolder and Database file
-    - `mkdir var`
-    - `./cc-backend -migrate-db`
-
-### Setup cc-metric-store
-* Clone Repository
-    - `git clone https://github.com/ClusterCockpit/cc-metric-store.git`
-    - `cd cc-metric-store`
-* Build Go Executable
-    - `go get`
-    - `go build`
-* Prepare Datafolders
-    - `mkdir -p var/checkpoints`
-    - `mkdir -p var/archive`
-* Update Config
-    - `vim config.json`
-    - Exchange existing setting in `metrics` with the following:
-```
-"clock":      { "frequency": 60, "aggregation": null },
-"cpi":        { "frequency": 60, "aggregation": null },
-"cpu_load":   { "frequency": 60, "aggregation": null },
-"flops_any":  { "frequency": 60, "aggregation": null },
-"flops_dp":   { "frequency": 60, "aggregation": null },
-"flops_sp":   { "frequency": 60, "aggregation": null },
-"ib_bw":      { "frequency": 60, "aggregation": null },
-"lustre_bw":  { "frequency": 60, "aggregation": null },
-"mem_bw":     { "frequency": 60, "aggregation": null },
-"mem_used":   { "frequency": 60, "aggregation": null },
-"rapl_power": { "frequency": 60, "aggregation": null }
-```
-* Back to toplevel `clustercockpit`
-    - `cd ..`
-
-### Setup Demo Data
-* `mkdir source-data`
-* `cd source-data`
-* Download JobArchive-Source:
-    - `wget https://hpc-mover.rrze.uni-erlangen.de/HPC-Data/0x7b58aefb/eig7ahyo6fo2bais0ephuf2aitohv1ai/job-archive-dev.tar.xz`
-    - `tar xJf job-archive-dev.tar.xz`
-    - `mv ./job-archive ./job-archive-source`
-    - `rm ./job-archive-dev.tar.xz`
-* Download CC-Metric-Store Checkpoints:
-    - `mkdir -p cc-metric-store-source/checkpoints`
-    - `cd cc-metric-store-source/checkpoints`
-    - `wget https://hpc-mover.rrze.uni-erlangen.de/HPC-Data/0x7b58aefb/eig7ahyo6fo2bais0ephuf2aitohv1ai/cc-metric-store-checkpoints.tar.xz`
-    - `tar xf cc-metric-store-checkpoints.tar.xz`
-    - `rm cc-metric-store-checkpoints.tar.xz`
-* Back to `source-data`
-    - `cd ../..`
-* Run timestamp migration script. This may take tens of minutes!
-    - `cp ../migrateTimestamps.pl .`
-    - `./migrateTimestamps.pl`
-    - Expected output:
-```
-Starting to update start- and stoptimes in job-archive for emmy
-Starting to update start- and stoptimes in job-archive for woody
-Done for job-archive
-Starting to update checkpoint filenames and data starttimes for emmy
-Starting to update checkpoint filenames and data starttimes for woody
-Done for checkpoints
-```
-* Copy `cluster.json` files from source to migrated folders
-    - `cp source-data/job-archive-source/emmy/cluster.json cc-backend/var/job-archive/emmy/`
-    - `cp source-data/job-archive-source/woody/cluster.json cc-backend/var/job-archive/woody/`
-* Initialize Job-Archive in SQLite3 job.db and add demo user
-    - `cd cc-backend`
-    - `./cc-backend -init-db -add-user demo:admin:demo`
-    - Expected output:
-```
-<6>[INFO]    new user "demo" created (roles: ["admin"], auth-source: 0)
-<6>[INFO]    Building job table...
-<6>[INFO]    A total of 3936 jobs have been registered in 1.791 seconds.
-```
-* Back to toplevel `clustercockpit`
-    - `cd ..`
-
-### Startup both Apps
-* In cc-backend root:  `$./cc-backend -server -dev`
-    - Starts Clustercockpit at `http:localhost:8080`
-        - Log: `<6>[INFO]    HTTP server listening at :8080...`
-    - Use local internet browser to access interface
-        - You should see and be able to browse finished Jobs
-        - Metadata is read from SQLite3 database
-        - Metricdata is read from job-archive/JSON-Files
-    - Create User in settings (top-right corner)
-        - Name `apiuser`
-        - Username `apiuser`
-        - Role `API`
-        - Submit & Refresh Page
-    - Create JTW for `apiuser`
-        - In Userlist, press `Gen. JTW` for `apiuser`
-        - Save JWT for later use
-* In cc-metric-store root:  `$./cc-metric-store`
-    - Start the cc-metric-store on `http:localhost:8081`, Log:
-```
-2022/07/15 17:17:42 Loading checkpoints newer than 2022-07-13T17:17:42+02:00
-2022/07/15 17:17:45 Checkpoints loaded (5621 files, 319 MB, that took 3.034652s)
-2022/07/15 17:17:45 API http endpoint listening on '0.0.0.0:8081'
-```
-    - Does *not* have a graphical interface
-    - Otpional: Test function by executing:
-```
-$ curl -H "Authorization: Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJFZERTQSJ9.eyJ1c2VyIjoiYWRtaW4iLCJyb2xlcyI6WyJST0xFX0FETUlOIiwiUk9MRV9BTkFMWVNUIiwiUk9MRV9VU0VSIl19.d-3_3FZTsadPjDEdsWrrQ7nS0edMAR4zjl-eK7rJU3HziNBfI9PDHDIpJVHTNN5E5SlLGLFXctWyKAkwhXL-Dw" -D - "http://localhost:8081/api/query" -d "{ \"cluster\": \"emmy\", \"from\": $(expr $(date +%s) - 60), \"to\": $(date +%s), \"queries\": [{
-  \"metric\": \"flops_any\",
-  \"host\": \"e1111\"
-}] }"
-
-HTTP/1.1 200 OK
-Content-Type: application/json
-Date: Fri, 15 Jul 2022 13:57:22 GMT
-Content-Length: 119
-{"results":[[JSON-DATA-ARRAY]]}
-```
-
-### Development API web interfaces
-The `-dev` flag enables web interfaces to document and test the apis:
-* http://localhost:8080/playground - A GraphQL playground. To use it you must have a authenticated session in the same browser.
-* http://localhost:8080/swagger - A Swagger UI. To use it you have to be logged out, so no user session in the same browser. Use the JWT token with role Api generate previously to authenticate via http header.
-
-### Use cc-backend API to start job
-* Enter the URL `http://localhost:8080/swagger/index.html` in your browser.
-* Enter your JWT token you generated for the API user by clicking the green Authorize button in the upper right part of the window.
-* Click the `/job/start_job` endpoint and click the Try it out button.
-* Enter the following json into the request body text area and fill in a recent start timestamp by executing `date +%s`.:
-```
-{
-    "jobId":         100000,
-    "arrayJobId":    0,
-    "user":          "ccdemouser",
-    "subCluster":    "main",
-    "cluster":       "emmy",
-    "startTime":    <date +%s>,
-    "project":       "ccdemoproject",
-    "resources":  [
-        {"hostname":  "e0601"},
-        {"hostname":  "e0823"},
-        {"hostname":  "e0337"},
-        {"hostname": "e1111"}],
-    "numNodes":      4,
-    "numHwthreads":  80,
-    "walltime":      86400
-}
-```
-* The response body should be the database id of the started job, for example:
-```
-{
-  "id": 3937
-}
-```
-* Check in ClusterCockpit
-    - User `ccdemouser` should appear in Users-Tab with one running job
-    - It could take up to 5 Minutes until the Job is displayed with some current data (5 Min Short-Job Filter)
-    - Job then is marked with a green `running` tag
-    - Metricdata displayed is read from cc-metric-store!
-
-
-### Use cc-backend API to stop job
-* Enter the URL `http://localhost:8080/swagger/index.html` in your browser.
-* Enter your JWT token you generated for the API user by clicking the green Authorize button in the upper right part of the window.
-* Click the `/job/stop_job/{id}` endpoint and click the Try it out button.
-* Enter the database id at id that was returned by `start_job` and copy the following into the request body. Replace the timestamp with a recent one:
-```
-{
-  "cluster": "emmy",
-  "jobState": "completed",
-  "stopTime": <RECENT TS>
-}
-```
-* On success a json document with the job meta data is returned.
-
-* Check in ClusterCockpit
-    - User `ccdemouser` should appear in Users-Tab with one completed job
-    - Job is no longer marked with a green `running` tag -> Completed!
-    - Metricdata displayed is now read from job-archive!
-* Check in job-archive
-    - `cd ./cc-backend/var/job-archive/emmy/100/000`
-    - `cd $STARTTIME`
-    - Inspect `meta.json` and `data.json`
-
-## Helper scripts
-* In this tarball you can find the perl script `generate_subcluster.pl` that helps to generate the subcluster section for your system.
-Usage:
-* Log into an exclusive cluster node.
-* The LIKWID tools likwid-topology and likwid-bench must be in the PATH!
-* `$./generate_subcluster.pl` outputs the subcluster section on `stdout`
-
-Please be aware that
-* You have to enter the name and node list for the subCluster manually.
-* GPU detection only works if LIKWID was build with Cuda avalable and you run likwid-topology also with Cuda loaded.
-* Do not blindly trust the measured peakflops values.
-* Because the script blindly relies on the CSV format output by likwid-topology this is a fragile undertaking!

docs/JWT-Handling.md

@@ -1,82 +0,0 @@
-## Introduction
-
-ClusterCockpit uses JSON Web Tokens (JWT) for authorization of its APIs.
-JSON Web Token (JWT) is an open standard (RFC 7519) that defines a compact and self-contained way for securely transmitting information between parties as a JSON object.
-This information can be verified and trusted because it is digitally signed.
-In ClusterCockpit JWTs are signed using a public/private key pair using ECDSA.
-Because tokens are signed using public/private key pairs, the signature also certifies that only the party holding the private key is the one that signed it.
-Currently JWT tokens in ClusterCockpit not yet expire.
-
-## JWT Payload
-
-You may view the payload of a JWT token at [https://jwt.io/#debugger-io](https://jwt.io/#debugger-io).
-Currently ClusterCockpit sets the following claims:
-* `iat`: Issued at claim. The “iat” claim is used to identify the the time at which the JWT was issued. This claim can be used to determine the age of the JWT.
-* `sub`: Subject claim. Identifies the subject of the JWT, in our case this is the username.
-* `roles`: An array of strings specifying the roles set for the subject.
-
-## Workflow
-
-1. Create a new ECDSA Public/private keypair:
-```
-$ go build ./tools/gen-keypair.go
-$ ./gen-keypair
-```
-2. Add keypair in your `.env` file. A template can be found in `./configs`.
-
-There are two usage scenarios:
-* The APIs are used during a browser session. In this case on login a JWT token is issued on login, that is used by the web frontend to authorize against the GraphQL and REST APIs.
-* The REST API is used outside a browser session, e.g. by scripts. In this case you have to issue a token manually. This possible from within the configuration view or on the command line. It is recommended to issue a JWT token in this case for a special user that only has the `api` role. By using different users for different purposes a fine grained access control and access revocation management is possible.
-
-The token is commonly specified in the Authorization HTTP header using the Bearer schema.
-
-## Setup user and JWT token for REST API authorization
-
-1. Create user:
-```
-$ ./cc-backend --add-user <username>:api:<Password> --no-server
-```
-2. Issue token for user:
-```
-$ ./cc-backend -jwt <username>  -no-server
-```
-3. Use issued token token on client side:
-```
-$ curl -X GET "<API ENDPOINT>" -H "accept: application/json"  -H "Content-Type: application/json"  -H "Authorization: Bearer <JWT TOKEN>"
-```
-
-## Accept externally generated JWTs provided via cookie
-If there is an external service like an AuthAPI that can generate JWTs and hand them over to ClusterCockpit via cookies, CC can be configured to accept them:
-
-1. `.env`: CC needs a public ed25519 key to verify foreign JWT signatures. Public keys in PEM format can be converted with the instructions in [/tools/convert-pem-pubkey-for-cc](../tools/convert-pem-pubkey-for-cc/Readme.md) .
-
-```
-CROSS_LOGIN_JWT_PUBLIC_KEY="+51iXX8BdLFocrppRxIw52xCOf8xFSH/eNilN5IHVGc="
-```
-
-2. `config.json`: Insert a name for the cookie (set by the external service) containing the JWT so that CC knows where to look at. Define a trusted issuer (JWT claim 'iss'), otherwise it will be rejected.
-If you want usernames and user roles from JWTs ('sub' and 'roles' claim) to be validated against CC's internal database, you need to enable it here. Unknown users will then be rejected and roles set via JWT will be ignored.
-
-```json
-"jwts": {
-    "cookieName": "access_cc",
-    "forceJWTValidationViaDatabase": true,
-    "trustedExternalIssuer": "auth.example.com"
-}
-```
-
-3. Make sure your external service includes the same issuer (`iss`) in its JWTs. Example JWT payload:
-
-```json
-{
-  "iat": 1668161471,
-  "nbf": 1668161471,
-  "exp": 1668161531,
-  "sub": "alice",
-  "roles": [
-    "user"
-  ],
-  "jti": "a1b2c3d4-1234-5678-abcd-a1b2c3d4e5f6",
-  "iss": "auth.example.com"
-}
-```

docs/Job-Archive.md

@@ -1,78 +0,0 @@
-The job archive specifies an exchange format for job meta and performance metric
-data. It consists of two parts:
-* a [SQLite database schema](https://github.com/ClusterCockpit/cc-backend/wiki/Job-Archive#sqlite-database-schema)  for job meta data and performance statistics
-* a [Json file format](https://github.com/ClusterCockpit/cc-backend/wiki/Job-Archive#json-file-format) together with a [Directory hierarchy specification](https://github.com/ClusterCockpit/cc-backend/wiki/Job-Archive#directory-hierarchy-specification)
-
-By using an open, portable and simple specification based on files it is
-possible to exchange job performance data for research and analysis purposes as
-well as use it as a robust way for archiving job performance data to disk.
-
-# SQLite database schema
-## Introduction
-
-A SQLite 3 database schema is provided to standardize the job meta data
-information in a portable way. The schema also includes optional columns for job
-performance statistics (called a job performance footprint). The database acts
-as a front end to filter and select subsets of job IDs, that are the keys to get
-the full job performance data in the job performance tree hierarchy.
-
-## Database schema
-
-The schema includes 3 tables: the job table, a tag table and a jobtag table
-representing the MANY-TO-MANY relation between jobs and tags. The SQL schema is
-specified
-[here](https://github.com/ClusterCockpit/cc-specifications/blob/master/schemas/jobs-sqlite.sql).
-Explanation of the various columns including the JSON datatypes is documented
-[here](https://github.com/ClusterCockpit/cc-specifications/blob/master/datastructures/job-meta.schema.json).
-
-# Directory hierarchy specification
-
-## Specification
-
-To manage the number of directories within a single directory a tree approach is
-used splitting the integer job ID. The job id is split in junks of 1000 each.
-Usually 2 layers of directories is sufficient but the concept can be used for an
-arbitrary number of layers.
-
-For a 2 layer schema this can be achieved with (code example in Perl):
-``` perl
-$level1 = $jobID/1000;
-$level2 = $jobID%1000;
-$dstPath = sprintf("%s/%s/%d/%03d", $trunk, $destdir, $level1, $level2);
-```
-
-## Example
-
-For the job ID 1034871 the directory path is `./1034/871/`.
-
-# Json file format
-## Overview
-
-Every cluster must be configured in a `cluster.json` file.
-
-The job data consists of two files:
-* `meta.json`: Contains job meta information and job statistics.
-* `data.json`: Contains complete job data with time series
-
-The description of the json format specification is available as [[json
-schema|https://json-schema.org/]] format file. The latest version of the json
-schema is part of the `cc-backend` source tree. For external reference it is
-also available in a separate repository.
-
-## Specification `cluster.json`
-
-The json schema specification is available
-[here](https://github.com/ClusterCockpit/cc-specifications/blob/master/datastructures/cluster.schema.json).
-
-## Specification `meta.json`
-
-The json schema specification is available
-[here](https://github.com/RRZE-HPC/HPCJobDatabase/blob/master/json-schema/job-meta.schema.json).
-
-## Specification `data.json`
-
-The json schema specification is available
-[here](https://github.com/RRZE-HPC/HPCJobDatabase/blob/master/json-schema/job-data.schema.json).
-Metric time series data is stored for a fixed time step. The time step is set
-per metric. If no value is available for a metric time series data timestamp
-`null` is entered. 

docs/adm-customization.md

@@ -1,24 +0,0 @@
-# Overview
-
-Customizing `cc-backend` means changing the logo, legal texts, and the login
-template instead of the placeholders.
-
-# Replace legal texts
-
-To replace the `imprint.tmpl` and `privacy.tmpl` legal texts, you can place your
-version in `./var/`. At startup `cc-backend` will check if `./var/imprint.tmpl` and/or
-`./var/privacy.tmpl` exist and use them instead of the built-in placeholders.
-You can use the placeholders in `web/templates` as a blueprint.
-
-# Replace login template
-To replace the default login layout and styling, you can place your version in
-`./var/`. At startup `cc-backend` will check if `./var/login.tmpl` exist and use
-it instead of the built-in placeholder. You can use the default temaplte
-`web/templates/login.tmpl` as a blueprint.
-
-# Replace logo
-To change the logo displayed in the navigation bar, you can provide the file
-`logo.png` in the folder `./var/img/`. On startup `cc-backend` will check if the
-folder exists and use the images provided there instead of the built-in images.
-You may also place additional images there you use in a custom login template.
-

docs/adm-upgrade.md

@@ -1,78 +0,0 @@
-In general, an upgrade is nothing more than a replacement of the binary file.
-All the necessary files, except the database file, the configuration file and
-the job archive, are embedded in the binary file. It is recommended to use a
-directory where the file names of the binary files are named with a version
-indicator. This can be, for example, the date or the Unix epoch time. A symbolic
-link points to the version to be used. This makes it easier to switch to earlier
-versions.
-
-The database and the job archive are versioned. Each release binary supports
-specific versions of the database and job archive. If a version mismatch is
-detected, the application is terminated and migration is required.
-
-**IMPORTANT NOTE**
-
-It is recommended to make a backup copy of the database before each update. This
-is mandatory in case the database needs to be migrated. In the case of sqlite,
-this means to stopping `cc-backend` and copying the sqlite database file
-somewhere.
-
-#  Migrating the database
-
-After you have backed up the database, run the following command to migrate the
-database to the latest version:
-```
-$ ./cc-backend -migrate-db
-```
-
-The migration files are embedded in the binary and can also be viewed in the cc
-backend [source tree](https://github.com/ClusterCockpit/cc-backend/tree/master/internal/repository/migrations).
-There are separate migration files for both supported
-database backends.
-We use the [migrate library](https://github.com/golang-migrate/migrate).
-
-If something goes wrong, you can check the status and get the current schema
-(here for sqlite):
-```
-$ sqlite3 var/job.db
-```
-In the sqlite console execute:
-```
-.schema
-```
-to get the current databse schema.
-You can query the current version and whether the migration failed with:
-```
-SELECT * FROM schema_migrations;
-```
-The first column indicates the current database version and the second column is
-a dirty flag indicating whether the migration was successful.
-
-# Migrating the job archive
-
-Job archive migration requires a separate tool (`archive-migration`), which is
-part of the cc-backend source tree (build with `go build ./tools/archive-migration`)
-and is also provided as part of the releases.
-
-Migration is supported only between two successive releases. The migration tool
-migrates the existing job archive to a new job archive. This means that there
-must be enough disk space for two complete job archives. If the tool is called
-without options:
-```
-$ ./archive-migration
-```
-
-it is assumed that a job archive exists in `./var/job-archive`. The new job
-archive is written to `./var/job-archive-new`. Since execution is threaded in case
-of a fatal error, it is impossible to determine in which job the error occurred.
-In this case, you can run the tool in debug mode (with the `-debug` flag). In
-debug mode, threading is disabled and the job ID of each migrated job is output.
-Jobs with empty files will be skipped. Between multiple runs of the tools, the
-`job-archive-new` directory must be moved or deleted.
-
-The `cluster.json` files in `job-archive-new` must be checked for errors, especially
-whether the aggregation attribute is set correctly for all metrics.
-
-Migration takes several hours for relatively large job archives (several hundred
-GB). A versioned job archive contains a version.txt file in the root directory
-of the job archive. This file contains the version as an unsigned integer.

docs/dev-frontend.md

@@ -1,33 +0,0 @@
-## Tips for frontend development
-
-The frontend assets including the Svelte js files are per default embedded in
-the bgo binary. To enable a quick turnaround cycle for web development of the
-frontend disable embedding of static assets in `config.json`:
-```
-"embed-static-files": false,
-"static-files": "./web/frontend/public/",
-
-```
-
-Start the node build process (in directory `./web/frontend`) in development mode:
-```
-$ npm run dev
-```
-
-This will start the build process in listen mode. Whenever you change a source
-files the depending javascript targets will be automatically rebuild.
-In case the javascript files are minified you may need to set the production
-flag by hand to false in `./web/frontend/rollup.config.mjs`:
-```
-const production = false
-```
-
-Usually this should work automatically.
-
-Because the files are still served by ./cc-backend you have to reload the view
-explicitly in your browser.
-
-A common setup is to have three terminals open:
-* One running cc-backend (working directory repository root): `./cc-backend -server -dev`
-* Another running npm in developer mode (working directory `./web/frontend`): `npm run dev`
-* And the last one editing the frontend source files

docs/dev-testing.md

@@ -1,34 +0,0 @@
-## Overview
-
-We use the standard golang testing environment.
-
-The following conventions are used:
-
-* *White box unit tests*: Tests for internal functionality are placed in files
-* *Black box unit tests*: Tests for public interfaces are placed in files
-with `<package name>_test.go` and belong to the package `<package_name>_test`.
-There only exists one package test file per package.
-* *Integration tests*: Tests that use multiple componenents are placed in a
-package test file. These are named `<package name>_test.go` and belong to the
-package `<package_name>_test`.
-* *Test assets*: Any required files are placed in a directory `./testdata`
-within each package directory.
-
-## Executing tests
-
-Visual Studio Code has a very good golang test integration.
-For debugging a test this is the recommended solution.
-
-The Makefile provided by us has a `test` target that executes:
-```
-$ go clean -testcache
-$ go build ./...
-$ go vet ./...
-$ go test ./...
-```
-
-Of course the commands can also be used on the command line.
-For details about golang testing refer to the standard documentation:
-
-* [Testing package](https://pkg.go.dev/testing)
-* [go test command](https://pkg.go.dev/cmd/go#hdr-Test_packages)

docs/migrateTimestamps.pl

@@ -1,229 +0,0 @@
-#!/usr/bin/env perl
-use strict;
-use warnings;
-use utf8;
-
-use JSON::PP; # from Perl default install
-use Time::Local qw( timelocal ); # from Perl default install
-use Time::Piece; # from Perl default install
-
-### JSON
-my $json = JSON::PP->new->allow_nonref;
-
-### TIME AND DATE
-# now
-my $localtime = localtime;
-my $epochtime = $localtime->epoch;
-# 5 days ago: Via epoch due to possible reverse month borders
-my $epochlessfive = $epochtime - (86400 * 5);
-my $locallessfive = localtime($epochlessfive);
-# Calc like `date --date 'TZ="Europe/Berlin" 0:00 5 days ago' +%s`)
-my ($day, $month, $year) = ($locallessfive->mday, $locallessfive->_mon, $locallessfive->year);
-my $checkpointStart = timelocal(0, 0, 0, $day, $month, $year);
-# for checkpoints
-my $halfday = 43200;
-
-### JOB-ARCHIVE
-my $archiveTarget = './cc-backend/var/job-archive';
-my $archiveSrc = './source-data/job-archive-source';
-my @ArchiveClusters;
-
-# Gen folder
-if ( not -d $archiveTarget ){
-    mkdir( $archiveTarget ) or die "Couldn't create $archiveTarget directory, $!";
-}
-
-# Get clusters by job-archive/$subfolder
-opendir my $dh, $archiveSrc  or die "can't open directory: $!";
-while ( readdir $dh ) {
-    chomp; next if $_ eq '.' or $_ eq '..'  or $_ eq 'job-archive';
-    my $cluster = $_;
-    push @ArchiveClusters, $cluster;
-}
-
-# start for jobarchive
-foreach my $cluster ( @ArchiveClusters ) {
-    print "Starting to update start- and stoptimes in job-archive for $cluster\n";
-
-    my $clusterTarget = "$archiveTarget/$cluster";
-
-    if ( not -d $clusterTarget ){
-        mkdir( $clusterTarget ) or die "Couldn't create $clusterTarget directory, $!";
-    }
-
-    opendir my $dhLevel1, "$archiveSrc/$cluster" or die "can't open directory: $!";
-    while ( readdir $dhLevel1 ) {
-        chomp; next if $_ eq '.' or $_ eq '..';
-        my $level1 = $_;
-
-        if ( -d "$archiveSrc/$cluster/$level1" ) {
-            opendir my $dhLevel2, "$archiveSrc/$cluster/$level1" or die "can't open directory: $!";
-            while ( readdir $dhLevel2 ) {
-                chomp; next if $_ eq '.' or $_ eq '..';
-                my $level2 = $_;
-                my $jobSource = "$archiveSrc/$cluster/$level1/$level2";
-                my $jobOrigin = "$jobSource";
-                my $jobTargetL1 = "$clusterTarget/$level1";
-                my $jobTargetL2 = "$jobTargetL1/$level2";
-
-                # check if files are directly accessible (old format) else get subfolders as file and update path
-                if ( ! -e "$jobSource/meta.json") {
-                    opendir(D, "$jobSource") || die "Can't open directory $jobSource: $!\n";
-                    my @folders = readdir(D);
-                    closedir(D);
-                    if (!@folders) {
-                        next;
-                    }
-
-                    foreach my $folder ( @folders ) {
-                        next if $folder eq '.' or $folder eq '..';
-                        $jobSource = "$jobSource/".$folder;
-                    }
-                }
-                # check if subfolder contains file, else skip
-                if ( ! -e "$jobSource/meta.json") {
-                    print "$jobSource skipped\n";
-                    next;
-                }
-
-                open my $metafh, '<', "$jobSource/meta.json" or die "Can't open file $!";
-                my $rawstr = do { local $/; <$metafh> };
-                close($metafh);
-                my $metadata = $json->decode($rawstr);
-
-                # NOTE Start meta.json iteration here
-                # my $random_number = int(rand(UPPERLIMIT)) + LOWERLIMIT;
-                # Set new startTime: Between 5 days and 1 day before now
-
-                #  Remove id from attributes
-                $metadata->{startTime} = $epochtime - (int(rand(432000)) + 86400);
-                $metadata->{stopTime} = $metadata->{startTime} + $metadata->{duration};
-
-                # Add starttime subfolder to target path
-                my $jobTargetL3 = "$jobTargetL2/".$metadata->{startTime};
-
-                if ( not -d $jobTargetL1 ){
-                    mkdir( $jobTargetL1 ) or die "Couldn't create $jobTargetL1 directory, $!";
-                }
-
-                if ( not -d $jobTargetL2 ){
-                    mkdir( $jobTargetL2 ) or die "Couldn't create $jobTargetL2 directory, $!";
-                }
-
-                # target is not directory
-                if ( not -d $jobTargetL3 ){
-                    mkdir( $jobTargetL3 ) or die "Couldn't create $jobTargetL3 directory, $!";
-
-                    my $outstr = $json->encode($metadata);
-                    open my $metaout, '>', "$jobTargetL3/meta.json" or die "Can't write to file $!";
-                    print $metaout $outstr;
-                    close($metaout);
-
-                    open my $datafh, '<', "$jobSource/data.json" or die "Can't open file $!";
-                    my $datastr = do { local $/; <$datafh> };
-                    close($datafh);
-
-                    open my $dataout, '>', "$jobTargetL3/data.json" or die "Can't write to file $!";
-                    print $dataout $datastr;
-                    close($dataout);
-                }
-            }
-        }
-    }
-}
-print "Done for job-archive\n";
-sleep(1);
-exit;
-
-## CHECKPOINTS
-my $checkpTarget = './cc-metric-store/var/checkpoints';
-my $checkpSource = './source-data/cc-metric-store-source/checkpoints';
-my @CheckpClusters;
-
-# Gen folder
-if ( not -d $checkpTarget ){
-    mkdir( $checkpTarget ) or die "Couldn't create $checkpTarget directory, $!";
-}
-
-# Get clusters by cc-metric-store/$subfolder
-opendir my $dhc, $checkpSource  or die "can't open directory: $!";
-while ( readdir $dhc ) {
-    chomp; next if $_ eq '.' or $_ eq '..'  or $_ eq 'job-archive';
-    my $cluster = $_;
-    push @CheckpClusters, $cluster;
-}
-closedir($dhc);
-
-# start for checkpoints
-foreach my $cluster ( @CheckpClusters ) {
-    print "Starting to update checkpoint filenames and data starttimes for $cluster\n";
-
-    my $clusterTarget = "$checkpTarget/$cluster";
-
-    if ( not -d $clusterTarget ){
-        mkdir( $clusterTarget ) or die "Couldn't create $clusterTarget directory, $!";
-    }
-
-    opendir my $dhLevel1, "$checkpSource/$cluster" or die "can't open directory: $!";
-    while ( readdir $dhLevel1 ) {
-        chomp; next if $_ eq '.' or $_ eq '..';
-        # Nodename as level1-folder
-        my $level1 = $_;
-
-        if ( -d "$checkpSource/$cluster/$level1" ) {
-
-            my $nodeSource = "$checkpSource/$cluster/$level1/";
-            my $nodeOrigin = "$nodeSource";
-            my $nodeTarget = "$clusterTarget/$level1";
-            my @files;
-
-            if ( -e "$nodeSource/1609459200.json") { # 1609459200 == First Checkpoint time in latest dump
-                opendir(D, "$nodeSource") || die "Can't open directory $nodeSource: $!\n";
-                while ( readdir D ) {
-                    chomp; next if $_ eq '.' or $_ eq '..';
-                    my $nodeFile = $_;
-                    push @files, $nodeFile;
-                }
-                closedir(D);
-                my $length = @files;
-                if (!@files || $length != 14) { # needs 14 files == 7 days worth of data
-                    next;
-                }
-            } else {
-                next;
-            }
-
-            # sort for integer timestamp-filename-part (moduleless): Guarantees start with index == 0 == 1609459200.json
-            my @sortedFiles = sort { ($a =~ /^([0-9]{10}).json$/)[0] <=> ($b =~ /^([0-9]{10}).json$/)[0] } @files;
-
-            if ( not -d $nodeTarget ){
-                mkdir( $nodeTarget ) or die "Couldn't create $nodeTarget directory, $!";
-
-                while (my ($index, $file) = each(@sortedFiles)) {
-                    open my $checkfh, '<', "$nodeSource/$file" or die "Can't open file $!";
-                    my $rawstr = do { local $/; <$checkfh> };
-                    close($checkfh);
-                    my $checkpdata = $json->decode($rawstr);
-
-                    my $newTimestamp = $checkpointStart + ($index * $halfday);
-                    # Get Diff from old Timestamp
-                    my $timeDiff = $newTimestamp - $checkpdata->{from};
-                    # Set new timestamp
-                    $checkpdata->{from} = $newTimestamp;
-
-                    foreach my $metric (keys %{$checkpdata->{metrics}}) {
-                        $checkpdata->{metrics}->{$metric}->{start} += $timeDiff;
-                    }
-
-                    my $outstr = $json->encode($checkpdata);
-
-                    open my $checkout, '>', "$nodeTarget/$newTimestamp.json" or die "Can't write to file $!";
-                    print $checkout $outstr;
-                    close($checkout);
-                }
-            }
-        }
-    }
-    closedir($dhLevel1);
-}
-print "Done for checkpoints\n";

docs/searchbar.md

@@ -1,35 +0,0 @@
-# Docs for ClusterCockpit Searchbar
-
-## Usage
-
-* Searchtags are implemented as `type:<query>` search-string
-  * Types `jobId, jobName, projectId, username, name` for roles `admin` and `support`
-    * `jobName` is jobName as persisted in `job.meta_data` table-column
-    * `username` is actual account identifier as persisted in `job.user` table-column
-    * `name` is account owners name as persisted in `user.name` table-column
-  * Types `jobId, jobName, projectId` for role `user`
-  * Examples:
-    * `jobName:myJob12`
-    * `jobId:123456`
-    * `username:abcd100`
-    * `name:Paul`
-* If no searchTag used: Best guess search with the following hierarchy
-  * `jobId -> username -> name -> projectId -> jobName`
-* Destinations:
-  * JobId: Job-Table (Allows multiple identical matches, e.g. JobIds from different clusters)
-  * JobName: Job-Table (Allows multiple identical matches, e.g. JobNames from different clusters)
-  * ProjectId: Job-Table
-  * Username: Users-Table
-    * **Please Note**: Only users with jobs will be shown in table! I.e., Users without jobs will be missing in table. Also, a `Last 30 Days` is active by default and might filter out expected users.
-  * Name: Users-Table
-    * **Please Note**: Only users with jobs will be shown in table! I.e., Users without jobs will be missing in table. Also, a `Last 30 Days` is active by default and might filter out expected users.
-  * Best guess search always redirects to Job-Table or `/monitoring/user/$USER` (first username match)
-  * Unprocessable queries will display messages detailing the cause (Info, Warning, Error)
-* Spaces trimmed (both for searchTag and queryString)
-  * `  job12` == `job12`
-  * `projectID : abcd ` == `projectId:abcd`
-* `jobName`- and `name-`queries work with a part of the target-string
-  * `jobName:myjob` for jobName "myjob_cluster1"
-  * `name:Paul` for name "Paul Atreides"
-
-* JobName GQL Query is resolved as matching the query as a part of the whole metaData-JSON in the SQL DB.

go.mod

@@ -1,88 +1,129 @@
 module github.com/ClusterCockpit/cc-backend
 
-go 1.18
+go 1.24.0
 
-require (
-	github.com/99designs/gqlgen v0.17.24
-	github.com/ClusterCockpit/cc-units v0.4.0
-	github.com/Masterminds/squirrel v1.5.3
-	github.com/go-co-op/gocron v1.25.0
-	github.com/go-ldap/ldap/v3 v3.4.4
-	github.com/go-sql-driver/mysql v1.7.0
-	github.com/golang-jwt/jwt/v4 v4.5.0
-	github.com/golang-migrate/migrate/v4 v4.15.2
-	github.com/google/gops v0.3.27
-	github.com/gorilla/handlers v1.5.1
-	github.com/gorilla/mux v1.8.0
-	github.com/gorilla/sessions v1.2.1
-	github.com/influxdata/influxdb-client-go/v2 v2.12.2
-	github.com/jmoiron/sqlx v1.3.5
-	github.com/mattn/go-sqlite3 v1.14.16
-	github.com/prometheus/client_golang v1.14.0
-	github.com/prometheus/common v0.40.0
-	github.com/qustavo/sqlhooks/v2 v2.1.0
-	github.com/santhosh-tekuri/jsonschema/v5 v5.2.0
-	github.com/swaggo/http-swagger v1.3.3
-	github.com/swaggo/swag v1.8.10
-	github.com/vektah/gqlparser/v2 v2.5.1
-	golang.org/x/crypto v0.6.0
-	golang.org/x/exp v0.0.0-20230510235704-dd950f8aeaea
+toolchain go1.24.1
+
+tool (
+	github.com/99designs/gqlgen
+	github.com/swaggo/swag/cmd/swag
 )
 
 require (
+	github.com/99designs/gqlgen v0.17.85
+	github.com/ClusterCockpit/cc-lib v1.0.2
+	github.com/Masterminds/squirrel v1.5.4
+	github.com/aws/aws-sdk-go-v2 v1.41.1
+	github.com/aws/aws-sdk-go-v2/config v1.32.7
+	github.com/aws/aws-sdk-go-v2/credentials v1.19.7
+	github.com/aws/aws-sdk-go-v2/service/s3 v1.95.0
+	github.com/coreos/go-oidc/v3 v3.17.0
+	github.com/expr-lang/expr v1.17.7
+	github.com/go-co-op/gocron/v2 v2.19.0
+	github.com/go-ldap/ldap/v3 v3.4.12
+	github.com/go-sql-driver/mysql v1.9.3
+	github.com/golang-jwt/jwt/v5 v5.3.0
+	github.com/golang-migrate/migrate/v4 v4.19.1
+	github.com/google/gops v0.3.28
+	github.com/gorilla/handlers v1.5.2
+	github.com/gorilla/mux v1.8.1
+	github.com/gorilla/sessions v1.4.0
+	github.com/influxdata/line-protocol/v2 v2.2.1
+	github.com/jmoiron/sqlx v1.4.0
+	github.com/joho/godotenv v1.5.1
+	github.com/linkedin/goavro/v2 v2.14.1
+	github.com/mattn/go-sqlite3 v1.14.33
+	github.com/nats-io/nats.go v1.47.0
+	github.com/prometheus/client_golang v1.23.2
+	github.com/prometheus/common v0.67.4
+	github.com/qustavo/sqlhooks/v2 v2.1.0
+	github.com/santhosh-tekuri/jsonschema/v5 v5.3.1
+	github.com/stretchr/testify v1.11.1
+	github.com/swaggo/http-swagger v1.3.4
+	github.com/swaggo/swag v1.16.6
+	github.com/vektah/gqlparser/v2 v2.5.31
+	golang.org/x/crypto v0.46.0
+	golang.org/x/oauth2 v0.34.0
+	golang.org/x/time v0.14.0
+)
+
+require (
+	filippo.io/edwards25519 v1.1.0 // indirect
 	github.com/Azure/go-ntlmssp v0.0.0-20221128193559-754e69321358 // indirect
 	github.com/KyleBanks/depth v1.2.1 // indirect
-	github.com/agnivade/levenshtein v1.1.1 // indirect
-	github.com/apapsch/go-jsonmerge/v2 v2.0.0 // indirect
+	github.com/agnivade/levenshtein v1.2.1 // indirect
+	github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.4 // indirect
+	github.com/aws/aws-sdk-go-v2/feature/ec2/imds v1.18.17 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/configsources v1.4.17 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.7.17 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/ini v1.8.4 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/v4a v1.4.16 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/accept-encoding v1.13.4 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/checksum v1.9.7 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/presigned-url v1.13.17 // indirect
+	github.com/aws/aws-sdk-go-v2/service/internal/s3shared v1.19.16 // indirect
+	github.com/aws/aws-sdk-go-v2/service/signin v1.0.5 // indirect
+	github.com/aws/aws-sdk-go-v2/service/sso v1.30.9 // indirect
+	github.com/aws/aws-sdk-go-v2/service/ssooidc v1.35.13 // indirect
+	github.com/aws/aws-sdk-go-v2/service/sts v1.41.6 // indirect
+	github.com/aws/smithy-go v1.24.0 // indirect
 	github.com/beorn7/perks v1.0.1 // indirect
-	github.com/cespare/xxhash/v2 v2.2.0 // indirect
-	github.com/containerd/containerd v1.6.18 // indirect
-	github.com/cpuguy83/go-md2man/v2 v2.0.2 // indirect
-	github.com/deepmap/oapi-codegen v1.12.4 // indirect
-	github.com/felixge/httpsnoop v1.0.3 // indirect
-	github.com/ghodss/yaml v1.0.0 // indirect
-	github.com/go-asn1-ber/asn1-ber v1.5.4 // indirect
-	github.com/go-openapi/jsonpointer v0.19.6 // indirect
-	github.com/go-openapi/jsonreference v0.20.2 // indirect
-	github.com/go-openapi/spec v0.20.8 // indirect
-	github.com/go-openapi/swag v0.22.3 // indirect
-	github.com/golang/protobuf v1.5.2 // indirect
-	github.com/google/uuid v1.3.0 // indirect
-	github.com/gorilla/securecookie v1.1.1 // indirect
-	github.com/gorilla/websocket v1.5.0 // indirect
-	github.com/hashicorp/errwrap v1.1.0 // indirect
-	github.com/hashicorp/go-multierror v1.1.1 // indirect
-	github.com/hashicorp/golang-lru v0.5.4 // indirect
-	github.com/influxdata/line-protocol v0.0.0-20210922203350-b1ad95c89adf // indirect
-	github.com/josharian/intern v1.0.0 // indirect
+	github.com/cespare/xxhash/v2 v2.3.0 // indirect
+	github.com/cpuguy83/go-md2man/v2 v2.0.7 // indirect
+	github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc // indirect
+	github.com/felixge/httpsnoop v1.0.4 // indirect
+	github.com/fsnotify/fsnotify v1.9.0 // indirect
+	github.com/go-asn1-ber/asn1-ber v1.5.8-0.20250403174932-29230038a667 // indirect
+	github.com/go-jose/go-jose/v4 v4.1.3 // indirect
+	github.com/go-openapi/jsonpointer v0.22.3 // indirect
+	github.com/go-openapi/jsonreference v0.21.3 // indirect
+	github.com/go-openapi/spec v0.22.1 // indirect
+	github.com/go-openapi/swag/conv v0.25.4 // indirect
+	github.com/go-openapi/swag/jsonname v0.25.4 // indirect
+	github.com/go-openapi/swag/jsonutils v0.25.4 // indirect
+	github.com/go-openapi/swag/loading v0.25.4 // indirect
+	github.com/go-openapi/swag/stringutils v0.25.4 // indirect
+	github.com/go-openapi/swag/typeutils v0.25.4 // indirect
+	github.com/go-openapi/swag/yamlutils v0.25.4 // indirect
+	github.com/go-viper/mapstructure/v2 v2.4.0 // indirect
+	github.com/goccy/go-yaml v1.19.0 // indirect
+	github.com/golang/snappy v0.0.4 // indirect
+	github.com/google/uuid v1.6.0 // indirect
+	github.com/gorilla/securecookie v1.1.2 // indirect
+	github.com/gorilla/websocket v1.5.3 // indirect
+	github.com/hashicorp/golang-lru/v2 v2.0.7 // indirect
+	github.com/jonboulle/clockwork v0.5.0 // indirect
 	github.com/jpillora/backoff v1.0.0 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
+	github.com/klauspost/compress v1.18.1 // indirect
 	github.com/lann/builder v0.0.0-20180802200727-47ae307949d0 // indirect
 	github.com/lann/ps v0.0.0-20150810152359-62de8c46ede0 // indirect
-	github.com/mailru/easyjson v0.7.7 // indirect
-	github.com/matttproud/golang_protobuf_extensions v1.0.4 // indirect
-	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
+	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
 	github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f // indirect
-	github.com/opencontainers/image-spec v1.0.3-0.20211202183452-c5a74bcca799 // indirect
-	github.com/pkg/errors v0.9.1 // indirect
-	github.com/prometheus/client_model v0.3.0 // indirect
-	github.com/prometheus/procfs v0.9.0 // indirect
+	github.com/nats-io/nkeys v0.4.11 // indirect
+	github.com/nats-io/nuid v1.0.1 // indirect
+	github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2 // indirect
+	github.com/prometheus/client_model v0.6.2 // indirect
+	github.com/prometheus/procfs v0.16.1 // indirect
 	github.com/robfig/cron/v3 v3.0.1 // indirect
 	github.com/russross/blackfriday/v2 v2.1.0 // indirect
-	github.com/swaggo/files v1.0.0 // indirect
-	github.com/urfave/cli/v2 v2.24.4 // indirect
-	github.com/xrash/smetrics v0.0.0-20201216005158-039620a65673 // indirect
-	go.uber.org/atomic v1.10.0 // indirect
-	golang.org/x/mod v0.8.0 // indirect
-	golang.org/x/net v0.7.0 // indirect
-	golang.org/x/oauth2 v0.5.0 // indirect
-	golang.org/x/sys v0.5.0 // indirect
-	golang.org/x/text v0.7.0 // indirect
-	golang.org/x/tools v0.6.0 // indirect
-	google.golang.org/appengine v1.6.7 // indirect
-	google.golang.org/protobuf v1.28.1 // indirect
-	gopkg.in/yaml.v2 v2.4.0 // indirect
+	github.com/sosodev/duration v1.3.1 // indirect
+	github.com/stretchr/objx v0.5.2 // indirect
+	github.com/swaggo/files v1.0.1 // indirect
+	github.com/urfave/cli/v2 v2.27.7 // indirect
+	github.com/urfave/cli/v3 v3.6.1 // indirect
+	github.com/xrash/smetrics v0.0.0-20250705151800-55b8f293f342 // indirect
+	go.yaml.in/yaml/v2 v2.4.3 // indirect
+	go.yaml.in/yaml/v3 v3.0.4 // indirect
+	golang.org/x/mod v0.31.0 // indirect
+	golang.org/x/net v0.48.0 // indirect
+	golang.org/x/sync v0.19.0 // indirect
+	golang.org/x/sys v0.39.0 // indirect
+	golang.org/x/text v0.32.0 // indirect
+	golang.org/x/tools v0.40.0 // indirect
+	google.golang.org/protobuf v1.36.11 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
+	sigs.k8s.io/yaml v1.6.0 // indirect
 )

gqlgen.yml

@@ -30,7 +30,9 @@ resolver:
 # gqlgen will search for any type names in the schema in these go packages
 # if they match it will use them, otherwise it will generate them.
 autobind:
+  - "github.com/99designs/gqlgen/graphql/introspection"
   - "github.com/ClusterCockpit/cc-backend/internal/graph/model"
+  - "github.com/ClusterCockpit/cc-backend/internal/config"
 
 # This section declares type mapping between the GraphQL and go type systems
 #
@@ -50,34 +52,51 @@ models:
       - github.com/99designs/gqlgen/graphql.Int64
       - github.com/99designs/gqlgen/graphql.Int32
   Job:
-    model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Job"
+    model: "github.com/ClusterCockpit/cc-lib/schema.Job"
     fields:
       tags:
         resolver: true
       metaData:
         resolver: true
   Cluster:
-    model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Cluster"
+    model: "github.com/ClusterCockpit/cc-lib/schema.Cluster"
     fields:
       partitions:
         resolver: true
-  NullableFloat: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Float" }
-  MetricScope: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricScope" }
-  MetricValue: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricValue" }
-  JobStatistics: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobStatistics" }
-  Tag: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Tag" }
-  Resource: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Resource" }
-  JobState: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobState" }
-  TimeRange: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.TimeRange" }
-  IntRange: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.IntRange" }
-  JobMetric: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.JobMetric" }
-  Series: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Series" }
-  MetricStatistics: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricStatistics" }
-  MetricConfig: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.MetricConfig" }
-  SubClusterConfig: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.SubClusterConfig" }
-  Accelerator: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Accelerator" }
-  Topology: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Topology" }
-  FilterRanges: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.FilterRanges" }
-  SubCluster: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.SubCluster" }
-  StatsSeries: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.StatsSeries" }
-  Unit: { model: "github.com/ClusterCockpit/cc-backend/pkg/schema.Unit" }
+  # Node:
+  #   model: "github.com/ClusterCockpit/cc-lib/schema.Node"
+  #   fields:
+  #     metaData:
+  #       resolver: true
+  NullableFloat: { model: "github.com/ClusterCockpit/cc-lib/schema.Float" }
+  MetricScope: { model: "github.com/ClusterCockpit/cc-lib/schema.MetricScope" }
+  MetricValue: { model: "github.com/ClusterCockpit/cc-lib/schema.MetricValue" }
+  JobStatistics:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.JobStatistics" }
+  GlobalMetricListItem:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.GlobalMetricListItem" }
+  ClusterSupport:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.ClusterSupport" }
+  Tag: { model: "github.com/ClusterCockpit/cc-lib/schema.Tag" }
+  Resource: { model: "github.com/ClusterCockpit/cc-lib/schema.Resource" }
+  JobState: { model: "github.com/ClusterCockpit/cc-lib/schema.JobState" }
+  Node: { model: "github.com/ClusterCockpit/cc-lib/schema.Node" }
+  SchedulerState:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.SchedulerState" }
+  HealthState:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.MonitoringState" }
+  JobMetric: { model: "github.com/ClusterCockpit/cc-lib/schema.JobMetric" }
+  Series: { model: "github.com/ClusterCockpit/cc-lib/schema.Series" }
+  MetricStatistics:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.MetricStatistics" }
+  MetricConfig:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.MetricConfig" }
+  SubClusterConfig:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.SubClusterConfig" }
+  Accelerator: { model: "github.com/ClusterCockpit/cc-lib/schema.Accelerator" }
+  Topology: { model: "github.com/ClusterCockpit/cc-lib/schema.Topology" }
+  FilterRanges:
+    { model: "github.com/ClusterCockpit/cc-lib/schema.FilterRanges" }
+  SubCluster: { model: "github.com/ClusterCockpit/cc-lib/schema.SubCluster" }
+  StatsSeries: { model: "github.com/ClusterCockpit/cc-lib/schema.StatsSeries" }
+  Unit: { model: "github.com/ClusterCockpit/cc-lib/schema.Unit" }

init/clustercockpit.service

@@ -1,5 +1,5 @@
 [Unit]
-Description=ClusterCockpit Web Server (Go edition)
+Description=ClusterCockpit Web Server
 Documentation=https://github.com/ClusterCockpit/cc-backend
 Wants=network-online.target
 After=network-online.target

internal/api/api_test.go

@@ -1,5 +1,5 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
 package api_test
@@ -14,31 +14,45 @@ import (
 	"os"
 	"path/filepath"
 	"reflect"
-	"strconv"
 	"strings"
 	"testing"
+	"time"
 
 	"github.com/ClusterCockpit/cc-backend/internal/api"
+	"github.com/ClusterCockpit/cc-backend/internal/archiver"
+	"github.com/ClusterCockpit/cc-backend/internal/auth"
 	"github.com/ClusterCockpit/cc-backend/internal/config"
 	"github.com/ClusterCockpit/cc-backend/internal/graph"
+	"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
 	"github.com/ClusterCockpit/cc-backend/internal/metricdata"
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
 	"github.com/ClusterCockpit/cc-backend/pkg/archive"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 	"github.com/gorilla/mux"
 
 	_ "github.com/mattn/go-sqlite3"
 )
 
-func setup(t *testing.T) *api.RestApi {
+func setup(t *testing.T) *api.RestAPI {
 	const testconfig = `{
+		"main": {
 	"addr":            "0.0.0.0:8080",
 	"validate": false,
+  "apiAllowedIPs": [
+    "*"
+  ]
+	},
 	"archive": {
 		"kind": "file",
 		"path": "./var/job-archive"
 	},
+	"auth": {
+  "jwts": {
+      "max-age": "2m"
+  }
+	},
 	"clusters": [
 	{
 	   "name": "testcluster",
@@ -51,7 +65,7 @@ func setup(t *testing.T) *api.RestApi {
 	}
 	]
 }`
-	const testclusterJson = `{
+	const testclusterJSON = `{
         "name": "testcluster",
 		"subClusters": [
 			{
@@ -107,22 +121,22 @@ func setup(t *testing.T) *api.RestApi {
 		]
 	}`
 
-	log.Init("info", true)
+	cclog.Init("info", true)
 	tmpdir := t.TempDir()
 	jobarchive := filepath.Join(tmpdir, "job-archive")
-	if err := os.Mkdir(jobarchive, 0777); err != nil {
+	if err := os.Mkdir(jobarchive, 0o777); err != nil {
 		t.Fatal(err)
 	}
 
-	if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), []byte(fmt.Sprintf("%d", 1)), 0666); err != nil {
+	if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), fmt.Appendf(nil, "%d", 3), 0o666); err != nil {
 		t.Fatal(err)
 	}
 
-	if err := os.Mkdir(filepath.Join(jobarchive, "testcluster"), 0777); err != nil {
+	if err := os.Mkdir(filepath.Join(jobarchive, "testcluster"), 0o777); err != nil {
 		t.Fatal(err)
 	}
 
-	if err := os.WriteFile(filepath.Join(jobarchive, "testcluster", "cluster.json"), []byte(testclusterJson), 0666); err != nil {
+	if err := os.WriteFile(filepath.Join(jobarchive, "testcluster", "cluster.json"), []byte(testclusterJSON), 0o666); err != nil {
 		t.Fatal(err)
 	}
 
@@ -133,34 +147,53 @@ func setup(t *testing.T) *api.RestApi {
 	}
 
 	cfgFilePath := filepath.Join(tmpdir, "config.json")
-	if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0666); err != nil {
+	if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0o666); err != nil {
 		t.Fatal(err)
 	}
 
-	config.Init(cfgFilePath)
+	ccconf.Init(cfgFilePath)
+
+	// Load and check main configuration
+	if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
+		if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
+			config.Init(cfg, clustercfg)
+		} else {
+			cclog.Abort("Cluster configuration must be present")
+		}
+	} else {
+		cclog.Abort("Main configuration must be present")
+	}
 	archiveCfg := fmt.Sprintf("{\"kind\": \"file\",\"path\": \"%s\"}", jobarchive)
 
 	repository.Connect("sqlite3", dbfilepath)
-	db := repository.GetConnection()
 
 	if err := archive.Init(json.RawMessage(archiveCfg), config.Keys.DisableArchive); err != nil {
 		t.Fatal(err)
 	}
 
-	if err := metricdata.Init(config.Keys.DisableArchive); err != nil {
+	if err := metricdata.Init(); err != nil {
 		t.Fatal(err)
 	}
 
-	jobRepo := repository.GetJobRepository()
-	resolver := &graph.Resolver{DB: db.DB, Repo: jobRepo}
+	archiver.Start(repository.GetJobRepository(), context.Background())
 
-	return &api.RestApi{
-		JobRepository: resolver.Repo,
-		Resolver:      resolver,
+	if cfg := ccconf.GetPackageConfig("auth"); cfg != nil {
+		auth.Init(&cfg)
+	} else {
+		cclog.Warn("Authentication disabled due to missing configuration")
+		auth.Init(nil)
 	}
+
+	graph.Init()
+
+	return api.New()
 }
 
 func cleanup() {
+	// Gracefully shutdown archiver with timeout
+	if err := archiver.Shutdown(5 * time.Second); err != nil {
+		cclog.Warnf("Archiver shutdown timeout in tests: %v", err)
+	}
 	// TODO: Clear all caches, reset all modules, etc...
 }
 
@@ -172,7 +205,6 @@ func cleanup() {
 func TestRestApi(t *testing.T) {
 	restapi := setup(t)
 	t.Cleanup(cleanup)
-
 	testData := schema.JobData{
 		"load_one": map[schema.MetricScope]*schema.JobMetric{
 			schema.MetricScopeNode: {
@@ -189,15 +221,21 @@ func TestRestApi(t *testing.T) {
 		},
 	}
 
-	metricdata.TestLoadDataCallback = func(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context) (schema.JobData, error) {
+	metricdata.TestLoadDataCallback = func(job *schema.Job, metrics []string, scopes []schema.MetricScope, ctx context.Context, resolution int) (schema.JobData, error) {
 		return testData, nil
 	}
 
 	r := mux.NewRouter()
-	restapi.MountRoutes(r)
+	r.PathPrefix("/api").Subrouter()
+	r.StrictSlash(true)
+	restapi.MountAPIRoutes(r)
+
+	var TestJobId int64 = 123
+	TestClusterName := "testcluster"
+	var TestStartTime int64 = 123456789
 
 	const startJobBody string = `{
-        "jobId":            123,
+    "jobId":            123,
 		"user":             "testuser",
 		"project":          "testproj",
 		"cluster":          "testcluster",
@@ -207,10 +245,9 @@ func TestRestApi(t *testing.T) {
 		"numNodes":         1,
 		"numHwthreads":     8,
 		"numAcc":           0,
-		"exclusive":        1,
+		"shared":           "none",
 		"monitoringStatus": 1,
 		"smt":              1,
-		"tags":             [{ "type": "testTagType", "name": "testTagName" }],
 		"resources": [
 			{
 				"hostname": "host123",
@@ -221,31 +258,37 @@ func TestRestApi(t *testing.T) {
 		"startTime": 123456789
 	}`
 
-	var dbid int64
+	const contextUserKey repository.ContextKey = "user"
+	contextUserValue := &schema.User{
+		Username:   "testuser",
+		Projects:   make([]string, 0),
+		Roles:      []string{"user"},
+		AuthType:   0,
+		AuthSource: 2,
+	}
+
 	if ok := t.Run("StartJob", func(t *testing.T) {
-		req := httptest.NewRequest(http.MethodPost, "/api/jobs/start_job/", bytes.NewBuffer([]byte(startJobBody)))
+		req := httptest.NewRequest(http.MethodPost, "/jobs/start_job/", bytes.NewBuffer([]byte(startJobBody)))
 		recorder := httptest.NewRecorder()
 
-		r.ServeHTTP(recorder, req)
+		ctx := context.WithValue(req.Context(), contextUserKey, contextUserValue)
+
+		r.ServeHTTP(recorder, req.WithContext(ctx))
 		response := recorder.Result()
 		if response.StatusCode != http.StatusCreated {
 			t.Fatal(response.Status, recorder.Body.String())
 		}
-
-		var res api.StartJobApiResponse
-		if err := json.Unmarshal(recorder.Body.Bytes(), &res); err != nil {
-			t.Fatal(err)
-		}
-
-		job, err := restapi.Resolver.Query().Job(context.Background(), strconv.Itoa(int(res.DBID)))
+		// resolver := graph.GetResolverInstance()
+		restapi.JobRepository.SyncJobs()
+		job, err := restapi.JobRepository.Find(&TestJobId, &TestClusterName, &TestStartTime)
 		if err != nil {
 			t.Fatal(err)
 		}
 
-		job.Tags, err = restapi.Resolver.Job().Tags(context.Background(), job)
-		if err != nil {
-			t.Fatal(err)
-		}
+		// job.Tags, err = resolver.Job().Tags(ctx, job)
+		// if err != nil {
+		// 	t.Fatal(err)
+		// }
 
 		if job.JobID != 123 ||
 			job.User != "testuser" ||
@@ -254,23 +297,20 @@ func TestRestApi(t *testing.T) {
 			job.SubCluster != "sc1" ||
 			job.Partition != "default" ||
 			job.Walltime != 3600 ||
-			job.ArrayJobId != 0 ||
+			job.ArrayJobID != 0 ||
 			job.NumNodes != 1 ||
 			job.NumHWThreads != 8 ||
 			job.NumAcc != 0 ||
-			job.Exclusive != 1 ||
 			job.MonitoringStatus != 1 ||
 			job.SMT != 1 ||
 			!reflect.DeepEqual(job.Resources, []*schema.Resource{{Hostname: "host123", HWThreads: []int{0, 1, 2, 3, 4, 5, 6, 7}}}) ||
-			job.StartTime.Unix() != 123456789 {
+			job.StartTime != 123456789 {
 			t.Fatalf("unexpected job properties: %#v", job)
 		}
 
-		if len(job.Tags) != 1 || job.Tags[0].Type != "testTagType" || job.Tags[0].Name != "testTagName" {
-			t.Fatalf("unexpected tags: %#v", job.Tags)
-		}
-
-		dbid = res.DBID
+		// if len(job.Tags) != 1 || job.Tags[0].Type != "testTagType" || job.Tags[0].Name != "testTagName" || job.Tags[0].Scope != "testuser" {
+		// 	t.Fatalf("unexpected tags: %#v", job.Tags)
+		// }
 	}); !ok {
 		return
 	}
@@ -286,17 +326,19 @@ func TestRestApi(t *testing.T) {
 
 	var stoppedJob *schema.Job
 	if ok := t.Run("StopJob", func(t *testing.T) {
-		req := httptest.NewRequest(http.MethodPost, "/api/jobs/stop_job/", bytes.NewBuffer([]byte(stopJobBody)))
+		req := httptest.NewRequest(http.MethodPost, "/jobs/stop_job/", bytes.NewBuffer([]byte(stopJobBody)))
 		recorder := httptest.NewRecorder()
 
-		r.ServeHTTP(recorder, req)
+		ctx := context.WithValue(req.Context(), contextUserKey, contextUserValue)
+
+		r.ServeHTTP(recorder, req.WithContext(ctx))
 		response := recorder.Result()
 		if response.StatusCode != http.StatusOK {
 			t.Fatal(response.Status, recorder.Body.String())
 		}
 
-		restapi.JobRepository.WaitForArchiving()
-		job, err := restapi.Resolver.Query().Job(context.Background(), strconv.Itoa(int(dbid)))
+		// Archiving happens asynchronously, will be completed in cleanup
+		job, err := restapi.JobRepository.Find(&TestJobId, &TestClusterName, &TestStartTime)
 		if err != nil {
 			t.Fatal(err)
 		}
@@ -324,7 +366,7 @@ func TestRestApi(t *testing.T) {
 	}
 
 	t.Run("CheckArchive", func(t *testing.T) {
-		data, err := metricdata.LoadData(stoppedJob, []string{"load_one"}, []schema.MetricScope{schema.MetricScopeNode}, context.Background())
+		data, err := metricDataDispatcher.LoadData(stoppedJob, []string{"load_one"}, []schema.MetricScope{schema.MetricScopeNode}, context.Background(), 60)
 		if err != nil {
 			t.Fatal(err)
 		}
@@ -336,12 +378,14 @@ func TestRestApi(t *testing.T) {
 
 	t.Run("CheckDoubleStart", func(t *testing.T) {
 		// Starting a job with the same jobId and cluster should only be allowed if the startTime is far appart!
-		body := strings.Replace(startJobBody, `"startTime": 123456789`, `"startTime": 123456790`, -1)
+		body := strings.ReplaceAll(startJobBody, `"startTime": 123456789`, `"startTime": 123456790`)
 
-		req := httptest.NewRequest(http.MethodPost, "/api/jobs/start_job/", bytes.NewBuffer([]byte(body)))
+		req := httptest.NewRequest(http.MethodPost, "/jobs/start_job/", bytes.NewBuffer([]byte(body)))
 		recorder := httptest.NewRecorder()
 
-		r.ServeHTTP(recorder, req)
+		ctx := context.WithValue(req.Context(), contextUserKey, contextUserValue)
+
+		r.ServeHTTP(recorder, req.WithContext(ctx))
 		response := recorder.Result()
 		if response.StatusCode != http.StatusUnprocessableEntity {
 			t.Fatal(response.Status, recorder.Body.String())
@@ -356,7 +400,7 @@ func TestRestApi(t *testing.T) {
 		"partition":        "default",
 		"walltime":         3600,
 		"numNodes":         1,
-		"exclusive":        1,
+		"shared":        	"none",
 		"monitoringStatus": 1,
 		"smt":              1,
 		"resources": [
@@ -368,10 +412,12 @@ func TestRestApi(t *testing.T) {
 	}`
 
 	ok := t.Run("StartJobFailed", func(t *testing.T) {
-		req := httptest.NewRequest(http.MethodPost, "/api/jobs/start_job/", bytes.NewBuffer([]byte(startJobBodyFailed)))
+		req := httptest.NewRequest(http.MethodPost, "/jobs/start_job/", bytes.NewBuffer([]byte(startJobBodyFailed)))
 		recorder := httptest.NewRecorder()
 
-		r.ServeHTTP(recorder, req)
+		ctx := context.WithValue(req.Context(), contextUserKey, contextUserValue)
+
+		r.ServeHTTP(recorder, req.WithContext(ctx))
 		response := recorder.Result()
 		if response.StatusCode != http.StatusCreated {
 			t.Fatal(response.Status, recorder.Body.String())
@@ -381,8 +427,11 @@ func TestRestApi(t *testing.T) {
 		t.Fatal("subtest failed")
 	}
 
+	time.Sleep(1 * time.Second)
+	restapi.JobRepository.SyncJobs()
+
 	const stopJobBodyFailed string = `{
-        "jobId":     12345,
+    "jobId":     12345,
 		"cluster":   "testcluster",
 
 		"jobState": "failed",
@@ -390,16 +439,18 @@ func TestRestApi(t *testing.T) {
 	}`
 
 	ok = t.Run("StopJobFailed", func(t *testing.T) {
-		req := httptest.NewRequest(http.MethodPost, "/api/jobs/stop_job/", bytes.NewBuffer([]byte(stopJobBodyFailed)))
+		req := httptest.NewRequest(http.MethodPost, "/jobs/stop_job/", bytes.NewBuffer([]byte(stopJobBodyFailed)))
 		recorder := httptest.NewRecorder()
 
-		r.ServeHTTP(recorder, req)
+		ctx := context.WithValue(req.Context(), contextUserKey, contextUserValue)
+
+		r.ServeHTTP(recorder, req.WithContext(ctx))
 		response := recorder.Result()
 		if response.StatusCode != http.StatusOK {
 			t.Fatal(response.Status, recorder.Body.String())
 		}
 
-		restapi.JobRepository.WaitForArchiving()
+		// Archiving happens asynchronously, will be completed in cleanup
 		jobid, cluster := int64(12345), "testcluster"
 		job, err := restapi.JobRepository.Find(&jobid, &cluster, nil)
 		if err != nil {

internal/api/cluster.go

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

internal/api/memorystore.go

@@ -0,0 +1,170 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package api
+
+import (
+	"bufio"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+
+	"github.com/ClusterCockpit/cc-backend/internal/memorystore"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+
+	"github.com/influxdata/line-protocol/v2/lineprotocol"
+)
+
+// handleFree godoc
+// @summary
+// @tags free
+// @description This endpoint allows the users to free the Buffers from the
+// metric store. This endpoint offers the users to remove then systematically
+// and also allows then to prune the data under node, if they do not want to
+// remove the whole node.
+// @produce     json
+// @param       to        query    string        false  "up to timestamp"
+// @success     200            {string} string  "ok"
+// @failure     400            {object} api.ErrorResponse       "Bad Request"
+// @failure     401            {object} api.ErrorResponse       "Unauthorized"
+// @failure     403            {object} api.ErrorResponse       "Forbidden"
+// @failure     500            {object} api.ErrorResponse       "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /free/ [post]
+func freeMetrics(rw http.ResponseWriter, r *http.Request) {
+	rawTo := r.URL.Query().Get("to")
+	if rawTo == "" {
+		handleError(errors.New("'to' is a required query parameter"), http.StatusBadRequest, rw)
+		return
+	}
+
+	to, err := strconv.ParseInt(rawTo, 10, 64)
+	if err != nil {
+		handleError(err, http.StatusInternalServerError, rw)
+		return
+	}
+
+	bodyDec := json.NewDecoder(r.Body)
+	var selectors [][]string
+	err = bodyDec.Decode(&selectors)
+	if err != nil {
+		http.Error(rw, err.Error(), http.StatusBadRequest)
+		return
+	}
+
+	ms := memorystore.GetMemoryStore()
+	n := 0
+	for _, sel := range selectors {
+		bn, err := ms.Free(sel, to)
+		if err != nil {
+			handleError(err, http.StatusInternalServerError, rw)
+			return
+		}
+
+		n += bn
+	}
+
+	rw.WriteHeader(http.StatusOK)
+	fmt.Fprintf(rw, "buffers freed: %d\n", n)
+}
+
+// handleWrite godoc
+// @summary Receive metrics in InfluxDB line-protocol
+// @tags write
+// @description Write data to the in-memory store in the InfluxDB line-protocol using [this format](https://github.com/ClusterCockpit/cc-specifications/blob/master/metrics/lineprotocol_alternative.md)
+
+// @accept      plain
+// @produce     json
+// @param       cluster        query string false "If the lines in the body do not have a cluster tag, use this value instead."
+// @success     200            {string} string  "ok"
+// @failure     400            {object} api.ErrorResponse       "Bad Request"
+// @failure     401            {object} api.ErrorResponse       "Unauthorized"
+// @failure     403            {object} api.ErrorResponse       "Forbidden"
+// @failure     500            {object} api.ErrorResponse       "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /write/ [post]
+func writeMetrics(rw http.ResponseWriter, r *http.Request) {
+	bytes, err := io.ReadAll(r.Body)
+	rw.Header().Add("Content-Type", "application/json")
+	if err != nil {
+		handleError(err, http.StatusInternalServerError, rw)
+		return
+	}
+
+	ms := memorystore.GetMemoryStore()
+	dec := lineprotocol.NewDecoderWithBytes(bytes)
+	if err := memorystore.DecodeLine(dec, ms, r.URL.Query().Get("cluster")); err != nil {
+		cclog.Errorf("/api/write error: %s", err.Error())
+		handleError(err, http.StatusBadRequest, rw)
+		return
+	}
+	rw.WriteHeader(http.StatusOK)
+}
+
+// handleDebug godoc
+// @summary Debug endpoint
+// @tags debug
+// @description This endpoint allows the users to print the content of
+// nodes/clusters/metrics to review the state of the data.
+// @produce     json
+// @param       selector        query    string            false "Selector"
+// @success     200            {string} string  "Debug dump"
+// @failure     400            {object} api.ErrorResponse       "Bad Request"
+// @failure     401            {object} api.ErrorResponse       "Unauthorized"
+// @failure     403            {object} api.ErrorResponse       "Forbidden"
+// @failure     500            {object} api.ErrorResponse       "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /debug/ [post]
+func debugMetrics(rw http.ResponseWriter, r *http.Request) {
+	raw := r.URL.Query().Get("selector")
+	rw.Header().Add("Content-Type", "application/json")
+	selector := []string{}
+	if len(raw) != 0 {
+		selector = strings.Split(raw, ":")
+	}
+
+	ms := memorystore.GetMemoryStore()
+	if err := ms.DebugDump(bufio.NewWriter(rw), selector); err != nil {
+		handleError(err, http.StatusBadRequest, rw)
+		return
+	}
+}
+
+// handleHealthCheck godoc
+// @summary HealthCheck endpoint
+// @tags healthcheck
+// @description This endpoint allows the users to check if a node is healthy
+// @produce     json
+// @param       selector        query    string            false "Selector"
+// @success     200            {string} string  "Debug dump"
+// @failure     400            {object} api.ErrorResponse       "Bad Request"
+// @failure     401            {object} api.ErrorResponse       "Unauthorized"
+// @failure     403            {object} api.ErrorResponse       "Forbidden"
+// @failure     500            {object} api.ErrorResponse       "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /healthcheck/ [get]
+func metricsHealth(rw http.ResponseWriter, r *http.Request) {
+	rawCluster := r.URL.Query().Get("cluster")
+	rawNode := r.URL.Query().Get("node")
+
+	if rawCluster == "" || rawNode == "" {
+		handleError(errors.New("'cluster' and 'node' are required query parameter"), http.StatusBadRequest, rw)
+		return
+	}
+
+	rw.Header().Add("Content-Type", "application/json")
+
+	selector := []string{rawCluster, rawNode}
+
+	ms := memorystore.GetMemoryStore()
+	if err := ms.HealthCheck(bufio.NewWriter(rw), selector); err != nil {
+		handleError(err, http.StatusBadRequest, rw)
+		return
+	}
+}

internal/api/nats.go

@@ -0,0 +1,231 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package api
+
+import (
+	"bytes"
+	"database/sql"
+	"encoding/json"
+	"sync"
+	"time"
+
+	"github.com/ClusterCockpit/cc-backend/internal/archiver"
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-backend/internal/importer"
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	"github.com/ClusterCockpit/cc-backend/pkg/nats"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+)
+
+// NatsAPI provides NATS subscription-based handlers for Job and Node operations.
+// It mirrors the functionality of the REST API but uses NATS messaging.
+type NatsAPI struct {
+	JobRepository *repository.JobRepository
+	// RepositoryMutex protects job creation operations from race conditions
+	// when checking for duplicate jobs during startJob calls.
+	RepositoryMutex sync.Mutex
+}
+
+// NewNatsAPI creates a new NatsAPI instance with default dependencies.
+func NewNatsAPI() *NatsAPI {
+	return &NatsAPI{
+		JobRepository: repository.GetJobRepository(),
+	}
+}
+
+// StartSubscriptions registers all NATS subscriptions for Job and Node APIs.
+// Returns an error if the NATS client is not available or subscription fails.
+func (api *NatsAPI) StartSubscriptions() error {
+	client := nats.GetClient()
+	if client == nil {
+		cclog.Warn("NATS client not available, skipping API subscriptions")
+		return nil
+	}
+
+	if config.Keys.APISubjects != nil {
+
+		s := config.Keys.APISubjects
+
+		if err := client.Subscribe(s.SubjectJobStart, api.handleStartJob); err != nil {
+			return err
+		}
+
+		if err := client.Subscribe(s.SubjectJobStop, api.handleStopJob); err != nil {
+			return err
+		}
+
+		if err := client.Subscribe(s.SubjectNodeState, api.handleNodeState); err != nil {
+			return err
+		}
+
+		cclog.Info("NATS API subscriptions started")
+	}
+	return nil
+}
+
+// handleStartJob processes job start messages received via NATS.
+// Expected JSON payload follows the schema.Job structure.
+func (api *NatsAPI) handleStartJob(subject string, data []byte) {
+	req := schema.Job{
+		Shared:           "none",
+		MonitoringStatus: schema.MonitoringStatusRunningOrArchiving,
+	}
+
+	dec := json.NewDecoder(bytes.NewReader(data))
+	dec.DisallowUnknownFields()
+	if err := dec.Decode(&req); err != nil {
+		cclog.Errorf("NATS %s: parsing request failed: %v", subject, err)
+		return
+	}
+
+	cclog.Debugf("NATS %s: %s", subject, req.GoString())
+	req.State = schema.JobStateRunning
+
+	if err := importer.SanityChecks(&req); err != nil {
+		cclog.Errorf("NATS %s: sanity check failed: %v", subject, err)
+		return
+	}
+
+	var unlockOnce sync.Once
+	api.RepositoryMutex.Lock()
+	defer unlockOnce.Do(api.RepositoryMutex.Unlock)
+
+	jobs, err := api.JobRepository.FindAll(&req.JobID, &req.Cluster, nil)
+	if err != nil && err != sql.ErrNoRows {
+		cclog.Errorf("NATS %s: checking for duplicate failed: %v", subject, err)
+		return
+	}
+	if err == nil {
+		for _, job := range jobs {
+			if (req.StartTime - job.StartTime) < secondsPerDay {
+				cclog.Errorf("NATS %s: job with jobId %d, cluster %s already exists (dbid: %d)",
+					subject, req.JobID, req.Cluster, job.ID)
+				return
+			}
+		}
+	}
+
+	id, err := api.JobRepository.Start(&req)
+	if err != nil {
+		cclog.Errorf("NATS %s: insert into database failed: %v", subject, err)
+		return
+	}
+	unlockOnce.Do(api.RepositoryMutex.Unlock)
+
+	for _, tag := range req.Tags {
+		if _, err := api.JobRepository.AddTagOrCreate(nil, id, tag.Type, tag.Name, tag.Scope); err != nil {
+			cclog.Errorf("NATS %s: adding tag to new job %d failed: %v", subject, id, err)
+			return
+		}
+	}
+
+	cclog.Infof("NATS: new job (id: %d): cluster=%s, jobId=%d, user=%s, startTime=%d",
+		id, req.Cluster, req.JobID, req.User, req.StartTime)
+}
+
+// handleStopJob processes job stop messages received via NATS.
+// Expected JSON payload follows the StopJobAPIRequest structure.
+func (api *NatsAPI) handleStopJob(subject string, data []byte) {
+	var req StopJobAPIRequest
+
+	dec := json.NewDecoder(bytes.NewReader(data))
+	dec.DisallowUnknownFields()
+	if err := dec.Decode(&req); err != nil {
+		cclog.Errorf("NATS %s: parsing request failed: %v", subject, err)
+		return
+	}
+
+	if req.JobID == nil {
+		cclog.Errorf("NATS %s: the field 'jobId' is required", subject)
+		return
+	}
+
+	job, err := api.JobRepository.Find(req.JobID, req.Cluster, req.StartTime)
+	if err != nil {
+		cachedJob, cachedErr := api.JobRepository.FindCached(req.JobID, req.Cluster, req.StartTime)
+		if cachedErr != nil {
+			cclog.Errorf("NATS %s: finding job failed: %v (cached lookup also failed: %v)",
+				subject, err, cachedErr)
+			return
+		}
+		job = cachedJob
+	}
+
+	if job.State != schema.JobStateRunning {
+		cclog.Errorf("NATS %s: jobId %d (id %d) on %s: job has already been stopped (state is: %s)",
+			subject, job.JobID, job.ID, job.Cluster, job.State)
+		return
+	}
+
+	if job.StartTime > req.StopTime {
+		cclog.Errorf("NATS %s: jobId %d (id %d) on %s: stopTime %d must be >= startTime %d",
+			subject, job.JobID, job.ID, job.Cluster, req.StopTime, job.StartTime)
+		return
+	}
+
+	if req.State != "" && !req.State.Valid() {
+		cclog.Errorf("NATS %s: jobId %d (id %d) on %s: invalid job state: %#v",
+			subject, job.JobID, job.ID, job.Cluster, req.State)
+		return
+	} else if req.State == "" {
+		req.State = schema.JobStateCompleted
+	}
+
+	job.Duration = int32(req.StopTime - job.StartTime)
+	job.State = req.State
+	api.JobRepository.Mutex.Lock()
+	defer api.JobRepository.Mutex.Unlock()
+
+	if err := api.JobRepository.Stop(*job.ID, job.Duration, job.State, job.MonitoringStatus); err != nil {
+		if err := api.JobRepository.StopCached(*job.ID, job.Duration, job.State, job.MonitoringStatus); err != nil {
+			cclog.Errorf("NATS %s: jobId %d (id %d) on %s: marking job as '%s' failed: %v",
+				subject, job.JobID, job.ID, job.Cluster, job.State, err)
+			return
+		}
+	}
+
+	cclog.Infof("NATS: archiving job (dbid: %d): cluster=%s, jobId=%d, user=%s, startTime=%d, duration=%d, state=%s",
+		job.ID, job.Cluster, job.JobID, job.User, job.StartTime, job.Duration, job.State)
+
+	if job.MonitoringStatus == schema.MonitoringStatusDisabled {
+		return
+	}
+
+	archiver.TriggerArchiving(job)
+}
+
+// handleNodeState processes node state update messages received via NATS.
+// Expected JSON payload follows the UpdateNodeStatesRequest structure.
+func (api *NatsAPI) handleNodeState(subject string, data []byte) {
+	var req UpdateNodeStatesRequest
+
+	dec := json.NewDecoder(bytes.NewReader(data))
+	dec.DisallowUnknownFields()
+	if err := dec.Decode(&req); err != nil {
+		cclog.Errorf("NATS %s: parsing request failed: %v", subject, err)
+		return
+	}
+
+	repo := repository.GetNodeRepository()
+
+	for _, node := range req.Nodes {
+		state := determineState(node.States)
+		nodeState := schema.NodeStateDB{
+			TimeStamp:       time.Now().Unix(),
+			NodeState:       state,
+			CpusAllocated:   node.CpusAllocated,
+			MemoryAllocated: node.MemoryAllocated,
+			GpusAllocated:   node.GpusAllocated,
+			HealthState:     schema.MonitoringStateFull,
+			JobsRunning:     node.JobsRunning,
+		}
+
+		repo.UpdateNodeState(node.Hostname, req.Cluster, &nodeState)
+	}
+
+	cclog.Debugf("NATS %s: updated %d node states for cluster %s", subject, len(req.Nodes), req.Cluster)
+}

internal/api/node.go

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

internal/api/user.go

@@ -0,0 +1,221 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package api
+
+import (
+	"encoding/json"
+	"fmt"
+	"net/http"
+
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/gorilla/mux"
+)
+
+type APIReturnedUser struct {
+	Username string   `json:"username"`
+	Name     string   `json:"name"`
+	Roles    []string `json:"roles"`
+	Email    string   `json:"email"`
+	Projects []string `json:"projects"`
+}
+
+// getUsers godoc
+// @summary     Returns a list of users
+// @tags User
+// @description Returns a JSON-encoded list of users.
+// @description Required query-parameter defines if all users or only users with additional special roles are returned.
+// @produce     json
+// @param       not-just-user query bool true "If returned list should contain all users or only users with additional special roles"
+// @success     200     {array} api.ApiReturnedUser "List of users returned successfully"
+// @failure     400     {string} string             "Bad Request"
+// @failure     401     {string} string             "Unauthorized"
+// @failure     403     {string} string             "Forbidden"
+// @failure     500     {string} string             "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /api/users/ [get]
+func (api *RestAPI) getUsers(rw http.ResponseWriter, r *http.Request) {
+	// SecuredCheck() only worked with TokenAuth: Removed
+
+	if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
+		handleError(fmt.Errorf("only admins are allowed to fetch a list of users"), http.StatusForbidden, rw)
+		return
+	}
+
+	users, err := repository.GetUserRepository().ListUsers(r.URL.Query().Get("not-just-user") == "true")
+	if err != nil {
+		handleError(fmt.Errorf("listing users failed: %w", err), http.StatusInternalServerError, rw)
+		return
+	}
+
+	rw.Header().Set("Content-Type", "application/json")
+	if err := json.NewEncoder(rw).Encode(users); err != nil {
+		cclog.Errorf("Failed to encode users response: %v", err)
+	}
+}
+
+// updateUser godoc
+// @summary     Update user roles and projects
+// @tags User
+// @description Allows admins to add/remove roles and projects for a user
+// @produce     plain
+// @param       id          path   string  true  "Username"
+// @param       add-role    formData string false "Role to add"
+// @param       remove-role formData string false "Role to remove"
+// @param       add-project formData string false "Project to add"
+// @param       remove-project formData string false "Project to remove"
+// @success     200     {string} string "Success message"
+// @failure     403     {object} api.ErrorResponse "Forbidden"
+// @failure     422     {object} api.ErrorResponse "Unprocessable Entity"
+// @security    ApiKeyAuth
+// @router      /api/user/{id} [post]
+func (api *RestAPI) updateUser(rw http.ResponseWriter, r *http.Request) {
+	// SecuredCheck() only worked with TokenAuth: Removed
+
+	if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
+		handleError(fmt.Errorf("only admins are allowed to update a user"), http.StatusForbidden, rw)
+		return
+	}
+
+	// Get Values
+	newrole := r.FormValue("add-role")
+	delrole := r.FormValue("remove-role")
+	newproj := r.FormValue("add-project")
+	delproj := r.FormValue("remove-project")
+
+	rw.Header().Set("Content-Type", "application/json")
+
+	// Handle role updates
+	if newrole != "" {
+		if err := repository.GetUserRepository().AddRole(r.Context(), mux.Vars(r)["id"], newrole); err != nil {
+			handleError(fmt.Errorf("adding role failed: %w", err), http.StatusUnprocessableEntity, rw)
+			return
+		}
+		if err := json.NewEncoder(rw).Encode(DefaultAPIResponse{Message: "Add Role Success"}); err != nil {
+			cclog.Errorf("Failed to encode response: %v", err)
+		}
+	} else if delrole != "" {
+		if err := repository.GetUserRepository().RemoveRole(r.Context(), mux.Vars(r)["id"], delrole); err != nil {
+			handleError(fmt.Errorf("removing role failed: %w", err), http.StatusUnprocessableEntity, rw)
+			return
+		}
+		if err := json.NewEncoder(rw).Encode(DefaultAPIResponse{Message: "Remove Role Success"}); err != nil {
+			cclog.Errorf("Failed to encode response: %v", err)
+		}
+	} else if newproj != "" {
+		if err := repository.GetUserRepository().AddProject(r.Context(), mux.Vars(r)["id"], newproj); err != nil {
+			handleError(fmt.Errorf("adding project failed: %w", err), http.StatusUnprocessableEntity, rw)
+			return
+		}
+		if err := json.NewEncoder(rw).Encode(DefaultAPIResponse{Message: "Add Project Success"}); err != nil {
+			cclog.Errorf("Failed to encode response: %v", err)
+		}
+	} else if delproj != "" {
+		if err := repository.GetUserRepository().RemoveProject(r.Context(), mux.Vars(r)["id"], delproj); err != nil {
+			handleError(fmt.Errorf("removing project failed: %w", err), http.StatusUnprocessableEntity, rw)
+			return
+		}
+		if err := json.NewEncoder(rw).Encode(DefaultAPIResponse{Message: "Remove Project Success"}); err != nil {
+			cclog.Errorf("Failed to encode response: %v", err)
+		}
+	} else {
+		handleError(fmt.Errorf("no operation specified: must provide add-role, remove-role, add-project, or remove-project"), http.StatusBadRequest, rw)
+	}
+}
+
+// createUser godoc
+// @summary     Create a new user
+// @tags User
+// @description Creates a new user with specified credentials and role
+// @produce     plain
+// @param       username formData string true  "Username"
+// @param       password formData string false "Password (not required for API users)"
+// @param       role     formData string true  "User role"
+// @param       name     formData string false "Full name"
+// @param       email    formData string false "Email address"
+// @param       project  formData string false "Project (required for managers)"
+// @success     200     {string} string "Success message"
+// @failure     400     {object} api.ErrorResponse "Bad Request"
+// @failure     403     {object} api.ErrorResponse "Forbidden"
+// @failure     422     {object} api.ErrorResponse "Unprocessable Entity"
+// @security    ApiKeyAuth
+// @router      /api/users/ [post]
+func (api *RestAPI) createUser(rw http.ResponseWriter, r *http.Request) {
+	// SecuredCheck() only worked with TokenAuth: Removed
+
+	rw.Header().Set("Content-Type", "text/plain")
+	me := repository.GetUserFromContext(r.Context())
+	if !me.HasRole(schema.RoleAdmin) {
+		handleError(fmt.Errorf("only admins are allowed to create new users"), http.StatusForbidden, rw)
+		return
+	}
+
+	username, password, role, name, email, project := r.FormValue("username"),
+		r.FormValue("password"), r.FormValue("role"), r.FormValue("name"),
+		r.FormValue("email"), r.FormValue("project")
+
+	// Validate username length
+	if len(username) == 0 || len(username) > 100 {
+		handleError(fmt.Errorf("username must be between 1 and 100 characters"), http.StatusBadRequest, rw)
+		return
+	}
+
+	if len(password) == 0 && role != schema.GetRoleString(schema.RoleApi) {
+		handleError(fmt.Errorf("only API users are allowed to have a blank password (login will be impossible)"), http.StatusBadRequest, rw)
+		return
+	}
+
+	if len(project) != 0 && role != schema.GetRoleString(schema.RoleManager) {
+		handleError(fmt.Errorf("only managers require a project (can be changed later)"), http.StatusBadRequest, rw)
+		return
+	} else if len(project) == 0 && role == schema.GetRoleString(schema.RoleManager) {
+		handleError(fmt.Errorf("managers require a project to manage (can be changed later)"), http.StatusBadRequest, rw)
+		return
+	}
+
+	if err := repository.GetUserRepository().AddUser(&schema.User{
+		Username: username,
+		Name:     name,
+		Password: password,
+		Email:    email,
+		Projects: []string{project},
+		Roles:    []string{role},
+	}); err != nil {
+		handleError(fmt.Errorf("adding user failed: %w", err), http.StatusUnprocessableEntity, rw)
+		return
+	}
+
+	fmt.Fprintf(rw, "User %v successfully created!\n", username)
+}
+
+// deleteUser godoc
+// @summary     Delete a user
+// @tags User
+// @description Deletes a user from the system
+// @produce     plain
+// @param       username formData string true "Username to delete"
+// @success     200     {string} string "Success"
+// @failure     403     {object} api.ErrorResponse "Forbidden"
+// @failure     422     {object} api.ErrorResponse "Unprocessable Entity"
+// @security    ApiKeyAuth
+// @router      /api/users/ [delete]
+func (api *RestAPI) deleteUser(rw http.ResponseWriter, r *http.Request) {
+	// SecuredCheck() only worked with TokenAuth: Removed
+
+	if user := repository.GetUserFromContext(r.Context()); !user.HasRole(schema.RoleAdmin) {
+		handleError(fmt.Errorf("only admins are allowed to delete a user"), http.StatusForbidden, rw)
+		return
+	}
+
+	username := r.FormValue("username")
+	if err := repository.GetUserRepository().DelUser(username); err != nil {
+		handleError(fmt.Errorf("deleting user failed: %w", err), http.StatusUnprocessableEntity, rw)
+		return
+	}
+
+	rw.WriteHeader(http.StatusOK)
+}

internal/archiver/README.md

@@ -0,0 +1,190 @@
+# Archiver Package
+
+The `archiver` package provides asynchronous job archiving functionality for ClusterCockpit. When jobs complete, their metric data is archived from the metric store to a persistent archive backend (filesystem, S3, SQLite, etc.).
+
+## Architecture
+
+### Producer-Consumer Pattern
+
+```
+┌──────────────┐     TriggerArchiving()      ┌───────────────┐
+│  API Handler │  ───────────────────────▶   │ archiveChannel│
+│ (Job Stop)   │                             │  (buffer: 128)│
+└──────────────┘                             └───────┬───────┘
+                                                     │
+                   ┌─────────────────────────────────┘
+                   │
+                   ▼
+         ┌──────────────────────┐
+         │  archivingWorker()   │
+         │   (goroutine)        │
+         └──────────┬───────────┘
+                    │
+                    ▼
+         1. Fetch job metadata
+         2. Load metric data
+         3. Calculate statistics
+         4. Archive to backend
+         5. Update database
+         6. Call hooks
+```
+
+### Components
+
+- **archiveChannel**: Buffered channel (128 jobs) for async communication
+- **archivePending**: WaitGroup tracking in-flight archiving operations
+- **archivingWorker**: Background goroutine processing archiving requests
+- **shutdownCtx**: Context for graceful cancellation during shutdown
+
+## Usage
+
+### Initialization
+
+```go
+// Start archiver with context for shutdown control
+ctx, cancel := context.WithCancel(context.Background())
+defer cancel()
+
+archiver.Start(jobRepository, ctx)
+```
+
+### Archiving a Job
+
+```go
+// Called automatically when a job completes
+archiver.TriggerArchiving(job)
+```
+
+The function returns immediately. Actual archiving happens in the background.
+
+### Graceful Shutdown
+
+```go
+// Shutdown with 10 second timeout
+if err := archiver.Shutdown(10 * time.Second); err != nil {
+    log.Printf("Archiver shutdown timeout: %v", err)
+}
+```
+
+**Shutdown process:**
+1. Closes channel (rejects new jobs)
+2. Waits for pending jobs (up to timeout)
+3. Cancels context if timeout exceeded
+4. Waits for worker to exit cleanly
+
+## Configuration
+
+### Channel Buffer Size
+
+The archiving channel has a buffer of 128 jobs. If more than 128 jobs are queued simultaneously, `TriggerArchiving()` will block until space is available.
+
+To adjust:
+```go
+// In archiveWorker.go Start() function
+archiveChannel = make(chan *schema.Job, 256) // Increase buffer
+```
+
+### Scope Selection
+
+Archive data scopes are automatically selected based on job size:
+
+- **Node scope**: Always included
+- **Core scope**: Included for jobs with ≤8 nodes (reduces data volume for large jobs)
+- **Accelerator scope**: Included if job used accelerators (`NumAcc > 0`)
+
+To adjust the node threshold:
+```go
+// In archiver.go ArchiveJob() function
+if job.NumNodes <= 16 { // Change from 8 to 16
+    scopes = append(scopes, schema.MetricScopeCore)
+}
+```
+
+### Resolution
+
+Data is archived at the highest available resolution (typically 60s intervals). To change:
+
+```go
+// In archiver.go ArchiveJob() function
+jobData, err := metricDataDispatcher.LoadData(job, allMetrics, scopes, ctx, 300)
+// 0 = highest resolution
+// 300 = 5-minute resolution
+```
+
+## Error Handling
+
+### Automatic Retry
+
+The archiver does **not** automatically retry failed archiving operations. If archiving fails:
+
+1. Error is logged
+2. Job is marked as `MonitoringStatusArchivingFailed` in database
+3. Worker continues processing other jobs
+
+### Manual Retry
+
+To re-archive failed jobs, query for jobs with `MonitoringStatusArchivingFailed` and call `TriggerArchiving()` again.
+
+## Performance Considerations
+
+### Single Worker Thread
+
+The archiver uses a single worker goroutine. For high-throughput systems:
+
+- Large channel buffer (128) prevents blocking
+- Archiving is typically I/O bound (writing to storage)
+- Single worker prevents overwhelming storage backend
+
+### Shutdown Timeout
+
+Recommended timeout values:
+- **Development**: 5-10 seconds
+- **Production**: 10-30 seconds
+- **High-load**: 30-60 seconds
+
+Choose based on:
+- Average archiving time per job
+- Storage backend latency
+- Acceptable shutdown delay
+
+## Monitoring
+
+### Logging
+
+The archiver logs:
+- **Info**: Startup, shutdown, successful completions
+- **Debug**: Individual job archiving times
+- **Error**: Archiving failures with job ID and reason
+- **Warn**: Shutdown timeout exceeded
+
+### Metrics
+
+Monitor these signals for archiver health:
+- Jobs with `MonitoringStatusArchivingFailed`
+- Time from job stop to successful archive
+- Shutdown timeout occurrences
+
+## Thread Safety
+
+All exported functions are safe for concurrent use:
+- `Start()` - Safe to call once
+- `TriggerArchiving()` - Safe from multiple goroutines
+- `Shutdown()` - Safe to call once
+- `WaitForArchiving()` - Deprecated, but safe
+
+Internal state is protected by:
+- Channel synchronization (`archiveChannel`)
+- WaitGroup for pending count (`archivePending`)
+- Context for cancellation (`shutdownCtx`)
+
+## Files
+
+- **archiveWorker.go**: Worker lifecycle, channel management, shutdown logic
+- **archiver.go**: Core archiving logic, metric loading, statistics calculation
+
+## Dependencies
+
+- `internal/repository`: Database operations for job metadata
+- `internal/metricDataDispatcher`: Loading metric data from various backends
+- `pkg/archive`: Archive backend abstraction (filesystem, S3, SQLite)
+- `cc-lib/schema`: Job and metric data structures

internal/archiver/archiveWorker.go

@@ -0,0 +1,250 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+// Package archiver provides asynchronous job archiving functionality for ClusterCockpit.
+//
+// The archiver runs a background worker goroutine that processes job archiving requests
+// from a buffered channel. When jobs complete, their metric data is archived from the
+// metric store to the configured archive backend (filesystem, S3, etc.).
+//
+// # Architecture
+//
+// The archiver uses a producer-consumer pattern:
+//   - Producer: TriggerArchiving() sends jobs to archiveChannel
+//   - Consumer: archivingWorker() processes jobs from the channel
+//   - Coordination: sync.WaitGroup tracks pending archive operations
+//
+// # Lifecycle
+//
+//  1. Start(repo, ctx) - Initialize worker with context for cancellation
+//  2. TriggerArchiving(job) - Queue job for archiving (called when job stops)
+//  3. archivingWorker() - Background goroutine processes jobs
+//  4. Shutdown(timeout) - Graceful shutdown with timeout
+//
+// # Graceful Shutdown
+//
+// The archiver supports graceful shutdown with configurable timeout:
+//   - Closes channel to reject new jobs
+//   - Waits for pending jobs to complete (up to timeout)
+//   - Cancels context if timeout exceeded
+//   - Ensures worker goroutine exits cleanly
+//
+// # Example Usage
+//
+//	// Initialize archiver
+//	ctx, cancel := context.WithCancel(context.Background())
+//	defer cancel()
+//	archiver.Start(jobRepository, ctx)
+//
+//	// Trigger archiving when job completes
+//	archiver.TriggerArchiving(job)
+//
+//	// Graceful shutdown with 10 second timeout
+//	if err := archiver.Shutdown(10 * time.Second); err != nil {
+//	    log.Printf("Archiver shutdown timeout: %v", err)
+//	}
+package archiver
+
+import (
+	"context"
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	sq "github.com/Masterminds/squirrel"
+)
+
+var (
+	archivePending sync.WaitGroup
+	archiveChannel chan *schema.Job
+	jobRepo        *repository.JobRepository
+	shutdownCtx    context.Context
+	shutdownCancel context.CancelFunc
+	workerDone     chan struct{}
+)
+
+// Start initializes the archiver and starts the background worker goroutine.
+//
+// The archiver processes job archiving requests asynchronously via a buffered channel.
+// Jobs are sent to the channel using TriggerArchiving() and processed by the worker.
+//
+// Parameters:
+//   - r: JobRepository instance for database operations
+//   - ctx: Context for cancellation (shutdown signal propagation)
+//
+// The worker goroutine will run until:
+//   - ctx is cancelled (via parent shutdown)
+//   - archiveChannel is closed (via Shutdown())
+//
+// Must be called before TriggerArchiving(). Safe to call only once.
+func Start(r *repository.JobRepository, ctx context.Context) {
+	shutdownCtx, shutdownCancel = context.WithCancel(ctx)
+	archiveChannel = make(chan *schema.Job, 128)
+	workerDone = make(chan struct{})
+	jobRepo = r
+
+	go archivingWorker()
+}
+
+// archivingWorker is the background goroutine that processes job archiving requests.
+//
+// The worker loop:
+//  1. Blocks waiting for jobs on archiveChannel or shutdown signal
+//  2. Fetches job metadata from repository
+//  3. Archives job data to configured backend (calls ArchiveJob)
+//  4. Updates job footprint and energy metrics in database
+//  5. Marks job as successfully archived
+//  6. Calls job stop hooks
+//
+// The worker exits when:
+//   - shutdownCtx is cancelled (timeout during shutdown)
+//   - archiveChannel is closed (normal shutdown)
+//
+// Errors during archiving are logged and the job is marked as failed,
+// but the worker continues processing other jobs.
+func archivingWorker() {
+	defer close(workerDone)
+
+	for {
+		select {
+		case <-shutdownCtx.Done():
+			cclog.Info("Archive worker received shutdown signal")
+			return
+
+		case job, ok := <-archiveChannel:
+			if !ok {
+				cclog.Info("Archive channel closed, worker exiting")
+				return
+			}
+
+			start := time.Now()
+			// not using meta data, called to load JobMeta into Cache?
+			// will fail if job meta not in repository
+			if _, err := jobRepo.FetchMetadata(job); err != nil {
+				cclog.Errorf("archiving job (dbid: %d) failed at check metadata step: %s", job.ID, err.Error())
+				jobRepo.UpdateMonitoringStatus(*job.ID, schema.MonitoringStatusArchivingFailed)
+				archivePending.Done()
+				continue
+			}
+
+			// ArchiveJob will fetch all the data from a MetricDataRepository and push into configured archive backend
+			// Use shutdown context to allow cancellation
+			jobMeta, err := ArchiveJob(job, shutdownCtx)
+			if err != nil {
+				cclog.Errorf("archiving job (dbid: %d) failed at archiving job step: %s", job.ID, err.Error())
+				jobRepo.UpdateMonitoringStatus(*job.ID, schema.MonitoringStatusArchivingFailed)
+				archivePending.Done()
+				continue
+			}
+
+			stmt := sq.Update("job").Where("job.id = ?", job.ID)
+
+			if stmt, err = jobRepo.UpdateFootprint(stmt, jobMeta); err != nil {
+				cclog.Errorf("archiving job (dbid: %d) failed at update Footprint step: %s", job.ID, err.Error())
+				archivePending.Done()
+				continue
+			}
+			if stmt, err = jobRepo.UpdateEnergy(stmt, jobMeta); err != nil {
+				cclog.Errorf("archiving job (dbid: %d) failed at update Energy step: %s", job.ID, err.Error())
+				archivePending.Done()
+				continue
+			}
+			// Update the jobs database entry one last time:
+			stmt = jobRepo.MarkArchived(stmt, schema.MonitoringStatusArchivingSuccessful)
+			if err := jobRepo.Execute(stmt); err != nil {
+				cclog.Errorf("archiving job (dbid: %d) failed at db execute: %s", job.ID, err.Error())
+				archivePending.Done()
+				continue
+			}
+			cclog.Debugf("archiving job %d took %s", job.JobID, time.Since(start))
+			cclog.Infof("archiving job (dbid: %d) successful", job.ID)
+
+			repository.CallJobStopHooks(job)
+			archivePending.Done()
+		}
+	}
+}
+
+// TriggerArchiving queues a job for asynchronous archiving.
+//
+// This function should be called when a job completes (stops) to archive its
+// metric data from the metric store to the configured archive backend.
+//
+// The function:
+//  1. Increments the pending job counter (WaitGroup)
+//  2. Sends the job to the archiving channel (buffered, capacity 128)
+//  3. Returns immediately (non-blocking unless channel is full)
+//
+// The actual archiving is performed asynchronously by the worker goroutine.
+// Upon completion, the worker will decrement the pending counter.
+//
+// Panics if Start() has not been called first.
+func TriggerArchiving(job *schema.Job) {
+	if archiveChannel == nil {
+		cclog.Fatal("Cannot archive without archiving channel. Did you Start the archiver?")
+	}
+
+	archivePending.Add(1)
+	archiveChannel <- job
+}
+
+// Shutdown performs a graceful shutdown of the archiver with a configurable timeout.
+//
+// The shutdown process:
+//  1. Closes archiveChannel - no new jobs will be accepted
+//  2. Waits for pending jobs to complete (up to timeout duration)
+//  3. If timeout is exceeded:
+//     - Cancels shutdownCtx to interrupt ongoing ArchiveJob operations
+//     - Returns error indicating timeout
+//  4. Waits for worker goroutine to exit cleanly
+//
+// Parameters:
+//   - timeout: Maximum duration to wait for pending jobs to complete
+//     (recommended: 10-30 seconds for production)
+//
+// Returns:
+//   - nil if all jobs completed within timeout
+//   - error if timeout was exceeded (some jobs may not have been archived)
+//
+// Jobs that don't complete within the timeout will be marked as failed.
+// The function always ensures the worker goroutine exits before returning.
+//
+// Example:
+//
+//	if err := archiver.Shutdown(10 * time.Second); err != nil {
+//	    log.Printf("Some jobs did not complete: %v", err)
+//	}
+func Shutdown(timeout time.Duration) error {
+	cclog.Info("Initiating archiver shutdown...")
+
+	// Close channel to signal no more jobs will be accepted
+	close(archiveChannel)
+
+	// Create a channel to signal when all jobs are done
+	done := make(chan struct{})
+	go func() {
+		archivePending.Wait()
+		close(done)
+	}()
+
+	// Wait for jobs to complete or timeout
+	select {
+	case <-done:
+		cclog.Info("All archive jobs completed successfully")
+		// Wait for worker to exit
+		<-workerDone
+		return nil
+	case <-time.After(timeout):
+		cclog.Warn("Archiver shutdown timeout exceeded, cancelling remaining operations")
+		// Cancel any ongoing operations
+		shutdownCancel()
+		// Wait for worker to exit
+		<-workerDone
+		return fmt.Errorf("archiver shutdown timeout after %v", timeout)
+	}
+}

internal/archiver/archiver.go

@@ -0,0 +1,105 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package archiver
+
+import (
+	"context"
+	"math"
+
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
+	"github.com/ClusterCockpit/cc-backend/pkg/archive"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+)
+
+// ArchiveJob archives a completed job's metric data to the configured archive backend.
+//
+// This function performs the following operations:
+//  1. Loads all metric data for the job from the metric data repository
+//  2. Calculates job-level statistics (avg, min, max) for each metric
+//  3. Stores the job metadata and metric data to the archive backend
+//
+// Metric data is retrieved at the highest available resolution (typically 60s)
+// for the following scopes:
+//   - Node scope (always)
+//   - Core scope (for jobs with ≤8 nodes, to reduce data volume)
+//   - Accelerator scope (if job used accelerators)
+//
+// The function respects context cancellation. If ctx is cancelled (e.g., during
+// shutdown timeout), the operation will be interrupted and return an error.
+//
+// Parameters:
+//   - job: The job to archive (must be a completed job)
+//   - ctx: Context for cancellation and timeout control
+//
+// Returns:
+//   - *schema.Job with populated Statistics field
+//   - error if data loading or archiving fails
+//
+// If config.Keys.DisableArchive is true, only job statistics are calculated
+// and returned (no data is written to archive backend).
+func ArchiveJob(job *schema.Job, ctx context.Context) (*schema.Job, error) {
+	allMetrics := make([]string, 0)
+	metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
+	for _, mc := range metricConfigs {
+		allMetrics = append(allMetrics, mc.Name)
+	}
+
+	scopes := []schema.MetricScope{schema.MetricScopeNode}
+	// FIXME: Add a config option for this
+	if job.NumNodes <= 8 {
+		// This will add the native scope if core scope is not available
+		scopes = append(scopes, schema.MetricScopeCore)
+	}
+
+	if job.NumAcc > 0 {
+		scopes = append(scopes, schema.MetricScopeAccelerator)
+	}
+
+	jobData, err := metricDataDispatcher.LoadData(job, allMetrics, scopes, ctx, 0) // 0 Resulotion-Value retrieves highest res (60s)
+	if err != nil {
+		cclog.Error("Error wile loading job data for archiving")
+		return nil, err
+	}
+
+	job.Statistics = make(map[string]schema.JobStatistics)
+
+	for metric, data := range jobData {
+		avg, min, max := 0.0, math.MaxFloat32, -math.MaxFloat32
+		nodeData, ok := data["node"]
+		if !ok {
+			// This should never happen ?
+			continue
+		}
+
+		for _, series := range nodeData.Series {
+			avg += series.Statistics.Avg
+			min = math.Min(min, series.Statistics.Min)
+			max = math.Max(max, series.Statistics.Max)
+		}
+
+		// Round AVG Result to 2 Digits
+		job.Statistics[metric] = schema.JobStatistics{
+			Unit: schema.Unit{
+				Prefix: archive.GetMetricConfig(job.Cluster, metric).Unit.Prefix,
+				Base:   archive.GetMetricConfig(job.Cluster, metric).Unit.Base,
+			},
+			Avg: (math.Round((avg/float64(job.NumNodes))*100) / 100),
+			Min: min,
+			Max: max,
+		}
+	}
+
+	// If the file based archive is disabled,
+	// only return the JobMeta structure as the
+	// statistics in there are needed.
+	if config.Keys.DisableArchive {
+		return job, nil
+	}
+
+	return job, archive.GetHandle().ImportJob(job, &jobData)
+}

internal/auth/auth.go

@@ -1,298 +1,137 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
+// Package auth implements various authentication methods
 package auth
 
 import (
+	"bytes"
 	"context"
 	"crypto/rand"
+	"database/sql"
 	"encoding/base64"
+	"encoding/json"
 	"errors"
 	"fmt"
+	"net"
 	"net/http"
 	"os"
-	"strings"
+	"sync"
 	"time"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
+	"golang.org/x/time/rate"
+
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/ClusterCockpit/cc-lib/util"
 	"github.com/gorilla/sessions"
-	"github.com/jmoiron/sqlx"
 )
 
-type AuthSource int
-
-const (
-	AuthViaLocalPassword AuthSource = iota
-	AuthViaLDAP
-	AuthViaToken
-)
-
-type User struct {
-	Username   string     `json:"username"`
-	Password   string     `json:"-"`
-	Name       string     `json:"name"`
-	Roles      []string   `json:"roles"`
-	AuthSource AuthSource `json:"via"`
-	Email      string     `json:"email"`
-	Projects   []string   `json:"projects"`
-	Expiration time.Time
-}
-
-type Role int
-
-const (
-	RoleAnonymous Role = iota
-	RoleApi
-	RoleUser
-	RoleManager
-	RoleSupport
-	RoleAdmin
-	RoleError
-)
-
-func GetRoleString(roleInt Role) string {
-	return [6]string{"anonymous", "api", "user", "manager", "support", "admin"}[roleInt]
-}
-
-func getRoleEnum(roleStr string) Role {
-	switch strings.ToLower(roleStr) {
-	case "admin":
-		return RoleAdmin
-	case "support":
-		return RoleSupport
-	case "manager":
-		return RoleManager
-	case "user":
-		return RoleUser
-	case "api":
-		return RoleApi
-	case "anonymous":
-		return RoleAnonymous
-	default:
-		return RoleError
-	}
-}
-
-func isValidRole(role string) bool {
-	if getRoleEnum(role) == RoleError {
-		return false
-	}
-	return true
-}
-
-func (u *User) HasValidRole(role string) (hasRole bool, isValid bool) {
-	if isValidRole(role) {
-		for _, r := range u.Roles {
-			if r == role {
-				return true, true
-			}
-		}
-		return false, true
-	}
-	return false, false
-}
-
-func (u *User) HasRole(role Role) bool {
-	for _, r := range u.Roles {
-		if r == GetRoleString(role) {
-			return true
-		}
-	}
-	return false
-}
-
-// Role-Arrays are short: performance not impacted by nested loop
-func (u *User) HasAnyRole(queryroles []Role) bool {
-	for _, ur := range u.Roles {
-		for _, qr := range queryroles {
-			if ur == GetRoleString(qr) {
-				return true
-			}
-		}
-	}
-	return false
-}
-
-// Role-Arrays are short: performance not impacted by nested loop
-func (u *User) HasAllRoles(queryroles []Role) bool {
-	target := len(queryroles)
-	matches := 0
-	for _, ur := range u.Roles {
-		for _, qr := range queryroles {
-			if ur == GetRoleString(qr) {
-				matches += 1
-				break
-			}
-		}
-	}
-
-	if matches == target {
-		return true
-	} else {
-		return false
-	}
-}
-
-// Role-Arrays are short: performance not impacted by nested loop
-func (u *User) HasNotRoles(queryroles []Role) bool {
-	matches := 0
-	for _, ur := range u.Roles {
-		for _, qr := range queryroles {
-			if ur == GetRoleString(qr) {
-				matches += 1
-				break
-			}
-		}
-	}
-
-	if matches == 0 {
-		return true
-	} else {
-		return false
-	}
-}
-
-// Called by API endpoint '/roles/' from frontend: Only required for admin config -> Check Admin Role
-func GetValidRoles(user *User) ([]string, error) {
-	var vals []string
-	if user.HasRole(RoleAdmin) {
-		for i := RoleApi; i < RoleError; i++ {
-			vals = append(vals, GetRoleString(i))
-		}
-		return vals, nil
-	}
-
-	return vals, fmt.Errorf("%s: only admins are allowed to fetch a list of roles", user.Username)
-}
-
-// Called by routerConfig web.page setup in backend: Only requires known user and/or not API user
-func GetValidRolesMap(user *User) (map[string]Role, error) {
-	named := make(map[string]Role)
-	if user.HasNotRoles([]Role{RoleApi, RoleAnonymous}) {
-		for i := RoleApi; i < RoleError; i++ {
-			named[GetRoleString(i)] = i
-		}
-		return named, nil
-	}
-	return named, fmt.Errorf("Only known users are allowed to fetch a list of roles")
-}
-
-// Find highest role
-func (u *User) GetAuthLevel() Role {
-	if u.HasRole(RoleAdmin) {
-		return RoleAdmin
-	} else if u.HasRole(RoleSupport) {
-		return RoleSupport
-	} else if u.HasRole(RoleManager) {
-		return RoleManager
-	} else if u.HasRole(RoleUser) {
-		return RoleUser
-	} else if u.HasRole(RoleApi) {
-		return RoleApi
-	} else if u.HasRole(RoleAnonymous) {
-		return RoleAnonymous
-	} else {
-		return RoleError
-	}
-}
-
-func (u *User) HasProject(project string) bool {
-	for _, p := range u.Projects {
-		if p == project {
-			return true
-		}
-	}
-	return false
-}
-
-func GetUser(ctx context.Context) *User {
-	x := ctx.Value(ContextUserKey)
-	if x == nil {
-		return nil
-	}
-
-	return x.(*User)
-}
-
+// Authenticator is the interface for all authentication methods.
+// Each authenticator determines if it can handle a login request (CanLogin)
+// and performs the actual authentication (Login).
 type Authenticator interface {
-	Init(auth *Authentication, config interface{}) error
-	CanLogin(user *User, rw http.ResponseWriter, r *http.Request) bool
-	Login(user *User, rw http.ResponseWriter, r *http.Request) (*User, error)
-	Auth(rw http.ResponseWriter, r *http.Request) (*User, error)
+	// CanLogin determines if this authenticator can handle the login request.
+	// It returns the user object if available and a boolean indicating if this
+	// authenticator should attempt the login. This method should not perform
+	// expensive operations or actual authentication.
+	CanLogin(user *schema.User, username string, rw http.ResponseWriter, r *http.Request) (*schema.User, bool)
+	
+	// Login performs the actually authentication for the user.
+	// It returns the authenticated user or an error if authentication fails.
+	// The user parameter may be nil if the user doesn't exist in the database yet.
+	Login(user *schema.User, rw http.ResponseWriter, r *http.Request) (*schema.User, error)
 }
 
-type ContextKey string
+var (
+	initOnce     sync.Once
+	authInstance *Authentication
+)
 
-const ContextUserKey ContextKey = "user"
+// rateLimiterEntry tracks a rate limiter and its last use time for cleanup
+type rateLimiterEntry struct {
+	limiter  *rate.Limiter
+	lastUsed time.Time
+}
 
+var ipUserLimiters sync.Map
+
+// getIPUserLimiter returns a rate limiter for the given IP and username combination.
+// Rate limiters are created on demand and track 5 attempts per 15 minutes.
+func getIPUserLimiter(ip, username string) *rate.Limiter {
+	key := ip + ":" + username
+	now := time.Now()
+	
+	if entry, ok := ipUserLimiters.Load(key); ok {
+		rle := entry.(*rateLimiterEntry)
+		rle.lastUsed = now
+		return rle.limiter
+	}
+	
+	// More aggressive rate limiting: 5 attempts per 15 minutes
+	newLimiter := rate.NewLimiter(rate.Every(15*time.Minute/5), 5)
+	ipUserLimiters.Store(key, &rateLimiterEntry{
+		limiter:  newLimiter,
+		lastUsed: now,
+	})
+	return newLimiter
+}
+
+// cleanupOldRateLimiters removes rate limiters that haven't been used recently
+func cleanupOldRateLimiters(olderThan time.Time) {
+	ipUserLimiters.Range(func(key, value any) bool {
+		entry := value.(*rateLimiterEntry)
+		if entry.lastUsed.Before(olderThan) {
+			ipUserLimiters.Delete(key)
+			cclog.Debugf("Cleaned up rate limiter for %v", key)
+		}
+		return true
+	})
+}
+
+// startRateLimiterCleanup starts a background goroutine to clean up old rate limiters
+func startRateLimiterCleanup() {
+	go func() {
+		ticker := time.NewTicker(1 * time.Hour)
+		defer ticker.Stop()
+		for range ticker.C {
+			// Clean up limiters not used in the last 24 hours
+			cleanupOldRateLimiters(time.Now().Add(-24 * time.Hour))
+		}
+	}()
+}
+
+// AuthConfig contains configuration for all authentication methods
+type AuthConfig struct {
+	LdapConfig   *LdapConfig    `json:"ldap"`
+	JwtConfig    *JWTAuthConfig `json:"jwts"`
+	OpenIDConfig *OpenIDConfig  `json:"oidc"`
+}
+
+// Keys holds the global authentication configuration
+var Keys AuthConfig
+
+// Authentication manages all authentication methods and session handling
 type Authentication struct {
-	db            *sqlx.DB
-	sessionStore  *sessions.CookieStore
-	SessionMaxAge time.Duration
-
-	authenticators []Authenticator
+	sessionStore   *sessions.CookieStore
 	LdapAuth       *LdapAuthenticator
 	JwtAuth        *JWTAuthenticator
 	LocalAuth      *LocalAuthenticator
-}
-
-func Init(db *sqlx.DB,
-	configs map[string]interface{}) (*Authentication, error) {
-	auth := &Authentication{}
-	auth.db = db
-
-	sessKey := os.Getenv("SESSION_KEY")
-	if sessKey == "" {
-		log.Warn("environment variable 'SESSION_KEY' not set (will use non-persistent random key)")
-		bytes := make([]byte, 32)
-		if _, err := rand.Read(bytes); err != nil {
-			log.Error("Error while initializing authentication -> failed to generate random bytes for session key")
-			return nil, err
-		}
-		auth.sessionStore = sessions.NewCookieStore(bytes)
-	} else {
-		bytes, err := base64.StdEncoding.DecodeString(sessKey)
-		if err != nil {
-			log.Error("Error while initializing authentication -> decoding session key failed")
-			return nil, err
-		}
-		auth.sessionStore = sessions.NewCookieStore(bytes)
-	}
-
-	auth.LocalAuth = &LocalAuthenticator{}
-	if err := auth.LocalAuth.Init(auth, nil); err != nil {
-		log.Error("Error while initializing authentication -> localAuth init failed")
-		return nil, err
-	}
-	auth.authenticators = append(auth.authenticators, auth.LocalAuth)
-
-	auth.JwtAuth = &JWTAuthenticator{}
-	if err := auth.JwtAuth.Init(auth, configs["jwt"]); err != nil {
-		log.Error("Error while initializing authentication -> jwtAuth init failed")
-		return nil, err
-	}
-	auth.authenticators = append(auth.authenticators, auth.JwtAuth)
-
-	if config, ok := configs["ldap"]; ok {
-		auth.LdapAuth = &LdapAuthenticator{}
-		if err := auth.LdapAuth.Init(auth, config); err != nil {
-			log.Error("Error while initializing authentication -> ldapAuth init failed")
-			return nil, err
-		}
-		auth.authenticators = append(auth.authenticators, auth.LdapAuth)
-	}
-
-	return auth, nil
+	authenticators []Authenticator
+	SessionMaxAge  time.Duration
 }
 
 func (auth *Authentication) AuthViaSession(
 	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
+	r *http.Request,
+) (*schema.User, error) {
 	session, err := auth.sessionStore.Get(r, "session")
 	if err != nil {
-		log.Error("Error while getting session store")
+		cclog.Error("Error while getting session store")
 		return nil, err
 	}
 
@@ -300,108 +139,441 @@ func (auth *Authentication) AuthViaSession(
 		return nil, nil
 	}
 
-	username, _ := session.Values["username"].(string)
-	projects, _ := session.Values["projects"].([]string)
-	roles, _ := session.Values["roles"].([]string)
-	return &User{
+	// Validate session data with proper type checking
+	username, ok := session.Values["username"].(string)
+	if !ok || username == "" {
+		cclog.Warn("Invalid session: missing or invalid username")
+		// Invalidate the corrupted session
+		session.Options.MaxAge = -1
+		_ = auth.sessionStore.Save(r, rw, session)
+		return nil, errors.New("invalid session data")
+	}
+
+	projects, ok := session.Values["projects"].([]string)
+	if !ok {
+		cclog.Warn("Invalid session: projects not found or invalid type, using empty list")
+		projects = []string{}
+	}
+
+	roles, ok := session.Values["roles"].([]string)
+	if !ok || len(roles) == 0 {
+		cclog.Warn("Invalid session: missing or invalid roles")
+		// Invalidate the corrupted session
+		session.Options.MaxAge = -1
+		_ = auth.sessionStore.Save(r, rw, session)
+		return nil, errors.New("invalid session data")
+	}
+
+	return &schema.User{
 		Username:   username,
 		Projects:   projects,
 		Roles:      roles,
+		AuthType:   schema.AuthSession,
 		AuthSource: -1,
 	}, nil
 }
 
-// Handle a POST request that should log the user in, starting a new session.
-func (auth *Authentication) Login(
-	onsuccess http.Handler,
-	onfailure func(rw http.ResponseWriter, r *http.Request, loginErr error)) http.Handler {
+func Init(authCfg *json.RawMessage) {
+	initOnce.Do(func() {
+		authInstance = &Authentication{}
+		
+		// Start background cleanup of rate limiters
+		startRateLimiterCleanup()
 
+		sessKey := os.Getenv("SESSION_KEY")
+		if sessKey == "" {
+			cclog.Warn("environment variable 'SESSION_KEY' not set (will use non-persistent random key)")
+			bytes := make([]byte, 32)
+			if _, err := rand.Read(bytes); err != nil {
+				cclog.Fatal("Error while initializing authentication -> failed to generate random bytes for session key")
+			}
+			authInstance.sessionStore = sessions.NewCookieStore(bytes)
+		} else {
+			bytes, err := base64.StdEncoding.DecodeString(sessKey)
+			if err != nil {
+				cclog.Fatal("Error while initializing authentication -> decoding session key failed")
+			}
+			authInstance.sessionStore = sessions.NewCookieStore(bytes)
+		}
+
+		if d, err := time.ParseDuration(config.Keys.SessionMaxAge); err == nil {
+			authInstance.SessionMaxAge = d
+		}
+
+		if authCfg == nil {
+			return
+		}
+
+		config.Validate(configSchema, *authCfg)
+		dec := json.NewDecoder(bytes.NewReader(*authCfg))
+		dec.DisallowUnknownFields()
+		if err := dec.Decode(&Keys); err != nil {
+			cclog.Errorf("error while decoding ldap config: %v", err)
+		}
+
+		if Keys.LdapConfig != nil {
+			ldapAuth := &LdapAuthenticator{}
+			if err := ldapAuth.Init(); err != nil {
+				cclog.Warn("Error while initializing authentication -> ldapAuth init failed")
+			} else {
+				authInstance.LdapAuth = ldapAuth
+				authInstance.authenticators = append(authInstance.authenticators, authInstance.LdapAuth)
+			}
+		} else {
+			cclog.Info("Missing LDAP configuration: No LDAP support!")
+		}
+
+		if Keys.JwtConfig != nil {
+			authInstance.JwtAuth = &JWTAuthenticator{}
+			if err := authInstance.JwtAuth.Init(); err != nil {
+				cclog.Fatal("Error while initializing authentication -> jwtAuth init failed")
+			}
+
+			jwtSessionAuth := &JWTSessionAuthenticator{}
+			if err := jwtSessionAuth.Init(); err != nil {
+				cclog.Info("jwtSessionAuth init failed: No JWT login support!")
+			} else {
+				authInstance.authenticators = append(authInstance.authenticators, jwtSessionAuth)
+			}
+
+			jwtCookieSessionAuth := &JWTCookieSessionAuthenticator{}
+			if err := jwtCookieSessionAuth.Init(); err != nil {
+				cclog.Info("jwtCookieSessionAuth init failed: No JWT cookie login support!")
+			} else {
+				authInstance.authenticators = append(authInstance.authenticators, jwtCookieSessionAuth)
+			}
+		} else {
+			cclog.Info("Missing JWT configuration: No JWT token support!")
+		}
+
+		authInstance.LocalAuth = &LocalAuthenticator{}
+		if err := authInstance.LocalAuth.Init(); err != nil {
+			cclog.Fatal("Error while initializing authentication -> localAuth init failed")
+		}
+		authInstance.authenticators = append(authInstance.authenticators, authInstance.LocalAuth)
+	})
+}
+
+func GetAuthInstance() *Authentication {
+	if authInstance == nil {
+		cclog.Fatal("Authentication module not initialized!")
+	}
+
+	return authInstance
+}
+
+// handleUserSync syncs or updates a user in the database based on configuration.
+// This is used for both JWT and OIDC authentication when syncUserOnLogin or updateUserOnLogin is enabled.
+func handleUserSync(user *schema.User, syncUserOnLogin, updateUserOnLogin bool) {
+	r := repository.GetUserRepository()
+	dbUser, err := r.GetUser(user.Username)
+
+	if err != nil && err != sql.ErrNoRows {
+		cclog.Errorf("Error while loading user '%s': %v", user.Username, err)
+		return
+	}
+	
+	if err == sql.ErrNoRows && syncUserOnLogin { // Add new user
+		if err := r.AddUser(user); err != nil {
+			cclog.Errorf("Error while adding user '%s' to DB: %v", user.Username, err)
+		}
+	} else if err == nil && updateUserOnLogin { // Update existing user
+		if err := r.UpdateUser(dbUser, user); err != nil {
+			cclog.Errorf("Error while updating user '%s' in DB: %v", dbUser.Username, err)
+		}
+	}
+}
+
+// handleTokenUser syncs JWT token user with database
+func handleTokenUser(tokenUser *schema.User) {
+	handleUserSync(tokenUser, Keys.JwtConfig.SyncUserOnLogin, Keys.JwtConfig.UpdateUserOnLogin)
+}
+
+// handleOIDCUser syncs OIDC user with database
+func handleOIDCUser(OIDCUser *schema.User) {
+	handleUserSync(OIDCUser, Keys.OpenIDConfig.SyncUserOnLogin, Keys.OpenIDConfig.UpdateUserOnLogin)
+}
+
+func (auth *Authentication) SaveSession(rw http.ResponseWriter, r *http.Request, user *schema.User) error {
+	session, err := auth.sessionStore.New(r, "session")
+	if err != nil {
+		cclog.Errorf("session creation failed: %s", err.Error())
+		http.Error(rw, err.Error(), http.StatusInternalServerError)
+		return err
+	}
+
+	if auth.SessionMaxAge != 0 {
+		session.Options.MaxAge = int(auth.SessionMaxAge.Seconds())
+	}
+	if config.Keys.HTTPSCertFile == "" && config.Keys.HTTPSKeyFile == "" {
+		cclog.Warn("HTTPS not configured - session cookies will not have Secure flag set (insecure for production)")
+		session.Options.Secure = false
+	}
+	session.Options.SameSite = http.SameSiteStrictMode
+	session.Values["username"] = user.Username
+	session.Values["projects"] = user.Projects
+	session.Values["roles"] = user.Roles
+	if err := auth.sessionStore.Save(r, rw, session); err != nil {
+		cclog.Warnf("session save failed: %s", err.Error())
+		http.Error(rw, err.Error(), http.StatusInternalServerError)
+		return err
+	}
+
+	return nil
+}
+
+func (auth *Authentication) Login(
+	onfailure func(rw http.ResponseWriter, r *http.Request, loginErr error),
+) http.Handler {
 	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
-		err := errors.New("no authenticator applied")
+		ip, _, err := net.SplitHostPort(r.RemoteAddr)
+		if err != nil {
+			ip = r.RemoteAddr
+		}
+
 		username := r.FormValue("username")
-		user := (*User)(nil)
+
+		limiter := getIPUserLimiter(ip, username)
+		if !limiter.Allow() {
+			cclog.Warnf("AUTH/RATE > Too many login attempts for combination IP: %s, Username: %s", ip, username)
+			onfailure(rw, r, errors.New("too many login attempts, try again in a few minutes"))
+			return
+		}
+
+		var dbUser *schema.User
 		if username != "" {
-			if user, _ = auth.GetUser(username); err != nil {
-				// log.Warnf("login of unkown user %v", username)
-				_ = err
+			var err error
+			dbUser, err = repository.GetUserRepository().GetUser(username)
+			if err != nil && err != sql.ErrNoRows {
+				cclog.Errorf("Error while loading user '%v'", username)
 			}
 		}
 
 		for _, authenticator := range auth.authenticators {
-			if !authenticator.CanLogin(user, rw, r) {
+			var ok bool
+			var user *schema.User
+			if user, ok = authenticator.CanLogin(dbUser, username, rw, r); !ok {
 				continue
+			} else {
+				cclog.Debugf("Can login with user %v", user)
 			}
 
-			user, err = authenticator.Login(user, rw, r)
+			user, err := authenticator.Login(user, rw, r)
 			if err != nil {
-				log.Warnf("user login failed: %s", err.Error())
+				cclog.Warnf("user login failed: %s", err.Error())
 				onfailure(rw, r, err)
 				return
 			}
 
-			session, err := auth.sessionStore.New(r, "session")
-			if err != nil {
-				log.Errorf("session creation failed: %s", err.Error())
-				http.Error(rw, err.Error(), http.StatusInternalServerError)
+			if err := auth.SaveSession(rw, r, user); err != nil {
 				return
 			}
 
-			if auth.SessionMaxAge != 0 {
-				session.Options.MaxAge = int(auth.SessionMaxAge.Seconds())
-			}
-			session.Values["username"] = user.Username
-			session.Values["projects"] = user.Projects
-			session.Values["roles"] = user.Roles
-			if err := auth.sessionStore.Save(r, rw, session); err != nil {
-				log.Warnf("session save failed: %s", err.Error())
-				http.Error(rw, err.Error(), http.StatusInternalServerError)
+			cclog.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
+			ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+
+			if r.FormValue("redirect") != "" {
+				http.RedirectHandler(r.FormValue("redirect"), http.StatusFound).ServeHTTP(rw, r.WithContext(ctx))
 				return
 			}
 
-			log.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
-			ctx := context.WithValue(r.Context(), ContextUserKey, user)
-			onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+			http.RedirectHandler("/", http.StatusFound).ServeHTTP(rw, r.WithContext(ctx))
 			return
 		}
 
-		log.Debugf("login failed: no authenticator applied")
-		onfailure(rw, r, err)
+		cclog.Debugf("login failed: no authenticator applied")
+		onfailure(rw, r, errors.New("no authenticator applied"))
 	})
 }
 
-// Authenticate the user and put a User object in the
-// context of the request. If authentication fails,
-// do not continue but send client to the login screen.
 func (auth *Authentication) Auth(
 	onsuccess http.Handler,
-	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error)) http.Handler {
-
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
 	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
-		for _, authenticator := range auth.authenticators {
-			user, err := authenticator.Auth(rw, r)
+		user, err := auth.JwtAuth.AuthViaJWT(rw, r)
+		if err != nil {
+			cclog.Infof("auth -> authentication failed: %s", err.Error())
+			http.Error(rw, err.Error(), http.StatusUnauthorized)
+			return
+		}
+		if user == nil {
+			user, err = auth.AuthViaSession(rw, r)
 			if err != nil {
-				log.Infof("authentication failed: %s", err.Error())
+				cclog.Infof("auth -> authentication failed: %s", err.Error())
 				http.Error(rw, err.Error(), http.StatusUnauthorized)
 				return
 			}
-			if user == nil {
-				continue
-			}
-
-			ctx := context.WithValue(r.Context(), ContextUserKey, user)
+		}
+		if user != nil {
+			ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
 			onsuccess.ServeHTTP(rw, r.WithContext(ctx))
 			return
 		}
 
-		log.Debugf("authentication failed: %s", "no authenticator applied")
-		// http.Error(rw, http.StatusText(http.StatusUnauthorized), http.StatusUnauthorized)
-		onfailure(rw, r, errors.New("unauthorized (login first or use a token)"))
+		cclog.Info("auth -> authentication failed")
+		onfailure(rw, r, errors.New("unauthorized (please login first)"))
 	})
 }
 
-// Clears the session cookie
-func (auth *Authentication) Logout(onsuccess http.Handler) http.Handler {
+func (auth *Authentication) AuthAPI(
+	onsuccess http.Handler,
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
+	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+		user, err := auth.JwtAuth.AuthViaJWT(rw, r)
+		if err != nil {
+			cclog.Infof("auth api -> authentication failed: %s", err.Error())
+			onfailure(rw, r, err)
+			return
+		}
 
+		ipErr := securedCheck(user, r)
+		if ipErr != nil {
+			cclog.Infof("auth api -> secured check failed: %s", ipErr.Error())
+			onfailure(rw, r, ipErr)
+			return
+		}
+
+		if user != nil {
+			switch {
+			case len(user.Roles) == 1:
+				if user.HasRole(schema.RoleApi) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			case len(user.Roles) >= 2:
+				if user.HasAllRoles([]schema.Role{schema.RoleAdmin, schema.RoleApi}) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			default:
+				cclog.Info("auth api -> authentication failed: missing role")
+				onfailure(rw, r, errors.New("unauthorized"))
+			}
+		}
+		cclog.Info("auth api -> authentication failed: no auth")
+		onfailure(rw, r, errors.New("unauthorized"))
+	})
+}
+
+func (auth *Authentication) AuthUserAPI(
+	onsuccess http.Handler,
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
+	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+		user, err := auth.JwtAuth.AuthViaJWT(rw, r)
+		if err != nil {
+			cclog.Infof("auth user api -> authentication failed: %s", err.Error())
+			onfailure(rw, r, err)
+			return
+		}
+
+		if user != nil {
+			switch {
+			case len(user.Roles) == 1:
+				if user.HasRole(schema.RoleApi) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			case len(user.Roles) >= 2:
+				if user.HasRole(schema.RoleApi) && user.HasAnyRole([]schema.Role{schema.RoleUser, schema.RoleManager, schema.RoleSupport, schema.RoleAdmin}) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			default:
+				cclog.Info("auth user api -> authentication failed: missing role")
+				onfailure(rw, r, errors.New("unauthorized"))
+			}
+		}
+		cclog.Info("auth user api -> authentication failed: no auth")
+		onfailure(rw, r, errors.New("unauthorized"))
+	})
+}
+
+func (auth *Authentication) AuthMetricStoreAPI(
+	onsuccess http.Handler,
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
+	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+		user, err := auth.JwtAuth.AuthViaJWT(rw, r)
+		if err != nil {
+			cclog.Infof("auth metricstore api -> authentication failed: %s", err.Error())
+			onfailure(rw, r, err)
+			return
+		}
+
+		if user != nil {
+			switch {
+			case len(user.Roles) == 1:
+				if user.HasRole(schema.RoleApi) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			case len(user.Roles) >= 2:
+				if user.HasRole(schema.RoleApi) && user.HasAnyRole([]schema.Role{schema.RoleUser, schema.RoleManager, schema.RoleAdmin}) {
+					ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+					onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+					return
+				}
+			default:
+				cclog.Info("auth metricstore api -> authentication failed: missing role")
+				onfailure(rw, r, errors.New("unauthorized"))
+			}
+		}
+		cclog.Info("auth metricstore api -> authentication failed: no auth")
+		onfailure(rw, r, errors.New("unauthorized"))
+	})
+}
+
+func (auth *Authentication) AuthConfigAPI(
+	onsuccess http.Handler,
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
+	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+		user, err := auth.AuthViaSession(rw, r)
+		if err != nil {
+			cclog.Infof("auth config api -> authentication failed: %s", err.Error())
+			onfailure(rw, r, err)
+			return
+		}
+		if user != nil && user.HasRole(schema.RoleAdmin) {
+			ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+			onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+			return
+		}
+		cclog.Info("auth config api -> authentication failed: no auth")
+		onfailure(rw, r, errors.New("unauthorized"))
+	})
+}
+
+func (auth *Authentication) AuthFrontendAPI(
+	onsuccess http.Handler,
+	onfailure func(rw http.ResponseWriter, r *http.Request, authErr error),
+) http.Handler {
+	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
+		user, err := auth.AuthViaSession(rw, r)
+		if err != nil {
+			cclog.Infof("auth frontend api -> authentication failed: %s", err.Error())
+			onfailure(rw, r, err)
+			return
+		}
+		if user != nil {
+			ctx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+			onsuccess.ServeHTTP(rw, r.WithContext(ctx))
+			return
+		}
+		cclog.Info("auth frontend api -> authentication failed: no auth")
+		onfailure(rw, r, errors.New("unauthorized"))
+	})
+}
+
+func (auth *Authentication) Logout(onsuccess http.Handler) http.Handler {
 	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
 		session, err := auth.sessionStore.Get(r, "session")
 		if err != nil {
@@ -420,3 +592,42 @@ func (auth *Authentication) Logout(onsuccess http.Handler) http.Handler {
 		onsuccess.ServeHTTP(rw, r)
 	})
 }
+
+// Helper Moved To MiddleWare Auth Handlers
+func securedCheck(user *schema.User, r *http.Request) error {
+	if user == nil {
+		return fmt.Errorf("no user for secured check")
+	}
+
+	// extract IP address for checking
+	IPAddress := r.Header.Get("X-Real-Ip")
+	if IPAddress == "" {
+		IPAddress = r.Header.Get("X-Forwarded-For")
+	}
+	if IPAddress == "" {
+		IPAddress = r.RemoteAddr
+	}
+
+	// Handle both IPv4 and IPv6 addresses properly
+	// For IPv6, this will strip the port and brackets
+	// For IPv4, this will strip the port
+	if host, _, err := net.SplitHostPort(IPAddress); err == nil {
+		IPAddress = host
+	}
+	// If SplitHostPort fails, IPAddress is already just a host (no port)
+
+	// If nothing declared in config: deny all request to this api endpoint
+	if len(config.Keys.APIAllowedIPs) == 0 {
+		return fmt.Errorf("missing configuration key ApiAllowedIPs")
+	}
+	// If wildcard declared in config: Continue
+	if config.Keys.APIAllowedIPs[0] == "*" {
+		return nil
+	}
+	// check if IP is allowed
+	if !util.Contains(config.Keys.APIAllowedIPs, IPAddress) {
+		return fmt.Errorf("unknown ip: %v", IPAddress)
+	}
+
+	return nil
+}

internal/auth/auth_test.go

@@ -1,129 +1,176 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package auth
 
 import (
+	"net"
 	"testing"
+	"time"
 )
 
-func TestHasValidRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	exists, _ := u.HasValidRole("user")
-
-	if !exists {
-		t.Fatalf(`User{Roles: ["user"]} -> HasValidRole("user"): EXISTS = %v, expected 'true'.`, exists)
+// TestGetIPUserLimiter tests the rate limiter creation and retrieval
+func TestGetIPUserLimiter(t *testing.T) {
+	ip := "192.168.1.1"
+	username := "testuser"
+	
+	// Get limiter for the first time
+	limiter1 := getIPUserLimiter(ip, username)
+	if limiter1 == nil {
+		t.Fatal("Expected limiter to be created")
+	}
+	
+	// Get the same limiter again
+	limiter2 := getIPUserLimiter(ip, username)
+	if limiter1 != limiter2 {
+		t.Error("Expected to get the same limiter instance")
+	}
+	
+	// Get a different limiter for different user
+	limiter3 := getIPUserLimiter(ip, "otheruser")
+	if limiter1 == limiter3 {
+		t.Error("Expected different limiter for different user")
+	}
+	
+	// Get a different limiter for different IP
+	limiter4 := getIPUserLimiter("192.168.1.2", username)
+	if limiter1 == limiter4 {
+		t.Error("Expected different limiter for different IP")
 	}
 }
 
-func TestHasNotValidRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	exists, _ := u.HasValidRole("manager")
-
-	if exists {
-		t.Fatalf(`User{Roles: ["user"]} -> HasValidRole("manager"): EXISTS = %v, expected 'false'.`, exists)
+// TestRateLimiterBehavior tests that rate limiting works correctly
+func TestRateLimiterBehavior(t *testing.T) {
+	ip := "10.0.0.1"
+	username := "ratelimituser"
+	
+	limiter := getIPUserLimiter(ip, username)
+	
+	// Should allow first 5 attempts
+	for i := 0; i < 5; i++ {
+		if !limiter.Allow() {
+			t.Errorf("Request %d should be allowed within rate limit", i+1)
+		}
+	}
+	
+	// 6th attempt should be blocked
+	if limiter.Allow() {
+		t.Error("Request 6 should be blocked by rate limiter")
 	}
 }
 
-func TestHasInvalidRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	_, valid := u.HasValidRole("invalid")
-
-	if valid {
-		t.Fatalf(`User{Roles: ["user"]} -> HasValidRole("invalid"): VALID = %v, expected 'false'.`, valid)
+// TestCleanupOldRateLimiters tests the cleanup function
+func TestCleanupOldRateLimiters(t *testing.T) {
+	// Clear all existing limiters first to avoid interference from other tests
+	cleanupOldRateLimiters(time.Now().Add(24 * time.Hour))
+	
+	// Create some new rate limiters
+	limiter1 := getIPUserLimiter("1.1.1.1", "user1")
+	limiter2 := getIPUserLimiter("2.2.2.2", "user2")
+	
+	if limiter1 == nil || limiter2 == nil {
+		t.Fatal("Failed to create test limiters")
+	}
+	
+	// Cleanup limiters older than 1 second from now (should keep both)
+	time.Sleep(10 * time.Millisecond) // Small delay to ensure timestamp difference
+	cleanupOldRateLimiters(time.Now().Add(-1 * time.Second))
+	
+	// Verify they still exist (should get same instance)
+	if getIPUserLimiter("1.1.1.1", "user1") != limiter1 {
+		t.Error("Limiter 1 was incorrectly cleaned up")
+	}
+	if getIPUserLimiter("2.2.2.2", "user2") != limiter2 {
+		t.Error("Limiter 2 was incorrectly cleaned up")
+	}
+	
+	// Cleanup limiters older than 1 hour from now (should remove both)
+	cleanupOldRateLimiters(time.Now().Add(2 * time.Hour))
+	
+	// Getting them again should create new instances
+	newLimiter1 := getIPUserLimiter("1.1.1.1", "user1")
+	if newLimiter1 == limiter1 {
+		t.Error("Old limiter should have been cleaned up")
 	}
 }
 
-func TestHasNotInvalidRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	_, valid := u.HasValidRole("user")
-
-	if !valid {
-		t.Fatalf(`User{Roles: ["user"]} -> HasValidRole("user"): VALID = %v, expected 'true'.`, valid)
+// TestIPv4Extraction tests extracting IPv4 addresses
+func TestIPv4Extraction(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"IPv4 with port", "192.168.1.1:8080", "192.168.1.1"},
+		{"IPv4 without port", "192.168.1.1", "192.168.1.1"},
+		{"Localhost with port", "127.0.0.1:3000", "127.0.0.1"},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := tt.input
+			if host, _, err := net.SplitHostPort(result); err == nil {
+				result = host
+			}
+			
+			if result != tt.expected {
+				t.Errorf("Expected %s, got %s", tt.expected, result)
+			}
+		})
 	}
 }
 
-func TestHasRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	exists := u.HasRole(RoleUser)
-
-	if !exists {
-		t.Fatalf(`User{Roles: ["user"]} -> HasRole(RoleUser): EXISTS = %v, expected 'true'.`, exists)
+// TestIPv6Extraction tests extracting IPv6 addresses  
+func TestIPv6Extraction(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"IPv6 with port", "[2001:db8::1]:8080", "2001:db8::1"},
+		{"IPv6 localhost with port", "[::1]:3000", "::1"},
+		{"IPv6 without port", "2001:db8::1", "2001:db8::1"},
+		{"IPv6 localhost", "::1", "::1"},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := tt.input
+			if host, _, err := net.SplitHostPort(result); err == nil {
+				result = host
+			}
+			
+			if result != tt.expected {
+				t.Errorf("Expected %s, got %s", tt.expected, result)
+			}
+		})
 	}
 }
 
-func TestHasNotRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user"}}
-
-	exists := u.HasRole(RoleManager)
-
-	if exists {
-		t.Fatalf(`User{Roles: ["user"]} -> HasRole(RoleManager): EXISTS = %v, expected 'false'.`, exists)
-	}
-}
-
-func TestHasAnyRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager"}}
-
-	result := u.HasAnyRole([]Role{RoleManager, RoleSupport, RoleAdmin})
-
-	if !result {
-		t.Fatalf(`User{Roles: ["user", "manager"]} -> HasAnyRole([]Role{RoleManager, RoleSupport, RoleAdmin}): RESULT = %v, expected 'true'.`, result)
-	}
-}
-
-func TestHasNotAnyRole(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager"}}
-
-	result := u.HasAnyRole([]Role{RoleSupport, RoleAdmin})
-
-	if result {
-		t.Fatalf(`User{Roles: ["user", "manager"]} -> HasAllRoles([]Role{RoleSupport, RoleAdmin}): RESULT = %v, expected 'false'.`, result)
-	}
-}
-
-func TestHasAllRoles(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager", "support"}}
-
-	result := u.HasAllRoles([]Role{RoleUser, RoleManager, RoleSupport})
-
-	if !result {
-		t.Fatalf(`User{Roles: ["user", "manager", "support"]} -> HasAllRoles([]Role{RoleUser, RoleManager, RoleSupport}): RESULT = %v, expected 'true'.`, result)
-	}
-}
-
-func TestHasNotAllRoles(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager"}}
-
-	result := u.HasAllRoles([]Role{RoleUser, RoleManager, RoleSupport})
-
-	if result {
-		t.Fatalf(`User{Roles: ["user", "manager"]} -> HasAllRoles([]Role{RoleUser, RoleManager, RoleSupport}): RESULT = %v, expected 'false'.`, result)
-	}
-}
-
-func TestHasNotRoles(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager"}}
-
-	result := u.HasNotRoles([]Role{RoleSupport, RoleAdmin})
-
-	if !result {
-		t.Fatalf(`User{Roles: ["user", "manager"]} -> HasNotRoles([]Role{RoleSupport, RoleAdmin}): RESULT = %v, expected 'true'.`, result)
-	}
-}
-
-func TestHasAllNotRoles(t *testing.T) {
-	u := User{Username: "testuser", Roles: []string{"user", "manager"}}
-
-	result := u.HasNotRoles([]Role{RoleUser, RoleManager})
-
-	if result {
-		t.Fatalf(`User{Roles: ["user", "manager"]} -> HasNotRoles([]Role{RoleUser, RoleManager}): RESULT = %v, expected 'false'.`, result)
+// TestIPExtractionEdgeCases tests edge cases for IP extraction
+func TestIPExtractionEdgeCases(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"Hostname without port", "example.com", "example.com"},
+		{"Empty string", "", ""},
+		{"Just port", ":8080", ""},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := tt.input
+			if host, _, err := net.SplitHostPort(result); err == nil {
+				result = host
+			}
+			
+			if result != tt.expected {
+				t.Errorf("Expected %s, got %s", tt.expected, result)
+			}
+		})
 	}
 }

internal/auth/jwt.go

@@ -1,321 +1,125 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package auth
 
 import (
 	"crypto/ed25519"
-	"database/sql"
 	"encoding/base64"
 	"errors"
-	"fmt"
 	"net/http"
 	"os"
 	"strings"
 	"time"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
-	"github.com/golang-jwt/jwt/v4"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/golang-jwt/jwt/v5"
 )
 
-type JWTAuthenticator struct {
-	auth *Authentication
+type JWTAuthConfig struct {
+	// Specifies for how long a JWT token shall be valid
+	// as a string parsable by time.ParseDuration().
+	MaxAge string `json:"max-age"`
 
-	publicKey           ed25519.PublicKey
-	privateKey          ed25519.PrivateKey
-	publicKeyCrossLogin ed25519.PublicKey // For accepting externally generated JWTs
+	// Specifies which cookie should be checked for a JWT token (if no authorization header is present)
+	CookieName string `json:"cookieName"`
 
-	loginTokenKey []byte // HS256 key
+	// Deny login for users not in database (but defined in JWT).
+	// Ignore user roles defined in JWTs ('roles' claim), get them from db.
+	ValidateUser bool `json:"validateUser"`
 
-	config *schema.JWTAuthConfig
+	// Specifies which issuer should be accepted when validating external JWTs ('iss' claim)
+	TrustedIssuer string `json:"trustedIssuer"`
+
+	// Should an non-existent user be added to the DB based on the information in the token
+	SyncUserOnLogin bool `json:"syncUserOnLogin"`
+
+	// Should an existent user be updated in the DB based on the information in the token
+	UpdateUserOnLogin bool `json:"updateUserOnLogin"`
 }
 
-var _ Authenticator = (*JWTAuthenticator)(nil)
-
-func (ja *JWTAuthenticator) Init(auth *Authentication, conf interface{}) error {
-
-	ja.auth = auth
-	ja.config = conf.(*schema.JWTAuthConfig)
+type JWTAuthenticator struct {
+	publicKey  ed25519.PublicKey
+	privateKey ed25519.PrivateKey
+}
 
+func (ja *JWTAuthenticator) Init() error {
 	pubKey, privKey := os.Getenv("JWT_PUBLIC_KEY"), os.Getenv("JWT_PRIVATE_KEY")
 	if pubKey == "" || privKey == "" {
-		log.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
+		cclog.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
 	} else {
 		bytes, err := base64.StdEncoding.DecodeString(pubKey)
 		if err != nil {
-			log.Warn("Could not decode JWT public key")
+			cclog.Warn("Could not decode JWT public key")
 			return err
 		}
 		ja.publicKey = ed25519.PublicKey(bytes)
 		bytes, err = base64.StdEncoding.DecodeString(privKey)
 		if err != nil {
-			log.Warn("Could not decode JWT private key")
+			cclog.Warn("Could not decode JWT private key")
 			return err
 		}
 		ja.privateKey = ed25519.PrivateKey(bytes)
 	}
 
-	if pubKey = os.Getenv("CROSS_LOGIN_JWT_HS512_KEY"); pubKey != "" {
-		bytes, err := base64.StdEncoding.DecodeString(pubKey)
-		if err != nil {
-			log.Warn("Could not decode cross login JWT HS512 key")
-			return err
-		}
-		ja.loginTokenKey = bytes
-	}
-
-	// Look for external public keys
-	pubKeyCrossLogin, keyFound := os.LookupEnv("CROSS_LOGIN_JWT_PUBLIC_KEY")
-	if keyFound && pubKeyCrossLogin != "" {
-		bytes, err := base64.StdEncoding.DecodeString(pubKeyCrossLogin)
-		if err != nil {
-			log.Warn("Could not decode cross login JWT public key")
-			return err
-		}
-		ja.publicKeyCrossLogin = ed25519.PublicKey(bytes)
-
-		// Warn if other necessary settings are not configured
-		if ja.config != nil {
-			if ja.config.CookieName == "" {
-				log.Warn("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
-			}
-			if !ja.config.ForceJWTValidationViaDatabase {
-				log.Warn("forceJWTValidationViaDatabase not set to true: CC will accept users and roles defined in JWTs regardless of its own database!")
-			}
-			if ja.config.TrustedExternalIssuer == "" {
-				log.Warn("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
-			}
-		} else {
-			log.Warn("cookieName and trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
-		}
-	} else {
-		ja.publicKeyCrossLogin = nil
-		log.Warn("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
-	}
-
 	return nil
 }
 
-func (ja *JWTAuthenticator) CanLogin(
-	user *User,
+func (ja *JWTAuthenticator) AuthViaJWT(
 	rw http.ResponseWriter,
-	r *http.Request) bool {
-
-	return (user != nil && user.AuthSource == AuthViaToken) || r.Header.Get("Authorization") != "" || r.URL.Query().Get("login-token") != ""
-}
-
-func (ja *JWTAuthenticator) Login(
-	user *User,
-	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
-	rawtoken := r.Header.Get("X-Auth-Token")
-	if rawtoken == "" {
-		rawtoken = r.Header.Get("Authorization")
-		rawtoken = strings.TrimPrefix(rawtoken, "Bearer ")
-		if rawtoken == "" {
-			rawtoken = r.URL.Query().Get("login-token")
-		}
-	}
-
-	token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (interface{}, error) {
-		if t.Method == jwt.SigningMethodEdDSA {
-			return ja.publicKey, nil
-		}
-		if t.Method == jwt.SigningMethodHS256 || t.Method == jwt.SigningMethodHS512 {
-			return ja.loginTokenKey, nil
-		}
-		return nil, fmt.Errorf("AUTH/JWT > unkown signing method for login token: %s (known: HS256, HS512, EdDSA)", t.Method.Alg())
-	})
-	if err != nil {
-		log.Warn("Error while parsing jwt token")
-		return nil, err
-	}
-
-	if err := token.Claims.Valid(); err != nil {
-		log.Warn("jwt token claims are not valid")
-		return nil, err
-	}
-
-	claims := token.Claims.(jwt.MapClaims)
-	sub, _ := claims["sub"].(string)
-	exp, _ := claims["exp"].(float64)
-	var roles []string
-	if rawroles, ok := claims["roles"].([]interface{}); ok {
-		for _, rr := range rawroles {
-			if r, ok := rr.(string); ok {
-				if isValidRole(r) {
-					roles = append(roles, r)
-				}
-			}
-		}
-	}
-	if rawrole, ok := claims["roles"].(string); ok {
-		if isValidRole(rawrole) {
-			roles = append(roles, rawrole)
-		}
-	}
-
-	if user == nil {
-		user, err = ja.auth.GetUser(sub)
-		if err != nil && err != sql.ErrNoRows {
-			log.Errorf("Error while loading user '%v'", sub)
-			return nil, err
-		} else if user == nil {
-			user = &User{
-				Username:   sub,
-				Roles:      roles,
-				AuthSource: AuthViaToken,
-			}
-			if err := ja.auth.AddUser(user); err != nil {
-				log.Errorf("Error while adding user '%v' to auth from token", user.Username)
-				return nil, err
-			}
-		}
-	}
-
-	user.Expiration = time.Unix(int64(exp), 0)
-	return user, nil
-}
-
-func (ja *JWTAuthenticator) Auth(
-	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
+	r *http.Request,
+) (*schema.User, error) {
 	rawtoken := r.Header.Get("X-Auth-Token")
 	if rawtoken == "" {
 		rawtoken = r.Header.Get("Authorization")
 		rawtoken = strings.TrimPrefix(rawtoken, "Bearer ")
 	}
 
-	// If no auth header was found, check for a certain cookie containing a JWT
-	cookieName := ""
-	cookieFound := false
-	if ja.config != nil && ja.config.CookieName != "" {
-		cookieName = ja.config.CookieName
-	}
-
-	// Try to read the JWT cookie
-	if rawtoken == "" && cookieName != "" {
-		jwtCookie, err := r.Cookie(cookieName)
-
-		if err == nil && jwtCookie.Value != "" {
-			rawtoken = jwtCookie.Value
-			cookieFound = true
-		}
-	}
-
-	// Because a user can also log in via a token, the
-	// session cookie must be checked here as well:
+	// there is no token
 	if rawtoken == "" {
-		return ja.auth.AuthViaSession(rw, r)
+		return nil, nil
 	}
 
-	// Try to parse JWT
-	token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (interface{}, error) {
+	token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
 		if t.Method != jwt.SigningMethodEdDSA {
 			return nil, errors.New("only Ed25519/EdDSA supported")
 		}
 
-		// Is there more than one public key?
-		if ja.publicKeyCrossLogin != nil && ja.config != nil && ja.config.TrustedExternalIssuer != "" {
-			// Determine whether to use the external public key
-			unvalidatedIssuer, success := t.Claims.(jwt.MapClaims)["iss"].(string)
-			if success && unvalidatedIssuer == ja.config.TrustedExternalIssuer {
-				// The (unvalidated) issuer seems to be the expected one,
-				// use public cross login key from config
-				return ja.publicKeyCrossLogin, nil
-			}
-		}
-
-		// No cross login key configured or issuer not expected
-		// Try own key
 		return ja.publicKey, nil
 	})
 	if err != nil {
-		log.Warn("Error while parsing token")
+		cclog.Warn("Error while parsing JWT token")
 		return nil, err
 	}
-
-	// Check token validity
-	if err := token.Claims.Valid(); err != nil {
-		log.Warn("jwt token claims are not valid")
-		return nil, err
+	if !token.Valid {
+		cclog.Warn("jwt token claims are not valid")
+		return nil, errors.New("jwt token claims are not valid")
 	}
 
 	// Token is valid, extract payload
 	claims := token.Claims.(jwt.MapClaims)
-	sub, _ := claims["sub"].(string)
-
-	var roles []string
-
-	// Validate user + roles from JWT against database?
-	if ja.config != nil && ja.config.ForceJWTValidationViaDatabase {
-		user, err := ja.auth.GetUser(sub)
-
-		// Deny any logins for unknown usernames
-		if err != nil {
-			log.Warn("Could not find user from JWT in internal database.")
-			return nil, errors.New("unknown user")
-		}
-
-		// Take user roles from database instead of trusting the JWT
-		roles = user.Roles
-	} else {
-		// Extract roles from JWT (if present)
-		if rawroles, ok := claims["roles"].([]interface{}); ok {
-			for _, rr := range rawroles {
-				if r, ok := rr.(string); ok {
-					roles = append(roles, r)
-				}
-			}
-		}
+	
+	// Use shared helper to get user from JWT claims
+	var user *schema.User
+	user, err = getUserFromJWT(claims, Keys.JwtConfig.ValidateUser, schema.AuthToken, -1)
+	if err != nil {
+		return nil, err
 	}
-
-	if cookieFound {
-		// Create a session so that we no longer need the JTW Cookie
-		session, err := ja.auth.sessionStore.New(r, "session")
-		if err != nil {
-			log.Errorf("session creation failed: %s", err.Error())
-			http.Error(rw, err.Error(), http.StatusInternalServerError)
-			return nil, err
-		}
-
-		if ja.auth.SessionMaxAge != 0 {
-			session.Options.MaxAge = int(ja.auth.SessionMaxAge.Seconds())
-		}
-		session.Values["username"] = sub
-		session.Values["roles"] = roles
-
-		if err := ja.auth.sessionStore.Save(r, rw, session); err != nil {
-			log.Warnf("session save failed: %s", err.Error())
-			http.Error(rw, err.Error(), http.StatusInternalServerError)
-			return nil, err
-		}
-
-		// (Ask browser to) Delete JWT cookie
-		deletedCookie := &http.Cookie{
-			Name:     cookieName,
-			Value:    "",
-			Path:     "/",
-			MaxAge:   -1,
-			HttpOnly: true,
-		}
-		http.SetCookie(rw, deletedCookie)
+	
+	// If not validating user, we only get roles from JWT (no projects for this auth method)
+	if !Keys.JwtConfig.ValidateUser {
+		user.Roles = extractRolesFromClaims(claims, false)
+		user.Projects = nil // Standard JWT auth doesn't include projects
 	}
-
-	return &User{
-		Username:   sub,
-		Roles:      roles,
-		AuthSource: AuthViaToken,
-	}, nil
+	
+	return user, nil
 }
 
-// Generate a new JWT that can be used for authentication
-func (ja *JWTAuthenticator) ProvideJWT(user *User) (string, error) {
-
+// ProvideJWT generates a new JWT that can be used for authentication
+func (ja *JWTAuthenticator) ProvideJWT(user *schema.User) (string, error) {
 	if ja.privateKey == nil {
 		return "", errors.New("environment variable 'JWT_PRIVATE_KEY' not set")
 	}
@@ -326,8 +130,12 @@ func (ja *JWTAuthenticator) ProvideJWT(user *User) (string, error) {
 		"roles": user.Roles,
 		"iat":   now.Unix(),
 	}
-	if ja.config != nil && ja.config.MaxAge != 0 {
-		claims["exp"] = now.Add(time.Duration(ja.config.MaxAge)).Unix()
+	if Keys.JwtConfig.MaxAge != "" {
+		d, err := time.ParseDuration(Keys.JwtConfig.MaxAge)
+		if err != nil {
+			return "", errors.New("cannot parse max-age config key")
+		}
+		claims["exp"] = now.Add(d).Unix()
 	}
 
 	return jwt.NewWithClaims(jwt.SigningMethodEdDSA, claims).SignedString(ja.privateKey)

internal/auth/jwtCookieSession.go

@@ -0,0 +1,172 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package auth
+
+import (
+	"crypto/ed25519"
+	"encoding/base64"
+	"errors"
+	"net/http"
+	"os"
+
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/golang-jwt/jwt/v5"
+)
+
+type JWTCookieSessionAuthenticator struct {
+	publicKey           ed25519.PublicKey
+	privateKey          ed25519.PrivateKey
+	publicKeyCrossLogin ed25519.PublicKey // For accepting externally generated JWTs
+}
+
+var _ Authenticator = (*JWTCookieSessionAuthenticator)(nil)
+
+func (ja *JWTCookieSessionAuthenticator) Init() error {
+	pubKey, privKey := os.Getenv("JWT_PUBLIC_KEY"), os.Getenv("JWT_PRIVATE_KEY")
+	if pubKey == "" || privKey == "" {
+		cclog.Warn("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
+		return errors.New("environment variables 'JWT_PUBLIC_KEY' or 'JWT_PRIVATE_KEY' not set (token based authentication will not work)")
+	} else {
+		bytes, err := base64.StdEncoding.DecodeString(pubKey)
+		if err != nil {
+			cclog.Warn("Could not decode JWT public key")
+			return err
+		}
+		ja.publicKey = ed25519.PublicKey(bytes)
+		bytes, err = base64.StdEncoding.DecodeString(privKey)
+		if err != nil {
+			cclog.Warn("Could not decode JWT private key")
+			return err
+		}
+		ja.privateKey = ed25519.PrivateKey(bytes)
+	}
+
+	// Look for external public keys
+	pubKeyCrossLogin, keyFound := os.LookupEnv("CROSS_LOGIN_JWT_PUBLIC_KEY")
+	if keyFound && pubKeyCrossLogin != "" {
+		bytes, err := base64.StdEncoding.DecodeString(pubKeyCrossLogin)
+		if err != nil {
+			cclog.Warn("Could not decode cross login JWT public key")
+			return err
+		}
+		ja.publicKeyCrossLogin = ed25519.PublicKey(bytes)
+	} else {
+		ja.publicKeyCrossLogin = nil
+		cclog.Debug("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
+		return errors.New("environment variable 'CROSS_LOGIN_JWT_PUBLIC_KEY' not set (cross login token based authentication will not work)")
+	}
+
+	// Warn if other necessary settings are not configured
+	if Keys.JwtConfig != nil {
+		if Keys.JwtConfig.CookieName == "" {
+			cclog.Info("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
+			return errors.New("cookieName for JWTs not configured (cross login via JWT cookie will fail)")
+		}
+		if !Keys.JwtConfig.ValidateUser {
+			cclog.Info("forceJWTValidationViaDatabase not set to true: CC will accept users and roles defined in JWTs regardless of its own database!")
+		}
+		if Keys.JwtConfig.TrustedIssuer == "" {
+			cclog.Info("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
+			return errors.New("trustedExternalIssuer for JWTs not configured (cross login via JWT cookie will fail)")
+		}
+	} else {
+		cclog.Warn("config for JWTs not configured (cross login via JWT cookie will fail)")
+		return errors.New("config for JWTs not configured (cross login via JWT cookie will fail)")
+	}
+
+	cclog.Info("JWT Cookie Session authenticator successfully registered")
+	return nil
+}
+
+func (ja *JWTCookieSessionAuthenticator) CanLogin(
+	user *schema.User,
+	username string,
+	rw http.ResponseWriter,
+	r *http.Request,
+) (*schema.User, bool) {
+	jc := Keys.JwtConfig
+	cookieName := ""
+	if jc.CookieName != "" {
+		cookieName = jc.CookieName
+	}
+
+	// Try to read the JWT cookie
+	if cookieName != "" {
+		jwtCookie, err := r.Cookie(cookieName)
+
+		if err == nil && jwtCookie.Value != "" {
+			return user, true
+		}
+	}
+
+	return nil, false
+}
+
+func (ja *JWTCookieSessionAuthenticator) Login(
+	user *schema.User,
+	rw http.ResponseWriter,
+	r *http.Request,
+) (*schema.User, error) {
+	jc := Keys.JwtConfig
+	jwtCookie, err := r.Cookie(jc.CookieName)
+	var rawtoken string
+
+	if err == nil && jwtCookie.Value != "" {
+		rawtoken = jwtCookie.Value
+	}
+
+	token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
+		if t.Method != jwt.SigningMethodEdDSA {
+			return nil, errors.New("only Ed25519/EdDSA supported")
+		}
+
+		unvalidatedIssuer, success := t.Claims.(jwt.MapClaims)["iss"].(string)
+		if success && unvalidatedIssuer == jc.TrustedIssuer {
+			// The (unvalidated) issuer seems to be the expected one,
+			// use public cross login key from config
+			return ja.publicKeyCrossLogin, nil
+		}
+
+		// No cross login key configured or issuer not expected
+		// Try own key
+		return ja.publicKey, nil
+	})
+	if err != nil {
+		cclog.Warn("JWT cookie session: error while parsing token")
+		return nil, err
+	}
+
+	if !token.Valid {
+		cclog.Warn("jwt token claims are not valid")
+		return nil, errors.New("jwt token claims are not valid")
+	}
+
+	claims := token.Claims.(jwt.MapClaims)
+	
+	// Use shared helper to get user from JWT claims
+	user, err = getUserFromJWT(claims, jc.ValidateUser, schema.AuthSession, schema.AuthViaToken)
+	if err != nil {
+		return nil, err
+	}
+	
+	// Sync or update user if configured
+	if !jc.ValidateUser && (jc.SyncUserOnLogin || jc.UpdateUserOnLogin) {
+		handleTokenUser(user)
+	}
+
+	// (Ask browser to) Delete JWT cookie
+	deletedCookie := &http.Cookie{
+		Name:     jc.CookieName,
+		Value:    "",
+		Path:     "/",
+		MaxAge:   -1,
+		HttpOnly: true,
+	}
+	http.SetCookie(rw, deletedCookie)
+
+	return user, nil
+}

internal/auth/jwtHelpers.go

@@ -0,0 +1,136 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package auth
+
+import (
+	"database/sql"
+	"errors"
+	"fmt"
+
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/golang-jwt/jwt/v5"
+)
+
+// extractStringFromClaims extracts a string value from JWT claims
+func extractStringFromClaims(claims jwt.MapClaims, key string) string {
+	if val, ok := claims[key].(string); ok {
+		return val
+	}
+	return ""
+}
+
+// extractRolesFromClaims extracts roles from JWT claims
+// If validateRoles is true, only valid roles are returned
+func extractRolesFromClaims(claims jwt.MapClaims, validateRoles bool) []string {
+	var roles []string
+	
+	if rawroles, ok := claims["roles"].([]any); ok {
+		for _, rr := range rawroles {
+			if r, ok := rr.(string); ok {
+				if validateRoles {
+					if schema.IsValidRole(r) {
+						roles = append(roles, r)
+					}
+				} else {
+					roles = append(roles, r)
+				}
+			}
+		}
+	}
+	
+	return roles
+}
+
+// extractProjectsFromClaims extracts projects from JWT claims
+func extractProjectsFromClaims(claims jwt.MapClaims) []string {
+	projects := make([]string, 0)
+	
+	if rawprojs, ok := claims["projects"].([]any); ok {
+		for _, pp := range rawprojs {
+			if p, ok := pp.(string); ok {
+				projects = append(projects, p)
+			}
+		}
+	} else if rawprojs, ok := claims["projects"]; ok {
+		if projSlice, ok := rawprojs.([]string); ok {
+			projects = append(projects, projSlice...)
+		}
+	}
+	
+	return projects
+}
+
+// extractNameFromClaims extracts name from JWT claims
+// Handles both simple string and complex nested structure
+func extractNameFromClaims(claims jwt.MapClaims) string {
+	// Try simple string first
+	if name, ok := claims["name"].(string); ok {
+		return name
+	}
+	
+	// Try nested structure: {name: {values: [...]}}
+	if wrap, ok := claims["name"].(map[string]any); ok {
+		if vals, ok := wrap["values"].([]any); ok {
+			if len(vals) == 0 {
+				return ""
+			}
+			
+			name := fmt.Sprintf("%v", vals[0])
+			for i := 1; i < len(vals); i++ {
+				name += fmt.Sprintf(" %v", vals[i])
+			}
+			return name
+		}
+	}
+	
+	return ""
+}
+
+// getUserFromJWT creates or retrieves a user based on JWT claims
+// If validateUser is true, the user must exist in the database
+// Otherwise, a new user object is created from claims
+// authSource should be a schema.AuthSource constant (like schema.AuthViaToken)
+func getUserFromJWT(claims jwt.MapClaims, validateUser bool, authType schema.AuthType, authSource schema.AuthSource) (*schema.User, error) {
+	sub := extractStringFromClaims(claims, "sub")
+	if sub == "" {
+		return nil, errors.New("missing 'sub' claim in JWT")
+	}
+	
+	if validateUser {
+		// Validate user against database
+		ur := repository.GetUserRepository()
+		user, err := ur.GetUser(sub)
+		if err != nil && err != sql.ErrNoRows {
+			cclog.Errorf("Error while loading user '%v': %v", sub, err)
+			return nil, fmt.Errorf("database error: %w", err)
+		}
+		
+		// Deny any logins for unknown usernames
+		if user == nil || err == sql.ErrNoRows {
+			cclog.Warn("Could not find user from JWT in internal database.")
+			return nil, errors.New("unknown user")
+		}
+		
+		// Return database user (with database roles)
+		return user, nil
+	}
+	
+	// Create user from JWT claims
+	name := extractNameFromClaims(claims)
+	roles := extractRolesFromClaims(claims, true) // Validate roles
+	projects := extractProjectsFromClaims(claims)
+	
+	return &schema.User{
+		Username:   sub,
+		Name:       name,
+		Roles:      roles,
+		Projects:   projects,
+		AuthType:   authType,
+		AuthSource: authSource,
+	}, nil
+}

internal/auth/jwtHelpers_test.go

@@ -0,0 +1,281 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package auth
+
+import (
+	"testing"
+
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/golang-jwt/jwt/v5"
+)
+
+// TestExtractStringFromClaims tests extracting string values from JWT claims
+func TestExtractStringFromClaims(t *testing.T) {
+	claims := jwt.MapClaims{
+		"sub":   "testuser",
+		"email": "test@example.com",
+		"age":   25, // not a string
+	}
+	
+	tests := []struct {
+		name     string
+		key      string
+		expected string
+	}{
+		{"Existing string", "sub", "testuser"},
+		{"Another string", "email", "test@example.com"},
+		{"Non-existent key", "missing", ""},
+		{"Non-string value", "age", ""},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := extractStringFromClaims(claims, tt.key)
+			if result != tt.expected {
+				t.Errorf("Expected %s, got %s", tt.expected, result)
+			}
+		})
+	}
+}
+
+// TestExtractRolesFromClaims tests role extraction and validation
+func TestExtractRolesFromClaims(t *testing.T) {
+	tests := []struct {
+		name          string
+		claims        jwt.MapClaims
+		validateRoles bool
+		expected      []string
+	}{
+		{
+			name: "Valid roles without validation",
+			claims: jwt.MapClaims{
+				"roles": []any{"admin", "user", "invalid_role"},
+			},
+			validateRoles: false,
+			expected:      []string{"admin", "user", "invalid_role"},
+		},
+		{
+			name: "Valid roles with validation",
+			claims: jwt.MapClaims{
+				"roles": []any{"admin", "user", "api"},
+			},
+			validateRoles: true,
+			expected:      []string{"admin", "user", "api"},
+		},
+		{
+			name: "Invalid roles with validation",
+			claims: jwt.MapClaims{
+				"roles": []any{"invalid_role", "fake_role"},
+			},
+			validateRoles: true,
+			expected:      []string{}, // Should filter out invalid roles
+		},
+		{
+			name:          "No roles claim",
+			claims:        jwt.MapClaims{},
+			validateRoles: false,
+			expected:      []string{},
+		},
+		{
+			name: "Non-array roles",
+			claims: jwt.MapClaims{
+				"roles": "admin",
+			},
+			validateRoles: false,
+			expected:      []string{},
+		},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := extractRolesFromClaims(tt.claims, tt.validateRoles)
+			
+			if len(result) != len(tt.expected) {
+				t.Errorf("Expected %d roles, got %d", len(tt.expected), len(result))
+				return
+			}
+			
+			for i, role := range result {
+				if i >= len(tt.expected) || role != tt.expected[i] {
+					t.Errorf("Expected role %s at position %d, got %s", tt.expected[i], i, role)
+				}
+			}
+		})
+	}
+}
+
+// TestExtractProjectsFromClaims tests project extraction from claims
+func TestExtractProjectsFromClaims(t *testing.T) {
+	tests := []struct {
+		name     string
+		claims   jwt.MapClaims
+		expected []string
+	}{
+		{
+			name: "Projects as array of interfaces",
+			claims: jwt.MapClaims{
+				"projects": []any{"project1", "project2", "project3"},
+			},
+			expected: []string{"project1", "project2", "project3"},
+		},
+		{
+			name: "Projects as string array",
+			claims: jwt.MapClaims{
+				"projects": []string{"projectA", "projectB"},
+			},
+			expected: []string{"projectA", "projectB"},
+		},
+		{
+			name:     "No projects claim",
+			claims:   jwt.MapClaims{},
+			expected: []string{},
+		},
+		{
+			name: "Mixed types in projects array",
+			claims: jwt.MapClaims{
+				"projects": []any{"project1", 123, "project2"},
+			},
+			expected: []string{"project1", "project2"}, // Should skip non-strings
+		},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := extractProjectsFromClaims(tt.claims)
+			
+			if len(result) != len(tt.expected) {
+				t.Errorf("Expected %d projects, got %d", len(tt.expected), len(result))
+				return
+			}
+			
+			for i, project := range result {
+				if i >= len(tt.expected) || project != tt.expected[i] {
+					t.Errorf("Expected project %s at position %d, got %s", tt.expected[i], i, project)
+				}
+			}
+		})
+	}
+}
+
+// TestExtractNameFromClaims tests name extraction from various formats
+func TestExtractNameFromClaims(t *testing.T) {
+	tests := []struct {
+		name     string
+		claims   jwt.MapClaims
+		expected string
+	}{
+		{
+			name: "Simple string name",
+			claims: jwt.MapClaims{
+				"name": "John Doe",
+			},
+			expected: "John Doe",
+		},
+		{
+			name: "Nested name structure",
+			claims: jwt.MapClaims{
+				"name": map[string]any{
+					"values": []any{"John", "Doe"},
+				},
+			},
+			expected: "John Doe",
+		},
+		{
+			name: "Nested name with single value",
+			claims: jwt.MapClaims{
+				"name": map[string]any{
+					"values": []any{"Alice"},
+				},
+			},
+			expected: "Alice",
+		},
+		{
+			name:     "No name claim",
+			claims:   jwt.MapClaims{},
+			expected: "",
+		},
+		{
+			name: "Empty nested values",
+			claims: jwt.MapClaims{
+				"name": map[string]any{
+					"values": []any{},
+				},
+			},
+			expected: "",
+		},
+		{
+			name: "Nested with non-string values",
+			claims: jwt.MapClaims{
+				"name": map[string]any{
+					"values": []any{123, "Smith"},
+				},
+			},
+			expected: "123 Smith", // Should convert to string
+		},
+	}
+	
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := extractNameFromClaims(tt.claims)
+			if result != tt.expected {
+				t.Errorf("Expected '%s', got '%s'", tt.expected, result)
+			}
+		})
+	}
+}
+
+// TestGetUserFromJWT_NoValidation tests getUserFromJWT without database validation
+func TestGetUserFromJWT_NoValidation(t *testing.T) {
+	claims := jwt.MapClaims{
+		"sub":      "testuser",
+		"name":     "Test User",
+		"roles":    []any{"user", "admin"},
+		"projects": []any{"project1", "project2"},
+	}
+	
+	user, err := getUserFromJWT(claims, false, schema.AuthToken, -1)
+	
+	if err != nil {
+		t.Fatalf("Unexpected error: %v", err)
+	}
+	
+	if user.Username != "testuser" {
+		t.Errorf("Expected username 'testuser', got '%s'", user.Username)
+	}
+	
+	if user.Name != "Test User" {
+		t.Errorf("Expected name 'Test User', got '%s'", user.Name)
+	}
+	
+	if len(user.Roles) != 2 {
+		t.Errorf("Expected 2 roles, got %d", len(user.Roles))
+	}
+	
+	if len(user.Projects) != 2 {
+		t.Errorf("Expected 2 projects, got %d", len(user.Projects))
+	}
+	
+	if user.AuthType != schema.AuthToken {
+		t.Errorf("Expected AuthType %v, got %v", schema.AuthToken, user.AuthType)
+	}
+}
+
+// TestGetUserFromJWT_MissingSub tests error when sub claim is missing
+func TestGetUserFromJWT_MissingSub(t *testing.T) {
+	claims := jwt.MapClaims{
+		"name": "Test User",
+	}
+	
+	_, err := getUserFromJWT(claims, false, schema.AuthToken, -1)
+	
+	if err == nil {
+		t.Error("Expected error for missing sub claim")
+	}
+	
+	if err.Error() != "missing 'sub' claim in JWT" {
+		t.Errorf("Expected specific error message, got: %v", err)
+	}
+}

internal/auth/jwtSession.go

@@ -0,0 +1,91 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package auth
+
+import (
+	"encoding/base64"
+	"errors"
+	"fmt"
+	"net/http"
+	"os"
+	"strings"
+
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/golang-jwt/jwt/v5"
+)
+
+type JWTSessionAuthenticator struct {
+	loginTokenKey []byte // HS256 key
+}
+
+var _ Authenticator = (*JWTSessionAuthenticator)(nil)
+
+func (ja *JWTSessionAuthenticator) Init() error {
+	if pubKey := os.Getenv("CROSS_LOGIN_JWT_HS512_KEY"); pubKey != "" {
+		bytes, err := base64.StdEncoding.DecodeString(pubKey)
+		if err != nil {
+			cclog.Warn("Could not decode cross login JWT HS512 key")
+			return err
+		}
+		ja.loginTokenKey = bytes
+	}
+
+	cclog.Info("JWT Session authenticator successfully registered")
+	return nil
+}
+
+func (ja *JWTSessionAuthenticator) CanLogin(
+	user *schema.User,
+	username string,
+	rw http.ResponseWriter,
+	r *http.Request,
+) (*schema.User, bool) {
+	return user, r.Header.Get("Authorization") != "" ||
+		r.URL.Query().Get("login-token") != ""
+}
+
+func (ja *JWTSessionAuthenticator) Login(
+	user *schema.User,
+	rw http.ResponseWriter,
+	r *http.Request,
+) (*schema.User, error) {
+	rawtoken := strings.TrimPrefix(r.Header.Get("Authorization"), "Bearer ")
+	if rawtoken == "" {
+		rawtoken = r.URL.Query().Get("login-token")
+	}
+
+	token, err := jwt.Parse(rawtoken, func(t *jwt.Token) (any, error) {
+		if t.Method == jwt.SigningMethodHS256 || t.Method == jwt.SigningMethodHS512 {
+			return ja.loginTokenKey, nil
+		}
+		return nil, fmt.Errorf("unkown signing method for login token: %s (known: HS256, HS512, EdDSA)", t.Method.Alg())
+	})
+	if err != nil {
+		cclog.Warn("Error while parsing jwt token")
+		return nil, err
+	}
+
+	if !token.Valid {
+		cclog.Warn("jwt token claims are not valid")
+		return nil, errors.New("jwt token claims are not valid")
+	}
+
+	claims := token.Claims.(jwt.MapClaims)
+	
+	// Use shared helper to get user from JWT claims
+	user, err = getUserFromJWT(claims, Keys.JwtConfig.ValidateUser, schema.AuthSession, schema.AuthViaToken)
+	if err != nil {
+		return nil, err
+	}
+	
+	// Sync or update user if configured
+	if !Keys.JwtConfig.ValidateUser && (Keys.JwtConfig.SyncUserOnLogin || Keys.JwtConfig.UpdateUserOnLogin) {
+		handleTokenUser(user)
+	}
+
+	return user, nil
+}

internal/auth/ldap.go

@@ -1,7 +1,8 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package auth
 
 import (
@@ -10,131 +11,168 @@ import (
 	"net/http"
 	"os"
 	"strings"
-	"time"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 	"github.com/go-ldap/ldap/v3"
 )
 
+type LdapConfig struct {
+	URL             string `json:"url"`
+	UserBase        string `json:"user_base"`
+	SearchDN        string `json:"search_dn"`
+	UserBind        string `json:"user_bind"`
+	UserFilter      string `json:"user_filter"`
+	UserAttr        string `json:"username_attr"`
+	SyncInterval    string `json:"sync_interval"` // Parsed using time.ParseDuration.
+	SyncDelOldUsers bool   `json:"sync_del_old_users"`
+
+	// Should an non-existent user be added to the DB if user exists in ldap directory
+	SyncUserOnLogin bool `json:"syncUserOnLogin"`
+}
+
 type LdapAuthenticator struct {
-	auth         *Authentication
-	config       *schema.LdapConfig
 	syncPassword string
+	UserAttr     string
 }
 
 var _ Authenticator = (*LdapAuthenticator)(nil)
 
-func (la *LdapAuthenticator) Init(
-	auth *Authentication,
-	conf interface{}) error {
-
-	la.auth = auth
-	la.config = conf.(*schema.LdapConfig)
-
+func (la *LdapAuthenticator) Init() error {
 	la.syncPassword = os.Getenv("LDAP_ADMIN_PASSWORD")
 	if la.syncPassword == "" {
-		log.Warn("environment variable 'LDAP_ADMIN_PASSWORD' not set (ldap sync will not work)")
+		cclog.Warn("environment variable 'LDAP_ADMIN_PASSWORD' not set (ldap sync will not work)")
 	}
 
-	if la.config != nil && la.config.SyncInterval != "" {
-		interval, err := time.ParseDuration(la.config.SyncInterval)
-		if err != nil {
-			log.Warnf("Could not parse duration for sync interval: %v", la.config.SyncInterval)
-			return err
-		}
-
-		if interval == 0 {
-			log.Info("Sync interval is zero")
-			return nil
-		}
-
-		go func() {
-			ticker := time.NewTicker(interval)
-			for t := range ticker.C {
-				log.Printf("sync started at %s", t.Format(time.RFC3339))
-				if err := la.Sync(); err != nil {
-					log.Errorf("sync failed: %s", err.Error())
-				}
-				log.Print("sync done")
-			}
-		}()
+	if Keys.LdapConfig.UserAttr != "" {
+		la.UserAttr = Keys.LdapConfig.UserAttr
+	} else {
+		la.UserAttr = "gecos"
 	}
 
 	return nil
 }
 
 func (la *LdapAuthenticator) CanLogin(
-	user *User,
+	user *schema.User,
+	username string,
 	rw http.ResponseWriter,
-	r *http.Request) bool {
+	r *http.Request,
+) (*schema.User, bool) {
+	lc := Keys.LdapConfig
 
-	return user != nil && user.AuthSource == AuthViaLDAP
+	if user != nil {
+		if user.AuthSource == schema.AuthViaLDAP {
+			return user, true
+		}
+	} else {
+		if lc.SyncUserOnLogin {
+			l, err := la.getLdapConnection(true)
+			if err != nil {
+				cclog.Error("LDAP connection error")
+				return nil, false
+			}
+			defer l.Close()
+
+			// Search for the given username
+			searchRequest := ldap.NewSearchRequest(
+				lc.UserBase,
+				ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
+				fmt.Sprintf("(&%s(uid=%s))", lc.UserFilter, username),
+				[]string{"dn", "uid", la.UserAttr}, nil)
+
+			sr, err := l.Search(searchRequest)
+			if err != nil {
+				cclog.Warn(err)
+				return nil, false
+			}
+
+			if len(sr.Entries) != 1 {
+				cclog.Warn("LDAP: User does not exist or too many entries returned")
+				return nil, false
+			}
+
+			entry := sr.Entries[0]
+			name := entry.GetAttributeValue(la.UserAttr)
+			var roles []string
+			roles = append(roles, schema.GetRoleString(schema.RoleUser))
+			projects := make([]string, 0)
+
+			user = &schema.User{
+				Username:   username,
+				Name:       name,
+				Roles:      roles,
+				Projects:   projects,
+				AuthType:   schema.AuthSession,
+				AuthSource: schema.AuthViaLDAP,
+			}
+
+			if err := repository.GetUserRepository().AddUser(user); err != nil {
+				cclog.Errorf("User '%s' LDAP: Insert into DB failed", username)
+				return nil, false
+			}
+
+			return user, true
+		}
+	}
+
+	return nil, false
 }
 
 func (la *LdapAuthenticator) Login(
-	user *User,
+	user *schema.User,
 	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
+	r *http.Request,
+) (*schema.User, error) {
 	l, err := la.getLdapConnection(false)
 	if err != nil {
-		log.Warn("Error while getting ldap connection")
+		cclog.Warn("Error while getting ldap connection")
 		return nil, err
 	}
 	defer l.Close()
 
-	userDn := strings.Replace(la.config.UserBind, "{username}", user.Username, -1)
+	userDn := strings.ReplaceAll(Keys.LdapConfig.UserBind, "{username}", user.Username)
 	if err := l.Bind(userDn, r.FormValue("password")); err != nil {
-		log.Errorf("AUTH/LOCAL > Authentication for user %s failed: %v", user.Username, err)
-		return nil, fmt.Errorf("AUTH/LDAP > Authentication failed")
+		cclog.Errorf("AUTH/LDAP > Authentication for user %s failed: %v",
+			user.Username, err)
+		return nil, fmt.Errorf("Authentication failed")
 	}
 
 	return user, nil
 }
 
-func (la *LdapAuthenticator) Auth(
-	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
-	return la.auth.AuthViaSession(rw, r)
-}
-
 func (la *LdapAuthenticator) Sync() error {
-
-	const IN_DB int = 1
-	const IN_LDAP int = 2
-	const IN_BOTH int = 3
+	const InDB int = 1
+	const InLdap int = 2
+	const InBoth int = 3
+	ur := repository.GetUserRepository()
+	lc := Keys.LdapConfig
 
 	users := map[string]int{}
-	rows, err := la.auth.db.Query(`SELECT username FROM user WHERE user.ldap = 1`)
+	usernames, err := ur.GetLdapUsernames()
 	if err != nil {
-		log.Warn("Error while querying LDAP users")
 		return err
 	}
 
-	for rows.Next() {
-		var username string
-		if err := rows.Scan(&username); err != nil {
-			log.Warnf("Error while scanning for user '%s'", username)
-			return err
-		}
-
-		users[username] = IN_DB
+	for _, username := range usernames {
+		users[username] = InDB
 	}
 
 	l, err := la.getLdapConnection(true)
 	if err != nil {
-		log.Error("LDAP connection error")
+		cclog.Error("LDAP connection error")
 		return err
 	}
 	defer l.Close()
 
 	ldapResults, err := l.Search(ldap.NewSearchRequest(
-		la.config.UserBase, ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
-		la.config.UserFilter, []string{"dn", "uid", "gecos"}, nil))
+		lc.UserBase,
+		ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
+		lc.UserFilter,
+		[]string{"dn", "uid", la.UserAttr}, nil))
 	if err != nil {
-		log.Warn("LDAP search error")
+		cclog.Warn("LDAP search error")
 		return err
 	}
 
@@ -147,26 +185,35 @@ func (la *LdapAuthenticator) Sync() error {
 
 		_, ok := users[username]
 		if !ok {
-			users[username] = IN_LDAP
-			newnames[username] = entry.GetAttributeValue("gecos")
+			users[username] = InLdap
+			newnames[username] = entry.GetAttributeValue(la.UserAttr)
 		} else {
-			users[username] = IN_BOTH
+			users[username] = InBoth
 		}
 	}
 
 	for username, where := range users {
-		if where == IN_DB && la.config.SyncDelOldUsers {
-			log.Debugf("sync: remove %v (does not show up in LDAP anymore)", username)
-			if _, err := la.auth.db.Exec(`DELETE FROM user WHERE user.username = ?`, username); err != nil {
-				log.Errorf("User '%s' not in LDAP anymore: Delete from DB failed", username)
-				return err
-			}
-		} else if where == IN_LDAP {
+		if where == InDB && lc.SyncDelOldUsers {
+			ur.DelUser(username)
+			cclog.Debugf("sync: remove %v (does not show up in LDAP anymore)", username)
+		} else if where == InLdap {
 			name := newnames[username]
-			log.Debugf("sync: add %v (name: %v, roles: [user], ldap: true)", username, name)
-			if _, err := la.auth.db.Exec(`INSERT INTO user (username, ldap, name, roles) VALUES (?, ?, ?, ?)`,
-				username, 1, name, "[\""+GetRoleString(RoleUser)+"\"]"); err != nil {
-				log.Errorf("User '%s' new in LDAP: Insert into DB failed", username)
+
+			var roles []string
+			roles = append(roles, schema.GetRoleString(schema.RoleUser))
+			projects := make([]string, 0)
+
+			user := &schema.User{
+				Username:   username,
+				Name:       name,
+				Roles:      roles,
+				Projects:   projects,
+				AuthSource: schema.AuthViaLDAP,
+			}
+
+			cclog.Debugf("sync: add %v (name: %v, roles: [user], ldap: true)", username, name)
+			if err := ur.AddUser(user); err != nil {
+				cclog.Errorf("User '%s' LDAP: Insert into DB failed", username)
 				return err
 			}
 		}
@@ -175,20 +222,18 @@ func (la *LdapAuthenticator) Sync() error {
 	return nil
 }
 
-// TODO: Add a connection pool or something like
-// that so that connections can be reused/cached.
 func (la *LdapAuthenticator) getLdapConnection(admin bool) (*ldap.Conn, error) {
-
-	conn, err := ldap.DialURL(la.config.Url)
+	lc := Keys.LdapConfig
+	conn, err := ldap.DialURL(lc.URL)
 	if err != nil {
-		log.Warn("LDAP URL dial failed")
+		cclog.Warn("LDAP URL dial failed")
 		return nil, err
 	}
 
 	if admin {
-		if err := conn.Bind(la.config.SearchDN, la.syncPassword); err != nil {
+		if err := conn.Bind(lc.SearchDN, la.syncPassword); err != nil {
 			conn.Close()
-			log.Warn("LDAP connection bind failed")
+			cclog.Warn("LDAP connection bind failed")
 			return nil, err
 		}
 	}

internal/auth/local.go

@@ -1,14 +1,16 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package auth
 
 import (
 	"fmt"
 	"net/http"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 	"golang.org/x/crypto/bcrypt"
 )
 
@@ -18,38 +20,29 @@ type LocalAuthenticator struct {
 
 var _ Authenticator = (*LocalAuthenticator)(nil)
 
-func (la *LocalAuthenticator) Init(
-	auth *Authentication,
-	_ interface{}) error {
-
-	la.auth = auth
+func (la *LocalAuthenticator) Init() error {
 	return nil
 }
 
 func (la *LocalAuthenticator) CanLogin(
-	user *User,
+	user *schema.User,
+	username string,
 	rw http.ResponseWriter,
-	r *http.Request) bool {
-
-	return user != nil && user.AuthSource == AuthViaLocalPassword
+	r *http.Request,
+) (*schema.User, bool) {
+	return user, user != nil && user.AuthSource == schema.AuthViaLocalPassword
 }
 
 func (la *LocalAuthenticator) Login(
-	user *User,
+	user *schema.User,
 	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
-	if e := bcrypt.CompareHashAndPassword([]byte(user.Password), []byte(r.FormValue("password"))); e != nil {
-		log.Errorf("AUTH/LOCAL > Authentication for user %s failed!", user.Username)
-		return nil, fmt.Errorf("AUTH/LOCAL > Authentication failed")
+	r *http.Request,
+) (*schema.User, error) {
+	if e := bcrypt.CompareHashAndPassword([]byte(user.Password),
+		[]byte(r.FormValue("password"))); e != nil {
+		cclog.Errorf("AUTH/LOCAL > Authentication for user %s failed!", user.Username)
+		return nil, fmt.Errorf("Authentication failed")
 	}
 
 	return user, nil
 }
-
-func (la *LocalAuthenticator) Auth(
-	rw http.ResponseWriter,
-	r *http.Request) (*User, error) {
-
-	return la.auth.AuthViaSession(rw, r)
-}

internal/auth/oidc.go

@@ -0,0 +1,212 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package auth
+
+import (
+	"context"
+	"crypto/rand"
+	"encoding/base64"
+	"io"
+	"net/http"
+	"os"
+	"time"
+
+	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/coreos/go-oidc/v3/oidc"
+	"github.com/gorilla/mux"
+	"golang.org/x/oauth2"
+)
+
+type OpenIDConfig struct {
+	Provider          string `json:"provider"`
+	SyncUserOnLogin   bool   `json:"syncUserOnLogin"`
+	UpdateUserOnLogin bool   `json:"updateUserOnLogin"`
+}
+
+type OIDC struct {
+	client         *oauth2.Config
+	provider       *oidc.Provider
+	authentication *Authentication
+	clientID       string
+}
+
+func randString(nByte int) (string, error) {
+	b := make([]byte, nByte)
+	if _, err := io.ReadFull(rand.Reader, b); err != nil {
+		return "", err
+	}
+	return base64.RawURLEncoding.EncodeToString(b), nil
+}
+
+func setCallbackCookie(w http.ResponseWriter, r *http.Request, name, value string) {
+	c := &http.Cookie{
+		Name:     name,
+		Value:    value,
+		MaxAge:   int(time.Hour.Seconds()),
+		Secure:   r.TLS != nil,
+		HttpOnly: true,
+	}
+	http.SetCookie(w, c)
+}
+
+// NewOIDC creates a new OIDC authenticator with the configured provider
+func NewOIDC(a *Authentication) *OIDC {
+	// Use context with timeout for provider initialization
+	ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+	defer cancel()
+	
+	provider, err := oidc.NewProvider(ctx, Keys.OpenIDConfig.Provider)
+	if err != nil {
+		cclog.Fatal(err)
+	}
+	clientID := os.Getenv("OID_CLIENT_ID")
+	if clientID == "" {
+		cclog.Warn("environment variable 'OID_CLIENT_ID' not set (Open ID connect auth will not work)")
+	}
+	clientSecret := os.Getenv("OID_CLIENT_SECRET")
+	if clientSecret == "" {
+		cclog.Warn("environment variable 'OID_CLIENT_SECRET' not set (Open ID connect auth will not work)")
+	}
+
+	client := &oauth2.Config{
+		ClientID:     clientID,
+		ClientSecret: clientSecret,
+		Endpoint:     provider.Endpoint(),
+		RedirectURL:  "oidc-callback",
+		Scopes:       []string{oidc.ScopeOpenID, "profile", "email"},
+	}
+
+	oa := &OIDC{provider: provider, client: client, clientID: clientID, authentication: a}
+
+	return oa
+}
+
+func (oa *OIDC) RegisterEndpoints(r *mux.Router) {
+	r.HandleFunc("/oidc-login", oa.OAuth2Login)
+	r.HandleFunc("/oidc-callback", oa.OAuth2Callback)
+}
+
+func (oa *OIDC) OAuth2Callback(rw http.ResponseWriter, r *http.Request) {
+	c, err := r.Cookie("state")
+	if err != nil {
+		http.Error(rw, "state cookie not found", http.StatusBadRequest)
+		return
+	}
+	state := c.Value
+
+	c, err = r.Cookie("verifier")
+	if err != nil {
+		http.Error(rw, "verifier cookie not found", http.StatusBadRequest)
+		return
+	}
+	codeVerifier := c.Value
+
+	_ = r.ParseForm()
+	if r.Form.Get("state") != state {
+		http.Error(rw, "State invalid", http.StatusBadRequest)
+		return
+	}
+	code := r.Form.Get("code")
+	if code == "" {
+		http.Error(rw, "Code not found", http.StatusBadRequest)
+		return
+	}
+	// Exchange authorization code for token with timeout
+	ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+	defer cancel()
+	
+	token, err := oa.client.Exchange(ctx, code, oauth2.VerifierOption(codeVerifier))
+	if err != nil {
+		http.Error(rw, "Failed to exchange token: "+err.Error(), http.StatusInternalServerError)
+		return
+	}
+
+	// Get user info from OIDC provider with same timeout
+	userInfo, err := oa.provider.UserInfo(ctx, oauth2.StaticTokenSource(token))
+	if err != nil {
+		http.Error(rw, "Failed to get userinfo: "+err.Error(), http.StatusInternalServerError)
+		return
+	}
+
+	// // Extract the ID Token from OAuth2 token.
+	// rawIDToken, ok := token.Extra("id_token").(string)
+	// if !ok {
+	// 	http.Error(rw, "Cannot access idToken", http.StatusInternalServerError)
+	// }
+	//
+	// verifier := oa.provider.Verifier(&oidc.Config{ClientID: oa.clientID})
+	// // Parse and verify ID Token payload.
+	// idToken, err := verifier.Verify(context.Background(), rawIDToken)
+	// if err != nil {
+	// 	http.Error(rw, "Failed to extract idToken: "+err.Error(), http.StatusInternalServerError)
+	// }
+
+	projects := make([]string, 0)
+
+	// Extract custom claims
+	var claims struct {
+		Username string `json:"preferred_username"`
+		Name     string `json:"name"`
+		Profile  struct {
+			Client struct {
+				Roles []string `json:"roles"`
+			} `json:"clustercockpit"`
+		} `json:"resource_access"`
+	}
+	if err := userInfo.Claims(&claims); err != nil {
+		http.Error(rw, "Failed to extract Claims: "+err.Error(), http.StatusInternalServerError)
+	}
+
+	var roles []string
+	for _, r := range claims.Profile.Client.Roles {
+		switch r {
+		case "user":
+			roles = append(roles, schema.GetRoleString(schema.RoleUser))
+		case "admin":
+			roles = append(roles, schema.GetRoleString(schema.RoleAdmin))
+		}
+	}
+
+	if len(roles) == 0 {
+		roles = append(roles, schema.GetRoleString(schema.RoleUser))
+	}
+
+	user := &schema.User{
+		Username:   claims.Username,
+		Name:       claims.Name,
+		Roles:      roles,
+		Projects:   projects,
+		AuthSource: schema.AuthViaOIDC,
+	}
+
+	if Keys.OpenIDConfig.SyncUserOnLogin || Keys.OpenIDConfig.UpdateUserOnLogin {
+		handleOIDCUser(user)
+	}
+
+	oa.authentication.SaveSession(rw, r, user)
+	cclog.Infof("login successfull: user: %#v (roles: %v, projects: %v)", user.Username, user.Roles, user.Projects)
+	userCtx := context.WithValue(r.Context(), repository.ContextUserKey, user)
+	http.RedirectHandler("/", http.StatusTemporaryRedirect).ServeHTTP(rw, r.WithContext(userCtx))
+}
+
+func (oa *OIDC) OAuth2Login(rw http.ResponseWriter, r *http.Request) {
+	state, err := randString(16)
+	if err != nil {
+		http.Error(rw, "Internal error", http.StatusInternalServerError)
+		return
+	}
+	setCallbackCookie(rw, r, "state", state)
+
+	// use PKCE to protect against CSRF attacks
+	codeVerifier := oauth2.GenerateVerifier()
+	setCallbackCookie(rw, r, "verifier", codeVerifier)
+
+	// Redirect user to consent page to ask for permission
+	url := oa.client.AuthCodeURL(state, oauth2.AccessTypeOffline, oauth2.S256ChallengeOption(codeVerifier))
+	http.Redirect(rw, r, url, http.StatusFound)
+}

internal/auth/schema.go

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

internal/auth/users.go

@@ -1,289 +0,0 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
-// Use of this source code is governed by a MIT-style
-// license that can be found in the LICENSE file.
-package auth
-
-import (
-	"context"
-	"database/sql"
-	"encoding/json"
-	"errors"
-	"fmt"
-	"strings"
-
-	"github.com/ClusterCockpit/cc-backend/internal/graph/model"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	sq "github.com/Masterminds/squirrel"
-	"github.com/jmoiron/sqlx"
-	"golang.org/x/crypto/bcrypt"
-)
-
-func (auth *Authentication) GetUser(username string) (*User, error) {
-
-	user := &User{Username: username}
-	var hashedPassword, name, rawRoles, email, rawProjects sql.NullString
-	if err := sq.Select("password", "ldap", "name", "roles", "email", "projects").From("user").
-		Where("user.username = ?", username).RunWith(auth.db).
-		QueryRow().Scan(&hashedPassword, &user.AuthSource, &name, &rawRoles, &email, &rawProjects); err != nil {
-		log.Warnf("Error while querying user '%v' from database", username)
-		return nil, err
-	}
-
-	user.Password = hashedPassword.String
-	user.Name = name.String
-	user.Email = email.String
-	if rawRoles.Valid {
-		if err := json.Unmarshal([]byte(rawRoles.String), &user.Roles); err != nil {
-			log.Warn("Error while unmarshaling raw roles from DB")
-			return nil, err
-		}
-	}
-	if rawProjects.Valid {
-		if err := json.Unmarshal([]byte(rawProjects.String), &user.Projects); err != nil {
-			return nil, err
-		}
-	}
-
-	return user, nil
-}
-
-func (auth *Authentication) AddUser(user *User) error {
-
-	rolesJson, _ := json.Marshal(user.Roles)
-	projectsJson, _ := json.Marshal(user.Projects)
-
-	cols := []string{"username", "roles", "projects"}
-	vals := []interface{}{user.Username, string(rolesJson), string(projectsJson)}
-
-	if user.Name != "" {
-		cols = append(cols, "name")
-		vals = append(vals, user.Name)
-	}
-	if user.Email != "" {
-		cols = append(cols, "email")
-		vals = append(vals, user.Email)
-	}
-	if user.Password != "" {
-		password, err := bcrypt.GenerateFromPassword([]byte(user.Password), bcrypt.DefaultCost)
-		if err != nil {
-			log.Error("Error while encrypting new user password")
-			return err
-		}
-		cols = append(cols, "password")
-		vals = append(vals, string(password))
-	}
-
-	if _, err := sq.Insert("user").Columns(cols...).Values(vals...).RunWith(auth.db).Exec(); err != nil {
-		log.Errorf("Error while inserting new user '%v' into DB", user.Username)
-		return err
-	}
-
-	log.Infof("new user %#v created (roles: %s, auth-source: %d, projects: %s)", user.Username, rolesJson, user.AuthSource, projectsJson)
-	return nil
-}
-
-func (auth *Authentication) DelUser(username string) error {
-
-	_, err := auth.db.Exec(`DELETE FROM user WHERE user.username = ?`, username)
-	log.Errorf("Error while deleting user '%s' from DB", username)
-	return err
-}
-
-func (auth *Authentication) ListUsers(specialsOnly bool) ([]*User, error) {
-
-	q := sq.Select("username", "name", "email", "roles", "projects").From("user")
-	if specialsOnly {
-		q = q.Where("(roles != '[\"user\"]' AND roles != '[]')")
-	}
-
-	rows, err := q.RunWith(auth.db).Query()
-	if err != nil {
-		log.Warn("Error while querying user list")
-		return nil, err
-	}
-
-	users := make([]*User, 0)
-	defer rows.Close()
-	for rows.Next() {
-		rawroles := ""
-		rawprojects := ""
-		user := &User{}
-		var name, email sql.NullString
-		if err := rows.Scan(&user.Username, &name, &email, &rawroles, &rawprojects); err != nil {
-			log.Warn("Error while scanning user list")
-			return nil, err
-		}
-
-		if err := json.Unmarshal([]byte(rawroles), &user.Roles); err != nil {
-			log.Warn("Error while unmarshaling raw role list")
-			return nil, err
-		}
-
-		if err := json.Unmarshal([]byte(rawprojects), &user.Projects); err != nil {
-			return nil, err
-		}
-
-		user.Name = name.String
-		user.Email = email.String
-		users = append(users, user)
-	}
-	return users, nil
-}
-
-func (auth *Authentication) AddRole(
-	ctx context.Context,
-	username string,
-	queryrole string) error {
-
-	newRole := strings.ToLower(queryrole)
-	user, err := auth.GetUser(username)
-	if err != nil {
-		log.Warnf("Could not load user '%s'", username)
-		return err
-	}
-
-	exists, valid := user.HasValidRole(newRole)
-
-	if !valid {
-		return fmt.Errorf("Supplied role is no valid option : %v", newRole)
-	}
-	if exists {
-		return fmt.Errorf("User %v already has role %v", username, newRole)
-	}
-
-	roles, _ := json.Marshal(append(user.Roles, newRole))
-	if _, err := sq.Update("user").Set("roles", roles).Where("user.username = ?", username).RunWith(auth.db).Exec(); err != nil {
-		log.Errorf("Error while adding new role for user '%s'", user.Username)
-		return err
-	}
-	return nil
-}
-
-func (auth *Authentication) RemoveRole(ctx context.Context, username string, queryrole string) error {
-	oldRole := strings.ToLower(queryrole)
-	user, err := auth.GetUser(username)
-	if err != nil {
-		log.Warnf("Could not load user '%s'", username)
-		return err
-	}
-
-	exists, valid := user.HasValidRole(oldRole)
-
-	if !valid {
-		return fmt.Errorf("Supplied role is no valid option : %v", oldRole)
-	}
-	if !exists {
-		return fmt.Errorf("Role already deleted for user '%v': %v", username, oldRole)
-	}
-
-	if oldRole == GetRoleString(RoleManager) && len(user.Projects) != 0 {
-		return fmt.Errorf("Cannot remove role 'manager' while user %s still has assigned project(s) : %v", username, user.Projects)
-	}
-
-	var newroles []string
-	for _, r := range user.Roles {
-		if r != oldRole {
-			newroles = append(newroles, r) // Append all roles not matching requested to be deleted role
-		}
-	}
-
-	var mroles, _ = json.Marshal(newroles)
-	if _, err := sq.Update("user").Set("roles", mroles).Where("user.username = ?", username).RunWith(auth.db).Exec(); err != nil {
-		log.Errorf("Error while removing role for user '%s'", user.Username)
-		return err
-	}
-	return nil
-}
-
-func (auth *Authentication) AddProject(
-	ctx context.Context,
-	username string,
-	project string) error {
-
-	user, err := auth.GetUser(username)
-	if err != nil {
-		return err
-	}
-
-	if !user.HasRole(RoleManager) {
-		return fmt.Errorf("user '%s' is not a manager!", username)
-	}
-
-	if user.HasProject(project) {
-		return fmt.Errorf("user '%s' already manages project '%s'", username, project)
-	}
-
-	projects, _ := json.Marshal(append(user.Projects, project))
-	if _, err := sq.Update("user").Set("projects", projects).Where("user.username = ?", username).RunWith(auth.db).Exec(); err != nil {
-		return err
-	}
-
-	return nil
-}
-
-func (auth *Authentication) RemoveProject(ctx context.Context, username string, project string) error {
-	user, err := auth.GetUser(username)
-	if err != nil {
-		return err
-	}
-
-	if !user.HasRole(RoleManager) {
-		return fmt.Errorf("user '%#v' is not a manager!", username)
-	}
-
-	if !user.HasProject(project) {
-		return fmt.Errorf("user '%#v': Cannot remove project '%#v' - Does not match!", username, project)
-	}
-
-	var exists bool
-	var newprojects []string
-	for _, p := range user.Projects {
-		if p != project {
-			newprojects = append(newprojects, p) // Append all projects not matching requested to be deleted project
-		} else {
-			exists = true
-		}
-	}
-
-	if exists == true {
-		var result interface{}
-		if len(newprojects) == 0 {
-			result = "[]"
-		} else {
-			result, _ = json.Marshal(newprojects)
-		}
-		if _, err := sq.Update("user").Set("projects", result).Where("user.username = ?", username).RunWith(auth.db).Exec(); err != nil {
-			return err
-		}
-		return nil
-	} else {
-		return fmt.Errorf("user %s already does not manage project %s", username, project)
-	}
-}
-
-func FetchUser(ctx context.Context, db *sqlx.DB, username string) (*model.User, error) {
-	me := GetUser(ctx)
-	if me != nil && me.Username != username && me.HasNotRoles([]Role{RoleAdmin, RoleSupport, RoleManager}) {
-		return nil, errors.New("forbidden")
-	}
-
-	user := &model.User{Username: username}
-	var name, email sql.NullString
-	if err := sq.Select("name", "email").From("user").Where("user.username = ?", username).
-		RunWith(db).QueryRow().Scan(&name, &email); err != nil {
-		if err == sql.ErrNoRows {
-			/* This warning will be logged *often* for non-local users, i.e. users mentioned only in job-table or archive, */
-			/* since FetchUser will be called to retrieve full name and mail for every job in query/list									 */
-			// log.Warnf("User '%s' Not found in DB", username)
-			return nil, nil
-		}
-
-		log.Warnf("Error while fetching user '%s'", username)
-		return nil, err
-	}
-
-	user.Name = name.String
-	user.Email = email.String
-	return user, nil
-}

internal/config/config.go

@@ -1,68 +1,158 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
+// Package config implements the program configuration data structures, validation and parsing
 package config
 
 import (
 	"bytes"
 	"encoding/json"
-	"log"
-	"os"
+	"time"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/resampler"
 )
 
-var Keys schema.ProgramConfig = schema.ProgramConfig{
+type ProgramConfig struct {
+	// Address where the http (or https) server will listen on (for example: 'localhost:80').
+	Addr string `json:"addr"`
+
+	// Addresses from which secured admin API endpoints can be reached, can be wildcard "*"
+	APIAllowedIPs []string `json:"apiAllowedIPs"`
+
+	APISubjects *NATSConfig `json:"apiSubjects"`
+
+	// Drop root permissions once .env was read and the port was taken.
+	User  string `json:"user"`
+	Group string `json:"group"`
+
+	// Disable authentication (for everything: API, Web-UI, ...)
+	DisableAuthentication bool `json:"disable-authentication"`
+
+	// If `embed-static-files` is true (default), the frontend files are directly
+	// embeded into the go binary and expected to be in web/frontend. Only if
+	// it is false the files in `static-files` are served instead.
+	EmbedStaticFiles bool   `json:"embed-static-files"`
+	StaticFiles      string `json:"static-files"`
+
+	// 'sqlite3' or 'mysql' (mysql will work for mariadb as well)
+	DBDriver string `json:"db-driver"`
+
+	// For sqlite3 a filename, for mysql a DSN in this format: https://github.com/go-sql-driver/mysql#dsn-data-source-name (Without query parameters!).
+	DB string `json:"db"`
+
+	// Keep all metric data in the metric data repositories,
+	// do not write to the job-archive.
+	DisableArchive bool `json:"disable-archive"`
+
+	EnableJobTaggers bool `json:"enable-job-taggers"`
+
+	// Validate json input against schema
+	Validate bool `json:"validate"`
+
+	// If 0 or empty, the session does not expire!
+	SessionMaxAge string `json:"session-max-age"`
+
+	// If both those options are not empty, use HTTPS using those certificates.
+	HTTPSCertFile string `json:"https-cert-file"`
+	HTTPSKeyFile  string `json:"https-key-file"`
+
+	// If not the empty string and `addr` does not end in ":80",
+	// redirect every request incoming at port 80 to that url.
+	RedirectHTTPTo string `json:"redirect-http-to"`
+
+	// Where to store MachineState files
+	MachineStateDir string `json:"machine-state-dir"`
+
+	// If not zero, automatically mark jobs as stopped running X seconds longer than their walltime.
+	StopJobsExceedingWalltime int `json:"stop-jobs-exceeding-walltime"`
+
+	// Defines time X in seconds in which jobs are considered to be "short" and will be filtered in specific views.
+	ShortRunningJobsDuration int `json:"short-running-jobs-duration"`
+
+	// Energy Mix CO2 Emission Constant [g/kWh]
+	// If entered, displays estimated CO2 emission for job based on jobs totalEnergy
+	EmissionConstant int `json:"emission-constant"`
+
+	// If exists, will enable dynamic zoom in frontend metric plots using the configured values
+	EnableResampling *ResampleConfig `json:"resampling"`
+}
+
+type ResampleConfig struct {
+	// Minimum number of points to trigger resampling of data
+	MinimumPoints int `json:"minimumPoints"`
+	// Array of resampling target resolutions, in seconds; Example: [600,300,60]
+	Resolutions []int `json:"resolutions"`
+	// Trigger next zoom level at less than this many visible datapoints
+	Trigger int `json:"trigger"`
+}
+
+type NATSConfig struct {
+	SubjectJobStart  string `json:"subjectJobStart"`
+	SubjectJobStop   string `json:"subjectJobStop"`
+	SubjectNodeState string `json:"subjectNodeState"`
+}
+
+type IntRange struct {
+	From int `json:"from"`
+	To   int `json:"to"`
+}
+
+type TimeRange struct {
+	From  *time.Time `json:"from"`
+	To    *time.Time `json:"to"`
+	Range string     `json:"range,omitempty"`
+}
+
+type FilterRanges struct {
+	Duration  *IntRange  `json:"duration"`
+	NumNodes  *IntRange  `json:"numNodes"`
+	StartTime *TimeRange `json:"startTime"`
+}
+
+type ClusterConfig struct {
+	Name                 string          `json:"name"`
+	FilterRanges         *FilterRanges   `json:"filterRanges"`
+	MetricDataRepository json.RawMessage `json:"metricDataRepository"`
+}
+
+var Clusters []*ClusterConfig
+
+var Keys ProgramConfig = ProgramConfig{
 	Addr:                      "localhost:8080",
 	DisableAuthentication:     false,
 	EmbedStaticFiles:          true,
 	DBDriver:                  "sqlite3",
 	DB:                        "./var/job.db",
-	Archive:                   json.RawMessage(`{\"kind\":\"file\",\"path\":\"./var/job-archive\"}`),
 	DisableArchive:            false,
 	Validate:                  false,
-	LdapConfig:                nil,
 	SessionMaxAge:             "168h",
 	StopJobsExceedingWalltime: 0,
 	ShortRunningJobsDuration:  5 * 60,
-	UiDefaults: map[string]interface{}{
-		"analysis_view_histogramMetrics":     []string{"flops_any", "mem_bw", "mem_used"},
-		"analysis_view_scatterPlotMetrics":   [][]string{{"flops_any", "mem_bw"}, {"flops_any", "cpu_load"}, {"cpu_load", "mem_bw"}},
-		"job_view_nodestats_selectedMetrics": []string{"flops_any", "mem_bw", "mem_used"},
-		"job_view_polarPlotMetrics":          []string{"flops_any", "mem_bw", "mem_used"},
-		"job_view_selectedMetrics":           []string{"flops_any", "mem_bw", "mem_used"},
-		"plot_general_colorBackground":       true,
-		"plot_general_colorscheme":           []string{"#00bfff", "#0000ff", "#ff00ff", "#ff0000", "#ff8000", "#ffff00", "#80ff00"},
-		"plot_general_lineWidth":             3,
-		"plot_list_jobsPerPage":              50,
-		"plot_list_selectedMetrics":          []string{"cpu_load", "mem_used", "flops_any", "mem_bw"},
-		"plot_view_plotsPerRow":              3,
-		"plot_view_showPolarplot":            true,
-		"plot_view_showRoofline":             true,
-		"plot_view_showStatTable":            true,
-		"system_view_selectedMetric":         "cpu_load",
-	},
 }
 
-func Init(flagConfigFile string) {
-	raw, err := os.ReadFile(flagConfigFile)
-	if err != nil {
-		if !os.IsNotExist(err) {
-			log.Fatalf("CONFIG ERROR: %v", err)
-		}
-	} else {
-		if err := schema.Validate(schema.Config, bytes.NewReader(raw)); err != nil {
-			log.Fatalf("Validate config: %v\n", err)
-		}
-		dec := json.NewDecoder(bytes.NewReader(raw))
-		dec.DisallowUnknownFields()
-		if err := dec.Decode(&Keys); err != nil {
-			log.Fatalf("could not decode: %v", err)
-		}
+func Init(mainConfig json.RawMessage, clusterConfig json.RawMessage) {
+	Validate(configSchema, mainConfig)
+	dec := json.NewDecoder(bytes.NewReader(mainConfig))
+	dec.DisallowUnknownFields()
+	if err := dec.Decode(&Keys); err != nil {
+		cclog.Abortf("Config Init: Could not decode config file '%s'.\nError: %s\n", mainConfig, err.Error())
+	}
 
-		if Keys.Clusters == nil || len(Keys.Clusters) < 1 {
-			log.Fatal("At least one cluster required in config!")
-		}
+	Validate(clustersSchema, clusterConfig)
+	dec = json.NewDecoder(bytes.NewReader(clusterConfig))
+	dec.DisallowUnknownFields()
+	if err := dec.Decode(&Clusters); err != nil {
+		cclog.Abortf("Config Init: Could not decode config file '%s'.\nError: %s\n", mainConfig, err.Error())
+	}
+
+	if len(Clusters) < 1 {
+		cclog.Abort("Config Init: At least one cluster required in config. Exited with error.")
+	}
+
+	if Keys.EnableResampling != nil && Keys.EnableResampling.MinimumPoints > 0 {
+		resampler.SetMinimumRequiredPoints(Keys.EnableResampling.MinimumPoints)
 	}
 }

internal/config/config_test.go

@@ -1,16 +1,30 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package config
 
 import (
 	"testing"
+
+	ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
 )
 
 func TestInit(t *testing.T) {
 	fp := "../../configs/config.json"
-	Init(fp)
+	ccconf.Init(fp)
+	if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
+		if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
+			Init(cfg, clustercfg)
+		} else {
+			cclog.Abort("Cluster configuration must be present")
+		}
+	} else {
+		cclog.Abort("Main configuration must be present")
+	}
+
 	if Keys.Addr != "0.0.0.0:443" {
 		t.Errorf("wrong addr\ngot: %s \nwant: 0.0.0.0:443", Keys.Addr)
 	}
@@ -18,7 +32,17 @@ func TestInit(t *testing.T) {
 
 func TestInitMinimal(t *testing.T) {
 	fp := "../../configs/config-demo.json"
-	Init(fp)
+	ccconf.Init(fp)
+	if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
+		if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
+			Init(cfg, clustercfg)
+		} else {
+			cclog.Abort("Cluster configuration must be present")
+		}
+	} else {
+		cclog.Abort("Main configuration must be present")
+	}
+
 	if Keys.Addr != "127.0.0.1:8080" {
 		t.Errorf("wrong addr\ngot: %s \nwant: 127.0.0.1:8080", Keys.Addr)
 	}

internal/config/default_metrics.go

@@ -0,0 +1,51 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package config
+
+import (
+	"encoding/json"
+	"os"
+	"strings"
+)
+
+// DEPRECATED: SUPERSEDED BY NEW USER CONFIG - userConfig.go / web.go
+
+type DefaultMetricsCluster struct {
+	Name           string `json:"name"`
+	DefaultMetrics string `json:"default_metrics"`
+}
+
+type DefaultMetricsConfig struct {
+	Clusters []DefaultMetricsCluster `json:"clusters"`
+}
+
+func LoadDefaultMetricsConfig() (*DefaultMetricsConfig, error) {
+	filePath := "default_metrics.json"
+	if _, err := os.Stat(filePath); os.IsNotExist(err) {
+		return nil, nil
+	}
+	data, err := os.ReadFile(filePath)
+	if err != nil {
+		return nil, err
+	}
+	var cfg DefaultMetricsConfig
+	if err := json.Unmarshal(data, &cfg); err != nil {
+		return nil, err
+	}
+	return &cfg, nil
+}
+
+func ParseMetricsString(s string) []string {
+	parts := strings.Split(s, ",")
+	var metrics []string
+	for _, p := range parts {
+		trimmed := strings.TrimSpace(p)
+		if trimmed != "" {
+			metrics = append(metrics, trimmed)
+		}
+	}
+	return metrics
+}

internal/config/schema.go

@@ -0,0 +1,205 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package config
+
+var configSchema = `
+	{
+  "type": "object",
+  "properties": {
+    "addr": {
+      "description": "Address where the http (or https) server will listen on (for example: 'localhost:80').",
+      "type": "string"
+    },
+    "apiAllowedIPs": {
+      "description": "Addresses from which secured API endpoints can be reached",
+      "type": "array",
+      "items": {
+        "type": "string"
+      }
+    },
+    "user": {
+      "description": "Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.",
+      "type": "string"
+    },
+    "group": {
+      "description": "Drop root permissions once .env was read and the port was taken. Only applicable if using privileged port.",
+      "type": "string"
+    },
+    "disable-authentication": {
+      "description": "Disable authentication (for everything: API, Web-UI, ...).",
+      "type": "boolean"
+    },
+    "embed-static-files": {
+      "description": "If all files in web/frontend/public should be served from within the binary itself (they are embedded) or not.",
+      "type": "boolean"
+    },
+    "static-files": {
+      "description": "Folder where static assets can be found, if embed-static-files is false.",
+      "type": "string"
+    },
+    "db": {
+      "description": "For sqlite3 a filename, for mysql a DSN in this format: https://github.com/go-sql-driver/mysql#dsn-data-source-name (Without query parameters!).",
+      "type": "string"
+    },
+    "disable-archive": {
+      "description": "Keep all metric data in the metric data repositories, do not write to the job-archive.",
+      "type": "boolean"
+    },
+    "enable-job-taggers": {
+      "description": "Turn on automatic application and jobclass taggers",
+      "type": "boolean"
+    },
+    "validate": {
+      "description": "Validate all input json documents against json schema.",
+      "type": "boolean"
+    },
+    "session-max-age": {
+      "description": "Specifies for how long a session shall be valid  as a string parsable by time.ParseDuration(). If 0 or empty, the session/token does not expire!",
+      "type": "string"
+    },
+    "https-cert-file": {
+      "description": "Filepath to SSL certificate. If also https-key-file is set use HTTPS using those certificates.",
+      "type": "string"
+    },
+    "https-key-file": {
+      "description": "Filepath to SSL key file. If also https-cert-file is set use HTTPS using those certificates.",
+      "type": "string"
+    },
+    "redirect-http-to": {
+      "description": "If not the empty string and addr does not end in :80, redirect every request incoming at port 80 to that url.",
+      "type": "string"
+    },
+    "stop-jobs-exceeding-walltime": {
+      "description": "If not zero, automatically mark jobs as stopped running X seconds longer than their walltime. Only applies if walltime is set for job.",
+      "type": "integer"
+    },
+    "short-running-jobs-duration": {
+      "description": "Do not show running jobs shorter than X seconds.",
+      "type": "integer"
+    },
+    "emission-constant": {
+      "description": ".",
+      "type": "integer"
+    },
+    "cron-frequency": {
+      "description": "Frequency of cron job workers.",
+      "type": "object",
+      "properties": {
+        "duration-worker": {
+          "description": "Duration Update Worker [Defaults to '5m']",
+          "type": "string"
+        },
+        "footprint-worker": {
+          "description": "Metric-Footprint Update Worker [Defaults to '10m']",
+          "type": "string"
+        }
+      }
+    },
+    "enable-resampling": {
+      "description": "Enable dynamic zoom in frontend metric plots.",
+      "type": "object",
+      "properties": {
+        "minimumPoints": {
+          "description": "Minimum points to trigger resampling of time-series data.",
+          "type": "integer"
+        },
+        "trigger": {
+          "description": "Trigger next zoom level at less than this many visible datapoints.",
+          "type": "integer"
+        },
+        "resolutions": {
+          "description": "Array of resampling target resolutions, in seconds.",
+          "type": "array",
+          "items": {
+            "type": "integer"
+          }
+        }
+      },
+      "required": ["trigger", "resolutions"]
+    }
+	},
+  "required": ["apiAllowedIPs"]
+	}`
+
+var clustersSchema = `
+  {
+    "type": "array",
+    "items": {
+      "type": "object",
+      "properties": {
+        "name": {
+          "description": "The name of the cluster.",
+          "type": "string"
+        },
+        "metricDataRepository": {
+          "description": "Type of the metric data repository for this cluster",
+          "type": "object",
+          "properties": {
+            "kind": {
+              "type": "string",
+                "enum": ["influxdb", "prometheus", "cc-metric-store", "cc-metric-store-internal", "test"]
+            },
+            "url": {
+              "type": "string"
+            },
+            "token": {
+              "type": "string"
+            }
+          },
+          "required": ["kind"]
+        },
+        "filterRanges": {
+          "description": "This option controls the slider ranges for the UI controls of numNodes, duration, and startTime.",
+          "type": "object",
+          "properties": {
+            "numNodes": {
+              "description": "UI slider range for number of nodes",
+              "type": "object",
+              "properties": {
+                "from": {
+                  "type": "integer"
+                },
+                "to": {
+                  "type": "integer"
+                }
+              },
+              "required": ["from", "to"]
+            },
+            "duration": {
+              "description": "UI slider range for duration",
+              "type": "object",
+              "properties": {
+                "from": {
+                  "type": "integer"
+                },
+                "to": {
+                  "type": "integer"
+                }
+              },
+              "required": ["from", "to"]
+            },
+            "startTime": {
+              "description": "UI slider range for start time",
+              "type": "object",
+              "properties": {
+                "from": {
+                  "type": "string",
+                  "format": "date-time"
+                },
+                "to": {
+                  "type": "null"
+                }
+              },
+              "required": ["from", "to"]
+            }
+          },
+          "required": ["numNodes", "duration", "startTime"]
+        }
+      },
+      "required": ["name", "metricDataRepository", "filterRanges"],
+      "minItems": 1
+    }
+  }`

internal/config/validate.go

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

internal/graph/model/models.go

@@ -1,5 +1,6 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package model

internal/graph/model/models_gen.go

@@ -3,27 +3,53 @@
 package model
 
 import (
+	"bytes"
 	"fmt"
 	"io"
 	"strconv"
 	"time"
 
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	"github.com/ClusterCockpit/cc-backend/internal/config"
+	"github.com/ClusterCockpit/cc-lib/schema"
 )
 
+type ClusterMetricWithName struct {
+	Name     string         `json:"name"`
+	Unit     *schema.Unit   `json:"unit,omitempty"`
+	Timestep int            `json:"timestep"`
+	Data     []schema.Float `json:"data"`
+}
+
+type ClusterMetrics struct {
+	NodeCount int                      `json:"nodeCount"`
+	Metrics   []*ClusterMetricWithName `json:"metrics"`
+}
+
 type Count struct {
 	Name  string `json:"name"`
 	Count int    `json:"count"`
 }
 
+type EnergyFootprintValue struct {
+	Hardware string  `json:"hardware"`
+	Metric   string  `json:"metric"`
+	Value    float64 `json:"value"`
+}
+
 type FloatRange struct {
 	From float64 `json:"from"`
 	To   float64 `json:"to"`
 }
 
+type FootprintValue struct {
+	Name  string  `json:"name"`
+	Stat  string  `json:"stat"`
+	Value float64 `json:"value"`
+}
+
 type Footprints struct {
-	Nodehours []schema.Float      `json:"nodehours"`
-	Metrics   []*MetricFootprints `json:"metrics"`
+	TimeWeights *TimeWeights        `json:"timeWeights"`
+	Metrics     []*MetricFootprints `json:"metrics"`
 }
 
 type HistoPoint struct {
@@ -37,27 +63,26 @@ type IntRangeOutput struct {
 }
 
 type JobFilter struct {
-	Tags            []string          `json:"tags"`
-	JobID           *StringInput      `json:"jobId"`
-	ArrayJobID      *int              `json:"arrayJobId"`
-	User            *StringInput      `json:"user"`
-	Project         *StringInput      `json:"project"`
-	JobName         *StringInput      `json:"jobName"`
-	Cluster         *StringInput      `json:"cluster"`
-	Partition       *StringInput      `json:"partition"`
-	Duration        *schema.IntRange  `json:"duration"`
-	MinRunningFor   *int              `json:"minRunningFor"`
-	NumNodes        *schema.IntRange  `json:"numNodes"`
-	NumAccelerators *schema.IntRange  `json:"numAccelerators"`
-	NumHWThreads    *schema.IntRange  `json:"numHWThreads"`
-	StartTime       *schema.TimeRange `json:"startTime"`
-	State           []schema.JobState `json:"state"`
-	FlopsAnyAvg     *FloatRange       `json:"flopsAnyAvg"`
-	MemBwAvg        *FloatRange       `json:"memBwAvg"`
-	LoadAvg         *FloatRange       `json:"loadAvg"`
-	MemUsedMax      *FloatRange       `json:"memUsedMax"`
-	Exclusive       *int              `json:"exclusive"`
-	Node            *StringInput      `json:"node"`
+	Tags            []string          `json:"tags,omitempty"`
+	DbID            []string          `json:"dbId,omitempty"`
+	JobID           *StringInput      `json:"jobId,omitempty"`
+	ArrayJobID      *int              `json:"arrayJobId,omitempty"`
+	User            *StringInput      `json:"user,omitempty"`
+	Project         *StringInput      `json:"project,omitempty"`
+	JobName         *StringInput      `json:"jobName,omitempty"`
+	Cluster         *StringInput      `json:"cluster,omitempty"`
+	Partition       *StringInput      `json:"partition,omitempty"`
+	Duration        *config.IntRange  `json:"duration,omitempty"`
+	Energy          *FloatRange       `json:"energy,omitempty"`
+	MinRunningFor   *int              `json:"minRunningFor,omitempty"`
+	NumNodes        *config.IntRange  `json:"numNodes,omitempty"`
+	NumAccelerators *config.IntRange  `json:"numAccelerators,omitempty"`
+	NumHWThreads    *config.IntRange  `json:"numHWThreads,omitempty"`
+	StartTime       *config.TimeRange `json:"startTime,omitempty"`
+	State           []schema.JobState `json:"state,omitempty"`
+	MetricStats     []*MetricStatItem `json:"metricStats,omitempty"`
+	Shared          *string           `json:"shared,omitempty"`
+	Node            *StringInput      `json:"node,omitempty"`
 }
 
 type JobLink struct {
@@ -66,9 +91,9 @@ type JobLink struct {
 }
 
 type JobLinkResultList struct {
-	ListQuery *string    `json:"listQuery"`
+	ListQuery *string    `json:"listQuery,omitempty"`
 	Items     []*JobLink `json:"items"`
-	Count     *int       `json:"count"`
+	Count     *int       `json:"count,omitempty"`
 }
 
 type JobMetricWithName struct {
@@ -78,24 +103,45 @@ type JobMetricWithName struct {
 }
 
 type JobResultList struct {
-	Items  []*schema.Job `json:"items"`
-	Offset *int          `json:"offset"`
-	Limit  *int          `json:"limit"`
-	Count  *int          `json:"count"`
+	Items       []*schema.Job `json:"items"`
+	Offset      *int          `json:"offset,omitempty"`
+	Limit       *int          `json:"limit,omitempty"`
+	Count       *int          `json:"count,omitempty"`
+	HasNextPage *bool         `json:"hasNextPage,omitempty"`
+}
+
+type JobStats struct {
+	ID              int           `json:"id"`
+	JobID           string        `json:"jobId"`
+	StartTime       int           `json:"startTime"`
+	Duration        int           `json:"duration"`
+	Cluster         string        `json:"cluster"`
+	SubCluster      string        `json:"subCluster"`
+	NumNodes        int           `json:"numNodes"`
+	NumHWThreads    *int          `json:"numHWThreads,omitempty"`
+	NumAccelerators *int          `json:"numAccelerators,omitempty"`
+	Stats           []*NamedStats `json:"stats"`
 }
 
 type JobsStatistics struct {
-	ID             string        `json:"id"`
-	Name           string        `json:"name"`
-	TotalJobs      int           `json:"totalJobs"`
-	RunningJobs    int           `json:"runningJobs"`
-	ShortJobs      int           `json:"shortJobs"`
-	TotalWalltime  int           `json:"totalWalltime"`
-	TotalNodeHours int           `json:"totalNodeHours"`
-	TotalCoreHours int           `json:"totalCoreHours"`
-	TotalAccHours  int           `json:"totalAccHours"`
-	HistDuration   []*HistoPoint `json:"histDuration"`
-	HistNumNodes   []*HistoPoint `json:"histNumNodes"`
+	ID             string               `json:"id"`
+	Name           string               `json:"name"`
+	TotalUsers     int                  `json:"totalUsers"`
+	TotalJobs      int                  `json:"totalJobs"`
+	RunningJobs    int                  `json:"runningJobs"`
+	ShortJobs      int                  `json:"shortJobs"`
+	TotalWalltime  int                  `json:"totalWalltime"`
+	TotalNodes     int                  `json:"totalNodes"`
+	TotalNodeHours int                  `json:"totalNodeHours"`
+	TotalCores     int                  `json:"totalCores"`
+	TotalCoreHours int                  `json:"totalCoreHours"`
+	TotalAccs      int                  `json:"totalAccs"`
+	TotalAccHours  int                  `json:"totalAccHours"`
+	HistDuration   []*HistoPoint        `json:"histDuration"`
+	HistNumNodes   []*HistoPoint        `json:"histNumNodes"`
+	HistNumCores   []*HistoPoint        `json:"histNumCores"`
+	HistNumAccs    []*HistoPoint        `json:"histNumAccs"`
+	HistMetrics    []*MetricHistoPoints `json:"histMetrics"`
 }
 
 type MetricFootprints struct {
@@ -103,14 +149,83 @@ type MetricFootprints struct {
 	Data   []schema.Float `json:"data"`
 }
 
+type MetricHistoPoint struct {
+	Bin   *int `json:"bin,omitempty"`
+	Count int  `json:"count"`
+	Min   *int `json:"min,omitempty"`
+	Max   *int `json:"max,omitempty"`
+}
+
+type MetricHistoPoints struct {
+	Metric string              `json:"metric"`
+	Unit   string              `json:"unit"`
+	Stat   *string             `json:"stat,omitempty"`
+	Data   []*MetricHistoPoint `json:"data,omitempty"`
+}
+
+type MetricStatItem struct {
+	MetricName string      `json:"metricName"`
+	Range      *FloatRange `json:"range"`
+}
+
+type Mutation struct {
+}
+
+type NamedStats struct {
+	Name string                   `json:"name"`
+	Data *schema.MetricStatistics `json:"data"`
+}
+
+type NamedStatsWithScope struct {
+	Name  string             `json:"name"`
+	Scope schema.MetricScope `json:"scope"`
+	Stats []*ScopedStats     `json:"stats"`
+}
+
+type NodeFilter struct {
+	Hostname       *StringInput           `json:"hostname,omitempty"`
+	Cluster        *StringInput           `json:"cluster,omitempty"`
+	Subcluster     *StringInput           `json:"subcluster,omitempty"`
+	SchedulerState *schema.SchedulerState `json:"schedulerState,omitempty"`
+	HealthState    *string                `json:"healthState,omitempty"`
+	TimeStart      *int                   `json:"timeStart,omitempty"`
+}
+
 type NodeMetrics struct {
 	Host       string               `json:"host"`
+	State      string               `json:"state"`
 	SubCluster string               `json:"subCluster"`
 	Metrics    []*JobMetricWithName `json:"metrics"`
 }
 
+type NodeStateResultList struct {
+	Items []*schema.Node `json:"items"`
+	Count *int           `json:"count,omitempty"`
+}
+
+type NodeStates struct {
+	State string `json:"state"`
+	Count int    `json:"count"`
+}
+
+type NodeStatesTimed struct {
+	State  string `json:"state"`
+	Counts []int  `json:"counts"`
+	Times  []int  `json:"times"`
+}
+
+type NodesResultList struct {
+	Items       []*NodeMetrics `json:"items"`
+	Offset      *int           `json:"offset,omitempty"`
+	Limit       *int           `json:"limit,omitempty"`
+	Count       *int           `json:"count,omitempty"`
+	TotalNodes  *int           `json:"totalNodes,omitempty"`
+	HasNextPage *bool          `json:"hasNextPage,omitempty"`
+}
+
 type OrderByInput struct {
 	Field string            `json:"field"`
+	Type  string            `json:"type"`
 	Order SortDirectionEnum `json:"order"`
 }
 
@@ -119,18 +234,31 @@ 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"`
-	Neq        *string  `json:"neq"`
-	Contains   *string  `json:"contains"`
-	StartsWith *string  `json:"startsWith"`
-	EndsWith   *string  `json:"endsWith"`
-	In         []string `json:"in"`
+	Eq         *string  `json:"eq,omitempty"`
+	Neq        *string  `json:"neq,omitempty"`
+	Contains   *string  `json:"contains,omitempty"`
+	StartsWith *string  `json:"startsWith,omitempty"`
+	EndsWith   *string  `json:"endsWith,omitempty"`
+	In         []string `json:"in,omitempty"`
 }
 
 type TimeRangeOutput struct {
-	From time.Time `json:"from"`
-	To   time.Time `json:"to"`
+	Range *string   `json:"range,omitempty"`
+	From  time.Time `json:"from"`
+	To    time.Time `json:"to"`
+}
+
+type TimeWeights struct {
+	NodeHours []schema.Float `json:"nodeHours"`
+	AccHours  []schema.Float `json:"accHours"`
+	CoreHours []schema.Float `json:"coreHours"`
 }
 
 type User struct {
@@ -142,20 +270,22 @@ type User struct {
 type Aggregate string
 
 const (
-	AggregateUser    Aggregate = "USER"
-	AggregateProject Aggregate = "PROJECT"
-	AggregateCluster Aggregate = "CLUSTER"
+	AggregateUser       Aggregate = "USER"
+	AggregateProject    Aggregate = "PROJECT"
+	AggregateCluster    Aggregate = "CLUSTER"
+	AggregateSubcluster Aggregate = "SUBCLUSTER"
 )
 
 var AllAggregate = []Aggregate{
 	AggregateUser,
 	AggregateProject,
 	AggregateCluster,
+	AggregateSubcluster,
 }
 
 func (e Aggregate) IsValid() bool {
 	switch e {
-	case AggregateUser, AggregateProject, AggregateCluster:
+	case AggregateUser, AggregateProject, AggregateCluster, AggregateSubcluster:
 		return true
 	}
 	return false
@@ -165,7 +295,7 @@ func (e Aggregate) String() string {
 	return string(e)
 }
 
-func (e *Aggregate) UnmarshalGQL(v interface{}) error {
+func (e *Aggregate) UnmarshalGQL(v any) error {
 	str, ok := v.(string)
 	if !ok {
 		return fmt.Errorf("enums must be strings")
@@ -182,6 +312,89 @@ func (e Aggregate) MarshalGQL(w io.Writer) {
 	fmt.Fprint(w, strconv.Quote(e.String()))
 }
 
+func (e *Aggregate) UnmarshalJSON(b []byte) error {
+	s, err := strconv.Unquote(string(b))
+	if err != nil {
+		return err
+	}
+	return e.UnmarshalGQL(s)
+}
+
+func (e Aggregate) MarshalJSON() ([]byte, error) {
+	var buf bytes.Buffer
+	e.MarshalGQL(&buf)
+	return buf.Bytes(), nil
+}
+
+type SortByAggregate string
+
+const (
+	SortByAggregateTotalwalltime  SortByAggregate = "TOTALWALLTIME"
+	SortByAggregateTotaljobs      SortByAggregate = "TOTALJOBS"
+	SortByAggregateTotalusers     SortByAggregate = "TOTALUSERS"
+	SortByAggregateTotalnodes     SortByAggregate = "TOTALNODES"
+	SortByAggregateTotalnodehours SortByAggregate = "TOTALNODEHOURS"
+	SortByAggregateTotalcores     SortByAggregate = "TOTALCORES"
+	SortByAggregateTotalcorehours SortByAggregate = "TOTALCOREHOURS"
+	SortByAggregateTotalaccs      SortByAggregate = "TOTALACCS"
+	SortByAggregateTotalacchours  SortByAggregate = "TOTALACCHOURS"
+)
+
+var AllSortByAggregate = []SortByAggregate{
+	SortByAggregateTotalwalltime,
+	SortByAggregateTotaljobs,
+	SortByAggregateTotalusers,
+	SortByAggregateTotalnodes,
+	SortByAggregateTotalnodehours,
+	SortByAggregateTotalcores,
+	SortByAggregateTotalcorehours,
+	SortByAggregateTotalaccs,
+	SortByAggregateTotalacchours,
+}
+
+func (e SortByAggregate) IsValid() bool {
+	switch e {
+	case SortByAggregateTotalwalltime, SortByAggregateTotaljobs, SortByAggregateTotalusers, SortByAggregateTotalnodes, SortByAggregateTotalnodehours, SortByAggregateTotalcores, SortByAggregateTotalcorehours, SortByAggregateTotalaccs, SortByAggregateTotalacchours:
+		return true
+	}
+	return false
+}
+
+func (e SortByAggregate) String() string {
+	return string(e)
+}
+
+func (e *SortByAggregate) UnmarshalGQL(v any) error {
+	str, ok := v.(string)
+	if !ok {
+		return fmt.Errorf("enums must be strings")
+	}
+
+	*e = SortByAggregate(str)
+	if !e.IsValid() {
+		return fmt.Errorf("%s is not a valid SortByAggregate", str)
+	}
+	return nil
+}
+
+func (e SortByAggregate) MarshalGQL(w io.Writer) {
+	fmt.Fprint(w, strconv.Quote(e.String()))
+}
+
+func (e *SortByAggregate) UnmarshalJSON(b []byte) error {
+	s, err := strconv.Unquote(string(b))
+	if err != nil {
+		return err
+	}
+	return e.UnmarshalGQL(s)
+}
+
+func (e SortByAggregate) MarshalJSON() ([]byte, error) {
+	var buf bytes.Buffer
+	e.MarshalGQL(&buf)
+	return buf.Bytes(), nil
+}
+
 type SortDirectionEnum string
 
 const (
@@ -206,7 +419,7 @@ func (e SortDirectionEnum) String() string {
 	return string(e)
 }
 
-func (e *SortDirectionEnum) UnmarshalGQL(v interface{}) error {
+func (e *SortDirectionEnum) UnmarshalGQL(v any) error {
 	str, ok := v.(string)
 	if !ok {
 		return fmt.Errorf("enums must be strings")
@@ -223,43 +436,16 @@ func (e SortDirectionEnum) MarshalGQL(w io.Writer) {
 	fmt.Fprint(w, strconv.Quote(e.String()))
 }
 
-type Weights string
-
-const (
-	WeightsNodeCount Weights = "NODE_COUNT"
-	WeightsNodeHours Weights = "NODE_HOURS"
-)
-
-var AllWeights = []Weights{
-	WeightsNodeCount,
-	WeightsNodeHours,
-}
-
-func (e Weights) IsValid() bool {
-	switch e {
-	case WeightsNodeCount, WeightsNodeHours:
-		return true
+func (e *SortDirectionEnum) UnmarshalJSON(b []byte) error {
+	s, err := strconv.Unquote(string(b))
+	if err != nil {
+		return err
 	}
-	return false
+	return e.UnmarshalGQL(s)
 }
 
-func (e Weights) String() string {
-	return string(e)
-}
-
-func (e *Weights) UnmarshalGQL(v interface{}) error {
-	str, ok := v.(string)
-	if !ok {
-		return fmt.Errorf("enums must be strings")
-	}
-
-	*e = Weights(str)
-	if !e.IsValid() {
-		return fmt.Errorf("%s is not a valid Weights", str)
-	}
-	return nil
-}
-
-func (e Weights) MarshalGQL(w io.Writer) {
-	fmt.Fprint(w, strconv.Quote(e.String()))
+func (e SortDirectionEnum) MarshalJSON() ([]byte, error) {
+	var buf bytes.Buffer
+	e.MarshalGQL(&buf)
+	return buf.Bytes(), nil
 }

internal/graph/resolver.go

@@ -1,15 +1,39 @@
 package graph
 
 import (
+	"sync"
+
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
 	"github.com/jmoiron/sqlx"
 )
 
 // This file will not be regenerated automatically.
 //
 // It serves as dependency injection for your app, add any dependencies you require here.
+var (
+	initOnce         sync.Once
+	resolverInstance *Resolver
+)
 
 type Resolver struct {
 	DB   *sqlx.DB
 	Repo *repository.JobRepository
 }
+
+func Init() {
+	initOnce.Do(func() {
+		db := repository.GetConnection()
+		resolverInstance = &Resolver{
+			DB: db.DB, Repo: repository.GetJobRepository(),
+		}
+	})
+}
+
+func GetResolverInstance() *Resolver {
+	if resolverInstance == nil {
+		cclog.Fatal("Authentication module not initialized!")
+	}
+
+	return resolverInstance
+}

internal/graph/util.go

@@ -1,20 +1,21 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
 package graph
 
 import (
 	"context"
-	"errors"
 	"fmt"
 	"math"
+	"slices"
 
 	"github.com/99designs/gqlgen/graphql"
 	"github.com/ClusterCockpit/cc-backend/internal/graph/model"
-	"github.com/ClusterCockpit/cc-backend/internal/metricdata"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	"github.com/ClusterCockpit/cc-backend/internal/metricDataDispatcher"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 )
 
 const MAX_JOBS_FOR_ANALYSIS = 500
@@ -24,15 +25,15 @@ func (r *queryResolver) rooflineHeatmap(
 	ctx context.Context,
 	filter []*model.JobFilter,
 	rows int, cols int,
-	minX float64, minY float64, maxX float64, maxY float64) ([][]float64, error) {
-
+	minX float64, minY float64, maxX float64, maxY float64,
+) ([][]float64, error) {
 	jobs, err := r.Repo.QueryJobs(ctx, filter, &model.PageRequest{Page: 1, ItemsPerPage: MAX_JOBS_FOR_ANALYSIS + 1}, nil)
 	if err != nil {
-		log.Error("Error while querying jobs for roofline")
+		cclog.Error("Error while querying jobs for roofline")
 		return nil, err
 	}
 	if len(jobs) > MAX_JOBS_FOR_ANALYSIS {
-		return nil, fmt.Errorf("GRAPH/STATS > too many jobs matched (max: %d)", MAX_JOBS_FOR_ANALYSIS)
+		return nil, fmt.Errorf("GRAPH/UTIL > too many jobs matched (max: %d)", MAX_JOBS_FOR_ANALYSIS)
 	}
 
 	fcols, frows := float64(cols), float64(rows)
@@ -47,22 +48,33 @@ func (r *queryResolver) rooflineHeatmap(
 			continue
 		}
 
-		jobdata, err := metricdata.LoadData(job, []string{"flops_any", "mem_bw"}, []schema.MetricScope{schema.MetricScopeNode}, ctx)
+		// metricConfigs := archive.GetCluster(job.Cluster).MetricConfig
+		// resolution := 0
+
+		// for _, mc := range metricConfigs {
+		// 	resolution = max(resolution, mc.Timestep)
+		// }
+
+		jobdata, err := metricDataDispatcher.LoadData(job, []string{"flops_any", "mem_bw"}, []schema.MetricScope{schema.MetricScopeNode}, ctx, 0)
 		if err != nil {
-			log.Error("Error while loading metrics for roofline")
+			cclog.Errorf("Error while loading roofline metrics for job %d", job.ID)
 			return nil, err
 		}
 
 		flops_, membw_ := jobdata["flops_any"], jobdata["mem_bw"]
 		if flops_ == nil && membw_ == nil {
-			return nil, fmt.Errorf("GRAPH/STATS > 'flops_any' or 'mem_bw' missing for job %d", job.ID)
+			cclog.Infof("rooflineHeatmap(): 'flops_any' or 'mem_bw' missing for job %d", job.ID)
+			continue
+			// return nil, fmt.Errorf("GRAPH/UTIL > 'flops_any' or 'mem_bw' missing for job %d", job.ID)
 		}
 
 		flops, ok1 := flops_["node"]
 		membw, ok2 := membw_["node"]
 		if !ok1 || !ok2 {
+			cclog.Info("rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
+			continue
 			// TODO/FIXME:
-			return nil, errors.New("GRAPH/STATS > todo: rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
+			// return nil, errors.New("GRAPH/UTIL > todo: rooflineHeatmap() query not implemented for where flops_any or mem_bw not available at 'node' level")
 		}
 
 		for n := 0; n < len(flops.Series); n++ {
@@ -94,11 +106,11 @@ func (r *queryResolver) rooflineHeatmap(
 func (r *queryResolver) jobsFootprints(ctx context.Context, filter []*model.JobFilter, metrics []string) (*model.Footprints, error) {
 	jobs, err := r.Repo.QueryJobs(ctx, filter, &model.PageRequest{Page: 1, ItemsPerPage: MAX_JOBS_FOR_ANALYSIS + 1}, nil)
 	if err != nil {
-		log.Error("Error while querying jobs for footprint")
+		cclog.Error("Error while querying jobs for footprint")
 		return nil, err
 	}
 	if len(jobs) > MAX_JOBS_FOR_ANALYSIS {
-		return nil, fmt.Errorf("GRAPH/STATS > too many jobs matched (max: %d)", MAX_JOBS_FOR_ANALYSIS)
+		return nil, fmt.Errorf("GRAPH/UTIL > too many jobs matched (max: %d)", MAX_JOBS_FOR_ANALYSIS)
 	}
 
 	avgs := make([][]schema.Float, len(metrics))
@@ -106,18 +118,33 @@ func (r *queryResolver) jobsFootprints(ctx context.Context, filter []*model.JobF
 		avgs[i] = make([]schema.Float, 0, len(jobs))
 	}
 
-	nodehours := make([]schema.Float, 0, len(jobs))
+	timeweights := new(model.TimeWeights)
+	timeweights.NodeHours = make([]schema.Float, 0, len(jobs))
+	timeweights.AccHours = make([]schema.Float, 0, len(jobs))
+	timeweights.CoreHours = make([]schema.Float, 0, len(jobs))
+
 	for _, job := range jobs {
 		if job.MonitoringStatus == schema.MonitoringStatusDisabled || job.MonitoringStatus == schema.MonitoringStatusArchivingFailed {
 			continue
 		}
 
-		if err := metricdata.LoadAverages(job, metrics, avgs, ctx); err != nil {
-			log.Error("Error while loading averages for footprint")
+		if err := metricDataDispatcher.LoadAverages(job, metrics, avgs, ctx); err != nil {
+			cclog.Error("Error while loading averages for footprint")
 			return nil, err
 		}
 
-		nodehours = append(nodehours, schema.Float(float64(job.Duration)/60.0*float64(job.NumNodes)))
+		// #166 collect arrays: Null values or no null values?
+		timeweights.NodeHours = append(timeweights.NodeHours, schema.Float(float64(job.Duration)/60.0*float64(job.NumNodes)))
+		if job.NumAcc > 0 {
+			timeweights.AccHours = append(timeweights.AccHours, schema.Float(float64(job.Duration)/60.0*float64(job.NumAcc)))
+		} else {
+			timeweights.AccHours = append(timeweights.AccHours, schema.Float(1.0))
+		}
+		if job.NumHWThreads > 0 {
+			timeweights.CoreHours = append(timeweights.CoreHours, schema.Float(float64(job.Duration)/60.0*float64(job.NumHWThreads))) // SQLite HWThreads == Cores; numCoresForJob(job)
+		} else {
+			timeweights.CoreHours = append(timeweights.CoreHours, schema.Float(1.0))
+		}
 	}
 
 	res := make([]*model.MetricFootprints, len(avgs))
@@ -129,19 +156,36 @@ func (r *queryResolver) jobsFootprints(ctx context.Context, filter []*model.JobF
 	}
 
 	return &model.Footprints{
-		Nodehours: nodehours,
-		Metrics:   res,
+		TimeWeights: timeweights,
+		Metrics:     res,
 	}, nil
 }
 
+// func numCoresForJob(job *schema.Job) (numCores int) {
+
+// 	subcluster, scerr := archive.GetSubCluster(job.Cluster, job.SubCluster)
+// 	if scerr != nil {
+// 		return 1
+// 	}
+
+// 	totalJobCores := 0
+// 	topology := subcluster.Topology
+
+// 	for _, host := range job.Resources {
+// 		hwthreads := host.HWThreads
+// 		if hwthreads == nil {
+// 			hwthreads = topology.Node
+// 		}
+
+// 		hostCores, _ := topology.GetCoresFromHWThreads(hwthreads)
+// 		totalJobCores += len(hostCores)
+// 	}
+
+// 	return totalJobCores
+// }
+
 func requireField(ctx context.Context, name string) bool {
 	fields := graphql.CollectAllFields(ctx)
 
-	for _, f := range fields {
-		if f == name {
-			return true
-		}
-	}
-
-	return false
+	return slices.Contains(fields, name)
 }

internal/importer/README.md

@@ -0,0 +1,132 @@
+# Importer Package
+
+The `importer` package provides functionality for importing job data into the ClusterCockpit database from archived job files.
+
+## Overview
+
+This package supports two primary import workflows:
+
+1. **Bulk Database Initialization** - Reinitialize the entire job database from archived jobs
+2. **Individual Job Import** - Import specific jobs from metadata/data file pairs
+
+Both workflows enrich job metadata by calculating performance footprints and energy consumption metrics before persisting to the database.
+
+## Main Entry Points
+
+### InitDB()
+
+Reinitializes the job database from all archived jobs.
+
+```go
+if err := importer.InitDB(); err != nil {
+    log.Fatal(err)
+}
+```
+
+This function:
+- Flushes existing job, tag, and jobtag tables
+- Iterates through all jobs in the configured archive
+- Enriches each job with calculated metrics
+- Inserts jobs into the database in batched transactions (100 jobs per batch)
+- Continues on individual job failures, logging errors
+
+**Use Case**: Initial database setup or complete database rebuild from archive.
+
+### HandleImportFlag(flag string)
+
+Imports jobs from specified file pairs.
+
+```go
+// Format: "<meta.json>:<data.json>[,<meta2.json>:<data2.json>,...]"
+flag := "/path/to/meta.json:/path/to/data.json"
+if err := importer.HandleImportFlag(flag); err != nil {
+    log.Fatal(err)
+}
+```
+
+This function:
+- Parses the comma-separated file pairs
+- Validates metadata and job data against schemas (if validation enabled)
+- Enriches each job with footprints and energy metrics
+- Imports jobs into both the archive and database
+- Fails fast on the first error
+
+**Use Case**: Importing specific jobs from external sources or manual job additions.
+
+## Job Enrichment
+
+Both import workflows use `enrichJobMetadata()` to calculate:
+
+### Performance Footprints
+
+Performance footprints are calculated from metric averages based on the subcluster configuration:
+
+```go
+job.Footprint["mem_used_avg"] = 45.2  // GB
+job.Footprint["cpu_load_avg"] = 0.87   // percentage
+```
+
+### Energy Metrics
+
+Energy consumption is calculated from power metrics using the formula:
+
+```
+Energy (kWh) = (Power (W) × Duration (s) / 3600) / 1000
+```
+
+For each energy metric:
+```go
+job.EnergyFootprint["acc_power"] = 12.5  // kWh
+job.Energy = 150.2  // Total energy in kWh
+```
+
+**Note**: Energy calculations for metrics with unit "energy" (Joules) are not yet implemented.
+
+## Data Validation
+
+### SanityChecks(job *schema.Job)
+
+Validates job metadata before database insertion:
+
+- Cluster exists in configuration
+- Subcluster is valid (assigns if needed)
+- Job state is valid
+- Resources and user fields are populated
+- Node counts and hardware thread counts are positive
+- Resource count matches declared node count
+
+## Normalization Utilities
+
+The package includes utilities for normalizing metric values to appropriate SI prefixes:
+
+### Normalize(avg float64, prefix string)
+
+Adjusts values and SI prefixes for readability:
+
+```go
+factor, newPrefix := importer.Normalize(2048.0, "M")  
+// Converts 2048 MB → ~2.0 GB
+// Returns: factor for conversion, "G"
+```
+
+This is useful for automatically scaling metrics (e.g., memory, storage) to human-readable units.
+
+## Dependencies
+
+- `github.com/ClusterCockpit/cc-backend/internal/repository` - Database operations
+- `github.com/ClusterCockpit/cc-backend/pkg/archive` - Job archive access
+- `github.com/ClusterCockpit/cc-lib/schema` - Job schema definitions
+- `github.com/ClusterCockpit/cc-lib/ccLogger` - Logging
+- `github.com/ClusterCockpit/cc-lib/ccUnits` - SI unit handling
+
+## Error Handling
+
+- **InitDB**: Continues processing on individual job failures, logs errors, returns summary
+- **HandleImportFlag**: Fails fast on first error, returns immediately
+- Both functions log detailed error context for debugging
+
+## Performance
+
+- **Transaction Batching**: InitDB processes jobs in batches of 100 for optimal database performance
+- **Tag Caching**: Tag IDs are cached during import to minimize database queries
+- **Progress Reporting**: InitDB prints progress updates during bulk operations

internal/importer/handleImport.go

@@ -1,5 +1,5 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
 package importer
@@ -10,16 +10,30 @@ import (
 	"fmt"
 	"os"
 	"strings"
-	"time"
 
 	"github.com/ClusterCockpit/cc-backend/internal/config"
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
 	"github.com/ClusterCockpit/cc-backend/pkg/archive"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 )
 
-// Import all jobs specified as `<path-to-meta.json>:<path-to-data.json>,...`
+// HandleImportFlag imports jobs from file pairs specified in a comma-separated flag string.
+//
+// The flag format is: "<path-to-meta.json>:<path-to-data.json>[,<path-to-meta2.json>:<path-to-data2.json>,...]"
+//
+// For each job pair, this function:
+//  1. Reads and validates the metadata JSON file (schema.Job)
+//  2. Reads and validates the job data JSON file (schema.JobData)
+//  3. Enriches the job with calculated footprints and energy metrics
+//  4. Validates the job using SanityChecks()
+//  5. Imports the job into the archive
+//  6. Inserts the job into the database with associated tags
+//
+// Schema validation is performed if config.Keys.Validate is true.
+//
+// Returns an error if file reading, validation, enrichment, or database operations fail.
+// The function stops processing on the first error encountered.
 func HandleImportFlag(flag string) error {
 	r := repository.GetJobRepository()
 
@@ -31,7 +45,7 @@ func HandleImportFlag(flag string) error {
 
 		raw, err := os.ReadFile(files[0])
 		if err != nil {
-			log.Warn("Error while reading metadata file for import")
+			cclog.Warn("Error while reading metadata file for import")
 			return err
 		}
 
@@ -42,15 +56,18 @@ func HandleImportFlag(flag string) error {
 		}
 		dec := json.NewDecoder(bytes.NewReader(raw))
 		dec.DisallowUnknownFields()
-		jobMeta := schema.JobMeta{BaseJob: schema.JobDefaults}
-		if err = dec.Decode(&jobMeta); err != nil {
-			log.Warn("Error while decoding raw json metadata for import")
+		job := schema.Job{
+			Shared:           "none",
+			MonitoringStatus: schema.MonitoringStatusRunningOrArchiving,
+		}
+		if err = dec.Decode(&job); err != nil {
+			cclog.Warn("Error while decoding raw json metadata for import")
 			return err
 		}
 
 		raw, err = os.ReadFile(files[1])
 		if err != nil {
-			log.Warn("Error while reading jobdata file for import")
+			cclog.Warn("Error while reading jobdata file for import")
 			return err
 		}
 
@@ -63,72 +80,41 @@ func HandleImportFlag(flag string) error {
 		dec.DisallowUnknownFields()
 		jobData := schema.JobData{}
 		if err = dec.Decode(&jobData); err != nil {
-			log.Warn("Error while decoding raw json jobdata for import")
+			cclog.Warn("Error while decoding raw json jobdata for import")
 			return err
 		}
 
-		// checkJobData(&jobData)
+		job.MonitoringStatus = schema.MonitoringStatusArchivingSuccessful
 
-		jobMeta.MonitoringStatus = schema.MonitoringStatusArchivingSuccessful
-
-		// if _, err = r.Find(&jobMeta.JobID, &jobMeta.Cluster, &jobMeta.StartTime); err != sql.ErrNoRows {
-		// 	if err != nil {
-		// 		log.Warn("Error while finding job in jobRepository")
-		// 		return err
-		// 	}
-		//
-		// 	return fmt.Errorf("REPOSITORY/INIT > a job with that jobId, cluster and startTime does already exist")
-		// }
-		//
-		job := schema.Job{
-			BaseJob:       jobMeta.BaseJob,
-			StartTime:     time.Unix(jobMeta.StartTime, 0),
-			StartTimeUnix: jobMeta.StartTime,
-		}
-
-		// TODO: Other metrics...
-		job.LoadAvg = loadJobStat(&jobMeta, "cpu_load")
-		job.FlopsAnyAvg = loadJobStat(&jobMeta, "flops_any")
-		job.MemUsedMax = loadJobStat(&jobMeta, "mem_used")
-		job.MemBwAvg = loadJobStat(&jobMeta, "mem_bw")
-		job.NetBwAvg = loadJobStat(&jobMeta, "net_bw")
-		job.FileBwAvg = loadJobStat(&jobMeta, "file_bw")
-
-		job.RawResources, err = json.Marshal(job.Resources)
-		if err != nil {
-			log.Warn("Error while marshaling job resources")
-			return err
-		}
-		job.RawMetaData, err = json.Marshal(job.MetaData)
-		if err != nil {
-			log.Warn("Error while marshaling job metadata")
+		if err = enrichJobMetadata(&job); err != nil {
+			cclog.Errorf("Error enriching job metadata: %v", err)
 			return err
 		}
 
-		if err = SanityChecks(&job.BaseJob); err != nil {
-			log.Warn("BaseJob SanityChecks failed")
+		if err = SanityChecks(&job); err != nil {
+			cclog.Warn("BaseJob SanityChecks failed")
 			return err
 		}
 
-		if err = archive.GetHandle().ImportJob(&jobMeta, &jobData); err != nil {
-			log.Error("Error while importing job")
+		if err = archive.GetHandle().ImportJob(&job, &jobData); err != nil {
+			cclog.Error("Error while importing job")
 			return err
 		}
 
 		id, err := r.InsertJob(&job)
 		if err != nil {
-			log.Warn("Error while job db insert")
+			cclog.Warn("Error while job db insert")
 			return err
 		}
 
 		for _, tag := range job.Tags {
-			if _, err := r.AddTagOrCreate(id, tag.Type, tag.Name); err != nil {
-				log.Error("Error while adding or creating tag")
+			if err := r.ImportTag(id, tag.Type, tag.Name, tag.Scope); err != nil {
+				cclog.Error("Error while adding or creating tag on import")
 				return err
 			}
 		}
 
-		log.Infof("successfully imported a new job (jobId: %d, cluster: %s, dbid: %d)", job.JobID, job.Cluster, id)
+		cclog.Infof("successfully imported a new job (jobId: %d, cluster: %s, dbid: %d)", job.JobID, job.Cluster, id)
 	}
 	return nil
 }

internal/importer/importer_test.go

@@ -1,5 +1,5 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
 package importer_test
@@ -16,9 +16,12 @@ import (
 	"github.com/ClusterCockpit/cc-backend/internal/importer"
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
 	"github.com/ClusterCockpit/cc-backend/pkg/archive"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
+	ccconf "github.com/ClusterCockpit/cc-lib/ccConfig"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
 )
 
+// copyFile copies a file from source path to destination path.
+// Used by tests to set up test fixtures.
 func copyFile(s string, d string) error {
 	r, err := os.Open(s)
 	if err != nil {
@@ -34,10 +37,22 @@ func copyFile(s string, d string) error {
 	return nil
 }
 
+// setup initializes a test environment for importer tests.
+//
+// Creates a temporary directory with:
+//   - A test job archive with cluster configuration
+//   - A SQLite database initialized with schema
+//   - Configuration files loaded
+//
+// Returns a JobRepository instance for test assertions.
 func setup(t *testing.T) *repository.JobRepository {
 	const testconfig = `{
+		"main": {
 	"addr":            "0.0.0.0:8080",
 	"validate": false,
+  "apiAllowedIPs": [
+    "*"
+  ]},
 	"archive": {
 		"kind": "file",
 		"path": "./var/job-archive"
@@ -72,18 +87,18 @@ func setup(t *testing.T) *repository.JobRepository {
 	 }
 	]}`
 
-	log.Init("info", true)
+	cclog.Init("info", true)
 	tmpdir := t.TempDir()
 
 	jobarchive := filepath.Join(tmpdir, "job-archive")
-	if err := os.Mkdir(jobarchive, 0777); err != nil {
+	if err := os.Mkdir(jobarchive, 0o777); err != nil {
 		t.Fatal(err)
 	}
-	if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), []byte(fmt.Sprintf("%d", 1)), 0666); err != nil {
+	if err := os.WriteFile(filepath.Join(jobarchive, "version.txt"), fmt.Appendf(nil, "%d", 3), 0o666); err != nil {
 		t.Fatal(err)
 	}
 	fritzArchive := filepath.Join(tmpdir, "job-archive", "fritz")
-	if err := os.Mkdir(fritzArchive, 0777); err != nil {
+	if err := os.Mkdir(fritzArchive, 0o777); err != nil {
 		t.Fatal(err)
 	}
 	if err := copyFile(filepath.Join("testdata", "cluster-fritz.json"),
@@ -98,11 +113,23 @@ func setup(t *testing.T) *repository.JobRepository {
 	}
 
 	cfgFilePath := filepath.Join(tmpdir, "config.json")
-	if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0666); err != nil {
+	if err := os.WriteFile(cfgFilePath, []byte(testconfig), 0o666); err != nil {
 		t.Fatal(err)
 	}
 
-	config.Init(cfgFilePath)
+	ccconf.Init(cfgFilePath)
+
+	// Load and check main configuration
+	if cfg := ccconf.GetPackageConfig("main"); cfg != nil {
+		if clustercfg := ccconf.GetPackageConfig("clusters"); clustercfg != nil {
+			config.Init(cfg, clustercfg)
+		} else {
+			t.Fatal("Cluster configuration must be present")
+		}
+	} else {
+		t.Fatal("Main configuration must be present")
+	}
+
 	archiveCfg := fmt.Sprintf("{\"kind\": \"file\",\"path\": \"%s\"}", jobarchive)
 
 	if err := archive.Init(json.RawMessage(archiveCfg), config.Keys.DisableArchive); err != nil {
@@ -113,6 +140,7 @@ func setup(t *testing.T) *repository.JobRepository {
 	return repository.GetJobRepository()
 }
 
+// Result represents the expected test result for job import verification.
 type Result struct {
 	JobId     int64
 	Cluster   string
@@ -120,6 +148,8 @@ type Result struct {
 	Duration  int32
 }
 
+// readResult reads the expected test result from a golden file.
+// Golden files contain the expected job attributes after import.
 func readResult(t *testing.T, testname string) Result {
 	var r Result
 
@@ -137,6 +167,13 @@ func readResult(t *testing.T, testname string) Result {
 	return r
 }
 
+// TestHandleImportFlag tests the HandleImportFlag function with various job import scenarios.
+//
+// The test uses golden files in testdata/ to verify that jobs are correctly:
+//   - Parsed from metadata and data JSON files
+//   - Enriched with footprints and energy metrics
+//   - Inserted into the database
+//   - Retrievable with correct attributes
 func TestHandleImportFlag(t *testing.T) {
 	r := setup(t)
 
@@ -160,7 +197,7 @@ func TestHandleImportFlag(t *testing.T) {
 			}
 
 			result := readResult(t, testname)
-			job, err := r.Find(&result.JobId, &result.Cluster, &result.StartTime)
+			job, err := r.FindCached(&result.JobId, &result.Cluster, &result.StartTime)
 			if err != nil {
 				t.Fatal(err)
 			}

internal/importer/initDB.go

@@ -1,40 +1,68 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
+
+// Package importer provides functionality for importing job data into the ClusterCockpit database.
+//
+// The package supports two primary use cases:
+//  1. Bulk database initialization from archived jobs via InitDB()
+//  2. Individual job import from file pairs via HandleImportFlag()
+//
+// Both operations enrich job metadata by calculating footprints and energy metrics
+// before persisting to the database.
 package importer
 
 import (
 	"encoding/json"
 	"fmt"
+	"math"
 	"strings"
 	"time"
 
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
 	"github.com/ClusterCockpit/cc-backend/pkg/archive"
-	"github.com/ClusterCockpit/cc-backend/pkg/log"
-	"github.com/ClusterCockpit/cc-backend/pkg/schema"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
 )
 
-// Delete the tables "job", "tag" and "jobtag" from the database and
-// repopulate them using the jobs found in `archive`.
+const (
+	addTagQuery = "INSERT INTO tag (tag_name, tag_type) VALUES (?, ?)"
+	setTagQuery = "INSERT INTO jobtag (job_id, tag_id) VALUES (?, ?)"
+)
+
+// InitDB reinitializes the job database from archived job data.
+//
+// This function performs the following operations:
+//  1. Flushes existing job, tag, and jobtag tables
+//  2. Iterates through all jobs in the archive
+//  3. Enriches each job with calculated footprints and energy metrics
+//  4. Inserts jobs and tags into the database in batched transactions
+//
+// Jobs are processed in batches of 100 for optimal performance. The function
+// continues processing even if individual jobs fail, logging errors and
+// returning a summary at the end.
+//
+// Returns an error if database initialization, transaction management, or
+// critical operations fail. Individual job failures are logged but do not
+// stop the overall import process.
 func InitDB() error {
 	r := repository.GetJobRepository()
 	if err := r.Flush(); err != nil {
-		log.Errorf("repository initDB(): %v", err)
+		cclog.Errorf("repository initDB(): %v", err)
 		return err
 	}
 	starttime := time.Now()
-	log.Print("Building job table...")
+	cclog.Print("Building job table...")
 
 	t, err := r.TransactionInit()
 	if err != nil {
-		log.Warn("Error while initializing SQL transactions")
+		cclog.Warn("Error while initializing SQL transactions")
 		return err
 	}
 	tags := make(map[string]int64)
 
-	// Not using log.Print because we want the line to end with `\r` and
+	// Not using cclog.Print because we want the line to end with `\r` and
 	// this function is only ever called when a special command line flag
 	// is passed anyways.
 	fmt.Printf("%d jobs inserted...\r", 0)
@@ -46,92 +74,195 @@ func InitDB() error {
 	for jobContainer := range ar.Iter(false) {
 
 		jobMeta := jobContainer.Meta
+		if jobMeta == nil {
+			cclog.Warn("skipping job with nil metadata")
+			errorOccured++
+			continue
+		}
 
 		// Bundle 100 inserts into one transaction for better performance
 		if i%100 == 0 {
-			r.TransactionCommit(t)
+			if i > 0 {
+				if err := t.Commit(); err != nil {
+					cclog.Errorf("transaction commit error: %v", err)
+					return err
+				}
+				// Start a new transaction for the next batch
+				t, err = r.TransactionInit()
+				if err != nil {
+					cclog.Errorf("transaction init error: %v", err)
+					return err
+				}
+			}
 			fmt.Printf("%d jobs inserted...\r", i)
 		}
 
 		jobMeta.MonitoringStatus = schema.MonitoringStatusArchivingSuccessful
-		job := schema.Job{
-			BaseJob:       jobMeta.BaseJob,
-			StartTime:     time.Unix(jobMeta.StartTime, 0),
-			StartTimeUnix: jobMeta.StartTime,
-		}
 
-		// TODO: Other metrics...
-		job.LoadAvg = loadJobStat(jobMeta, "cpu_load")
-		job.FlopsAnyAvg = loadJobStat(jobMeta, "flops_any")
-		job.MemUsedMax = loadJobStat(jobMeta, "mem_used")
-		job.MemBwAvg = loadJobStat(jobMeta, "mem_bw")
-		job.NetBwAvg = loadJobStat(jobMeta, "net_bw")
-		job.FileBwAvg = loadJobStat(jobMeta, "file_bw")
-
-		job.RawResources, err = json.Marshal(job.Resources)
-		if err != nil {
-			log.Errorf("repository initDB(): %v", err)
+		if err := enrichJobMetadata(jobMeta); err != nil {
+			cclog.Errorf("repository initDB(): %v", err)
 			errorOccured++
 			continue
 		}
 
-		job.RawMetaData, err = json.Marshal(job.MetaData)
-		if err != nil {
-			log.Errorf("repository initDB(): %v", err)
+		if err := SanityChecks(jobMeta); err != nil {
+			cclog.Errorf("repository initDB(): %v", err)
 			errorOccured++
 			continue
 		}
 
-		if err := SanityChecks(&job.BaseJob); err != nil {
-			log.Errorf("repository initDB(): %v", err)
+		id, jobErr := r.TransactionAddNamed(t,
+			repository.NamedJobInsert, jobMeta)
+		if jobErr != nil {
+			cclog.Errorf("repository initDB(): %v", jobErr)
 			errorOccured++
 			continue
 		}
 
-		id, err := r.TransactionAdd(t, job)
-		if err != nil {
-			log.Errorf("repository initDB(): %v", err)
-			errorOccured++
-			continue
-		}
+		// Job successfully inserted, increment counter
+		i += 1
 
-		for _, tag := range job.Tags {
+		for _, tag := range jobMeta.Tags {
 			tagstr := tag.Name + ":" + tag.Type
-			tagId, ok := tags[tagstr]
+			tagID, ok := tags[tagstr]
 			if !ok {
-				tagId, err = r.TransactionAddTag(t, tag)
+				var err error
+				tagID, err = r.TransactionAdd(t,
+					addTagQuery,
+					tag.Name, tag.Type)
 				if err != nil {
-					log.Errorf("Error adding tag: %v", err)
+					cclog.Errorf("Error adding tag: %v", err)
 					errorOccured++
 					continue
 				}
-				tags[tagstr] = tagId
+				tags[tagstr] = tagID
 			}
 
-			r.TransactionSetTag(t, id, tagId)
-		}
-
-		if err == nil {
-			i += 1
+			r.TransactionAdd(t,
+				setTagQuery,
+				id, tagID)
 		}
 	}
 
 	if errorOccured > 0 {
-		log.Warnf("Error in import of %d jobs!", errorOccured)
+		cclog.Warnf("Error in import of %d jobs!", errorOccured)
 	}
 
 	r.TransactionEnd(t)
-	log.Printf("A total of %d jobs have been registered in %.3f seconds.\n", i, time.Since(starttime).Seconds())
+	cclog.Infof("A total of %d jobs have been registered in %.3f seconds.", i, time.Since(starttime).Seconds())
 	return nil
 }
 
-// This function also sets the subcluster if necessary!
-func SanityChecks(job *schema.BaseJob) error {
+// enrichJobMetadata calculates and populates job footprints, energy metrics, and serialized fields.
+//
+// This function performs the following enrichment operations:
+//  1. Calculates job footprint metrics based on the subcluster configuration
+//  2. Computes energy footprint and total energy consumption in kWh
+//  3. Marshals footprints, resources, and metadata into JSON for database storage
+//
+// The function expects the job's MonitoringStatus and SubCluster to be already set.
+// Energy calculations convert power metrics (Watts) to energy (kWh) using the formula:
+//
+//	Energy (kWh) = (Power (W) * Duration (s) / 3600) / 1000
+//
+// Returns an error if subcluster retrieval, metric indexing, or JSON marshaling fails.
+func enrichJobMetadata(job *schema.Job) error {
+	sc, err := archive.GetSubCluster(job.Cluster, job.SubCluster)
+	if err != nil {
+		cclog.Errorf("cannot get subcluster: %s", err.Error())
+		return err
+	}
+
+	job.Footprint = make(map[string]float64)
+
+	for _, fp := range sc.Footprint {
+		statType := "avg"
+
+		if i, err := archive.MetricIndex(sc.MetricConfig, fp); err != nil {
+			statType = sc.MetricConfig[i].Footprint
+		}
+
+		name := fmt.Sprintf("%s_%s", fp, statType)
+
+		job.Footprint[name] = repository.LoadJobStat(job, fp, statType)
+	}
+
+	job.RawFootprint, err = json.Marshal(job.Footprint)
+	if err != nil {
+		cclog.Warn("Error while marshaling job footprint")
+		return err
+	}
+
+	job.EnergyFootprint = make(map[string]float64)
+
+	// Total Job Energy Outside Loop
+	totalEnergy := 0.0
+	for _, fp := range sc.EnergyFootprint {
+		// Always Init Metric Energy Inside Loop
+		metricEnergy := 0.0
+		if i, err := archive.MetricIndex(sc.MetricConfig, fp); err == nil {
+			// Note: For DB data, calculate and save as kWh
+			switch sc.MetricConfig[i].Energy {
+			case "energy": // this metric has energy as unit (Joules)
+				cclog.Warnf("Update EnergyFootprint for Job %d and Metric %s on cluster %s: Set to 'energy' in cluster.json: Not implemented, will return 0.0", job.JobID, job.Cluster, fp)
+				// FIXME: Needs sum as stats type
+			case "power": // this metric has power as unit (Watt)
+				// Energy: Power (in Watts) * Time (in Seconds)
+				// Unit: (W * (s / 3600)) / 1000 = kWh
+				// Round 2 Digits: round(Energy * 100) / 100
+				// Here: (All-Node Metric Average * Number of Nodes) * (Job Duration in Seconds / 3600) / 1000
+				// Note: Shared Jobs handled correctly since "Node Average" is based on partial resources, while "numNodes" factor is 1
+				rawEnergy := ((repository.LoadJobStat(job, fp, "avg") * float64(job.NumNodes)) * (float64(job.Duration) / 3600.0)) / 1000.0
+				metricEnergy = math.Round(rawEnergy*100.0) / 100.0
+			}
+		} else {
+			cclog.Warnf("Error while collecting energy metric %s for job, DB ID '%v', return '0.0'", fp, job.ID)
+		}
+
+		job.EnergyFootprint[fp] = metricEnergy
+		totalEnergy += metricEnergy
+	}
+
+	job.Energy = (math.Round(totalEnergy*100.0) / 100.0)
+	if job.RawEnergyFootprint, err = json.Marshal(job.EnergyFootprint); err != nil {
+		cclog.Warnf("Error while marshaling energy footprint for job INTO BYTES, DB ID '%v'", job.ID)
+		return err
+	}
+
+	job.RawResources, err = json.Marshal(job.Resources)
+	if err != nil {
+		cclog.Warn("Error while marshaling job resources")
+		return err
+	}
+
+	job.RawMetaData, err = json.Marshal(job.MetaData)
+	if err != nil {
+		cclog.Warn("Error while marshaling job metadata")
+		return err
+	}
+
+	return nil
+}
+
+// SanityChecks validates job metadata and ensures cluster/subcluster configuration is valid.
+//
+// This function performs the following validations:
+//  1. Verifies the cluster exists in the archive configuration
+//  2. Assigns and validates the subcluster (may modify job.SubCluster)
+//  3. Validates job state is a recognized value
+//  4. Ensures resources and user fields are populated
+//  5. Validates node counts and hardware thread counts are positive
+//  6. Verifies the number of resources matches the declared node count
+//
+// The function may modify the job's SubCluster field if it needs to be assigned.
+//
+// Returns an error if any validation check fails.
+func SanityChecks(job *schema.Job) error {
 	if c := archive.GetCluster(job.Cluster); c == nil {
 		return fmt.Errorf("no such cluster: %v", job.Cluster)
 	}
 	if err := archive.AssignSubCluster(job); err != nil {
-		log.Warn("Error while assigning subcluster to job")
+		cclog.Warn("Error while assigning subcluster to job")
 		return err
 	}
 	if !job.State.Valid() {
@@ -150,18 +281,14 @@ func SanityChecks(job *schema.BaseJob) error {
 	return nil
 }
 
-func loadJobStat(job *schema.JobMeta, metric string) float64 {
-	if stats, ok := job.Statistics[metric]; ok {
-		if metric == "mem_used" {
-			return stats.Max
-		} else {
-			return stats.Avg
-		}
-	}
-
-	return 0.0
-}
-
+// checkJobData normalizes metric units in job data based on average values.
+//
+// NOTE: This function is currently unused and contains incomplete implementation.
+// It was intended to normalize byte and file-related metrics to appropriate SI prefixes,
+// but the normalization logic is commented out. Consider removing or completing this
+// function based on project requirements.
+//
+// TODO: Either implement the metric normalization or remove this dead code.
 func checkJobData(d *schema.JobData) error {
 	for _, scopes := range *d {
 		// var newUnit schema.Unit

internal/importer/normalize.go

@@ -1,5 +1,5 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
 package importer
@@ -7,13 +7,27 @@ package importer
 import (
 	"math"
 
-	ccunits "github.com/ClusterCockpit/cc-units"
+	ccunits "github.com/ClusterCockpit/cc-lib/ccUnits"
 )
 
+// getNormalizationFactor calculates the scaling factor needed to normalize a value
+// to a more readable range (typically between 1.0 and 1000.0).
+//
+// For values greater than 1000, the function scales down by factors of 1000 (returns negative exponent).
+// For values less than 1.0, the function scales up by factors of 1000 (returns positive exponent).
+//
+// Returns:
+//   - factor: The multiplicative factor to apply (10^(count*scale))
+//   - exponent: The power of 10 representing the adjustment (multiple of 3 for SI prefixes)
 func getNormalizationFactor(v float64) (float64, int) {
 	count := 0
 	scale := -3
 
+	// Prevent infinite loop for zero or negative values
+	if v <= 0.0 {
+		return 1.0, 0
+	}
+
 	if v > 1000.0 {
 		for v > 1000.0 {
 			v *= 1e-3
@@ -29,9 +43,22 @@ func getNormalizationFactor(v float64) (float64, int) {
 	return math.Pow10(count * scale), count * scale
 }
 
+// getExponent calculates the SI prefix exponent from a numeric prefix value.
+//
+// For example:
+//   - Input: 1000.0 (kilo) returns 3
+//   - Input: 1000000.0 (mega) returns 6
+//   - Input: 1000000000.0 (giga) returns 9
+//
+// Returns the exponent representing the power of 10 for the SI prefix.
 func getExponent(p float64) int {
 	count := 0
 
+	// Prevent infinite loop for infinity or NaN values
+	if math.IsInf(p, 0) || math.IsNaN(p) || p <= 0.0 {
+		return 0
+	}
+
 	for p > 1.0 {
 		p = p / 1000.0
 		count++
@@ -40,12 +67,42 @@ func getExponent(p float64) int {
 	return count * 3
 }
 
+// newPrefixFromFactor computes a new SI unit prefix after applying a normalization factor.
+//
+// Given an original prefix and an exponent adjustment, this function calculates
+// the resulting SI prefix. For example, if normalizing from bytes (no prefix) by
+// a factor of 10^9, the result would be the "G" (giga) prefix.
+//
+// Parameters:
+//   - op: The original SI prefix value
+//   - e: The exponent adjustment to apply
+//
+// Returns the new SI prefix after adjustment.
 func newPrefixFromFactor(op ccunits.Prefix, e int) ccunits.Prefix {
 	f := float64(op)
 	exp := math.Pow10(getExponent(f) - e)
 	return ccunits.Prefix(exp)
 }
 
+// Normalize adjusts a metric value and its SI unit prefix to a more readable range.
+//
+// This function is useful for automatically scaling metrics to appropriate units.
+// For example, normalizing 2048 MiB might result in ~2.0 GiB.
+//
+// The function analyzes the average value and determines if a different SI prefix
+// would make the number more human-readable (typically keeping values between 1 and 1000).
+//
+// Parameters:
+//   - avg: The metric value to normalize
+//   - p: The current SI prefix as a string (e.g., "K", "M", "G")
+//
+// Returns:
+//   - factor: The multiplicative factor to apply to convert the value
+//   - newPrefix: The new SI prefix string to use
+//
+// Example:
+//
+//	factor, newPrefix := Normalize(2048.0, "M")  // returns factor for MB->GB conversion, "G"
 func Normalize(avg float64, p string) (float64, string) {
 	f, e := getNormalizationFactor(avg)
 

internal/importer/normalize_test.go

@@ -1,5 +1,5 @@
-// Copyright (C) 2022 NHR@FAU, University Erlangen-Nuremberg.
-// All rights reserved.
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
 // Use of this source code is governed by a MIT-style
 // license that can be found in the LICENSE file.
 package importer
@@ -8,9 +8,11 @@ import (
 	"fmt"
 	"testing"
 
-	ccunits "github.com/ClusterCockpit/cc-units"
+	ccunits "github.com/ClusterCockpit/cc-lib/ccUnits"
 )
 
+// TestNormalizeFactor tests the normalization of large byte values to gigabyte prefix.
+// Verifies that values in the billions are correctly scaled to the "G" (giga) prefix.
 func TestNormalizeFactor(t *testing.T) {
 	// var us string
 	s := []float64{2890031237, 23998994567, 389734042344, 390349424345}
@@ -38,6 +40,8 @@ func TestNormalizeFactor(t *testing.T) {
 	}
 }
 
+// TestNormalizeKeep tests that values already in an appropriate range maintain their prefix.
+// Verifies that when values don't require rescaling, the original "G" prefix is preserved.
 func TestNormalizeKeep(t *testing.T) {
 	s := []float64{3.0, 24.0, 390.0, 391.0}
 

internal/importer/testdata/meta-fritzError.input

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

internal/importer/testdata/meta-fritzMinimal.input

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

internal/memorystore/api.go

@@ -0,0 +1,227 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"errors"
+	"math"
+
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/ClusterCockpit/cc-lib/util"
+)
+
+var (
+	ErrInvalidTimeRange = errors.New("[METRICSTORE]> invalid time range: 'from' must be before 'to'")
+	ErrEmptyCluster     = errors.New("[METRICSTORE]> cluster name cannot be empty")
+)
+
+type APIMetricData struct {
+	Error      *string           `json:"error,omitempty"`
+	Data       schema.FloatArray `json:"data,omitempty"`
+	From       int64             `json:"from"`
+	To         int64             `json:"to"`
+	Resolution int64             `json:"resolution"`
+	Avg        schema.Float      `json:"avg"`
+	Min        schema.Float      `json:"min"`
+	Max        schema.Float      `json:"max"`
+}
+
+type APIQueryRequest struct {
+	Cluster     string     `json:"cluster"`
+	Queries     []APIQuery `json:"queries"`
+	ForAllNodes []string   `json:"for-all-nodes"`
+	From        int64      `json:"from"`
+	To          int64      `json:"to"`
+	WithStats   bool       `json:"with-stats"`
+	WithData    bool       `json:"with-data"`
+	WithPadding bool       `json:"with-padding"`
+}
+
+type APIQueryResponse struct {
+	Queries []APIQuery        `json:"queries,omitempty"`
+	Results [][]APIMetricData `json:"results"`
+}
+
+type APIQuery struct {
+	Type        *string      `json:"type,omitempty"`
+	SubType     *string      `json:"subtype,omitempty"`
+	Metric      string       `json:"metric"`
+	Hostname    string       `json:"host"`
+	Resolution  int64        `json:"resolution"`
+	TypeIds     []string     `json:"type-ids,omitempty"`
+	SubTypeIds  []string     `json:"subtype-ids,omitempty"`
+	ScaleFactor schema.Float `json:"scale-by,omitempty"`
+	Aggregate   bool         `json:"aggreg"`
+}
+
+// TODO: Optimize this, just like the stats endpoint!
+func (data *APIMetricData) AddStats() {
+	n := 0
+	sum, min, max := 0.0, math.MaxFloat64, -math.MaxFloat64
+	for _, x := range data.Data {
+		if x.IsNaN() {
+			continue
+		}
+
+		n += 1
+		sum += float64(x)
+		min = math.Min(min, float64(x))
+		max = math.Max(max, float64(x))
+	}
+
+	if n > 0 {
+		avg := sum / float64(n)
+		data.Avg = schema.Float(avg)
+		data.Min = schema.Float(min)
+		data.Max = schema.Float(max)
+	} else {
+		data.Avg, data.Min, data.Max = schema.NaN, schema.NaN, schema.NaN
+	}
+}
+
+func (data *APIMetricData) ScaleBy(f schema.Float) {
+	if f == 0 || f == 1 {
+		return
+	}
+
+	data.Avg *= f
+	data.Min *= f
+	data.Max *= f
+	for i := 0; i < len(data.Data); i++ {
+		data.Data[i] *= f
+	}
+}
+
+func (data *APIMetricData) PadDataWithNull(ms *MemoryStore, from, to int64, metric string) {
+	minfo, ok := ms.Metrics[metric]
+	if !ok {
+		return
+	}
+
+	if (data.From / minfo.Frequency) > (from / minfo.Frequency) {
+		padfront := int((data.From / minfo.Frequency) - (from / minfo.Frequency))
+		ndata := make([]schema.Float, 0, padfront+len(data.Data))
+		for range padfront {
+			ndata = append(ndata, schema.NaN)
+		}
+		for j := 0; j < len(data.Data); j++ {
+			ndata = append(ndata, data.Data[j])
+		}
+		data.Data = ndata
+	}
+}
+
+func FetchData(req APIQueryRequest) (*APIQueryResponse, error) {
+	if req.From > req.To {
+		return nil, ErrInvalidTimeRange
+	}
+	if req.Cluster == "" && req.ForAllNodes != nil {
+		return nil, ErrEmptyCluster
+	}
+
+	req.WithData = true
+	ms := GetMemoryStore()
+
+	response := APIQueryResponse{
+		Results: make([][]APIMetricData, 0, len(req.Queries)),
+	}
+	if req.ForAllNodes != nil {
+		nodes := ms.ListChildren([]string{req.Cluster})
+		for _, node := range nodes {
+			for _, metric := range req.ForAllNodes {
+				q := APIQuery{
+					Metric:   metric,
+					Hostname: node,
+				}
+				req.Queries = append(req.Queries, q)
+				response.Queries = append(response.Queries, q)
+			}
+		}
+	}
+
+	for _, query := range req.Queries {
+		sels := make([]util.Selector, 0, 1)
+		if query.Aggregate || query.Type == nil {
+			sel := util.Selector{{String: req.Cluster}, {String: query.Hostname}}
+			if query.Type != nil {
+				if len(query.TypeIds) == 1 {
+					sel = append(sel, util.SelectorElement{String: *query.Type + query.TypeIds[0]})
+				} else {
+					ids := make([]string, len(query.TypeIds))
+					for i, id := range query.TypeIds {
+						ids[i] = *query.Type + id
+					}
+					sel = append(sel, util.SelectorElement{Group: ids})
+				}
+
+				if query.SubType != nil {
+					if len(query.SubTypeIds) == 1 {
+						sel = append(sel, util.SelectorElement{String: *query.SubType + query.SubTypeIds[0]})
+					} else {
+						ids := make([]string, len(query.SubTypeIds))
+						for i, id := range query.SubTypeIds {
+							ids[i] = *query.SubType + id
+						}
+						sel = append(sel, util.SelectorElement{Group: ids})
+					}
+				}
+			}
+			sels = append(sels, sel)
+		} else {
+			for _, typeID := range query.TypeIds {
+				if query.SubType != nil {
+					for _, subTypeID := range query.SubTypeIds {
+						sels = append(sels, util.Selector{
+							{String: req.Cluster},
+							{String: query.Hostname},
+							{String: *query.Type + typeID},
+							{String: *query.SubType + subTypeID},
+						})
+					}
+				} else {
+					sels = append(sels, util.Selector{
+						{String: req.Cluster},
+						{String: query.Hostname},
+						{String: *query.Type + typeID},
+					})
+				}
+			}
+		}
+
+		// log.Printf("query: %#v\n", query)
+		// log.Printf("sels: %#v\n", sels)
+		var err error
+		res := make([]APIMetricData, 0, len(sels))
+		for _, sel := range sels {
+			data := APIMetricData{}
+
+			data.Data, data.From, data.To, data.Resolution, err = ms.Read(sel, query.Metric, req.From, req.To, query.Resolution)
+			if err != nil {
+				msg := err.Error()
+				data.Error = &msg
+				res = append(res, data)
+				continue
+			}
+
+			if req.WithStats {
+				data.AddStats()
+			}
+			if query.ScaleFactor != 0 {
+				data.ScaleBy(query.ScaleFactor)
+			}
+			if req.WithPadding {
+				data.PadDataWithNull(ms, req.From, req.To, query.Metric)
+			}
+			if !req.WithData {
+				data.Data = nil
+			}
+			res = append(res, data)
+		}
+		response.Results = append(response.Results, res)
+	}
+
+	return &response, nil
+}

internal/memorystore/archive.go

@@ -0,0 +1,196 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"archive/zip"
+	"bufio"
+	"context"
+	"errors"
+	"fmt"
+	"io"
+	"os"
+	"path/filepath"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+)
+
+func Archiving(wg *sync.WaitGroup, ctx context.Context) {
+	go func() {
+		defer wg.Done()
+		d, err := time.ParseDuration(Keys.Archive.Interval)
+		if err != nil {
+			cclog.Fatalf("[METRICSTORE]> error parsing archive interval duration: %v\n", err)
+		}
+		if d <= 0 {
+			return
+		}
+
+		ticker := time.NewTicker(d)
+		defer ticker.Stop()
+
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case <-ticker.C:
+				t := time.Now().Add(-d)
+				cclog.Infof("[METRICSTORE]> start archiving checkpoints (older than %s)...", t.Format(time.RFC3339))
+				n, err := ArchiveCheckpoints(Keys.Checkpoints.RootDir,
+					Keys.Archive.RootDir, t.Unix(), Keys.Archive.DeleteInstead)
+
+				if err != nil {
+					cclog.Errorf("[METRICSTORE]> archiving failed: %s", err.Error())
+				} else {
+					cclog.Infof("[METRICSTORE]> done: %d files zipped and moved to archive", n)
+				}
+			}
+		}
+	}()
+}
+
+var ErrNoNewArchiveData error = errors.New("all data already archived")
+
+// ZIP all checkpoint files older than `from` together and write them to the `archiveDir`,
+// deleting them from the `checkpointsDir`.
+func ArchiveCheckpoints(checkpointsDir, archiveDir string, from int64, deleteInstead bool) (int, error) {
+	entries1, err := os.ReadDir(checkpointsDir)
+	if err != nil {
+		return 0, err
+	}
+
+	type workItem struct {
+		cdir, adir    string
+		cluster, host string
+	}
+
+	var wg sync.WaitGroup
+	n, errs := int32(0), int32(0)
+	work := make(chan workItem, Keys.NumWorkers)
+
+	wg.Add(Keys.NumWorkers)
+	for worker := 0; worker < Keys.NumWorkers; worker++ {
+		go func() {
+			defer wg.Done()
+			for workItem := range work {
+				m, err := archiveCheckpoints(workItem.cdir, workItem.adir, from, deleteInstead)
+				if err != nil {
+					cclog.Errorf("error while archiving %s/%s: %s", workItem.cluster, workItem.host, err.Error())
+					atomic.AddInt32(&errs, 1)
+				}
+				atomic.AddInt32(&n, int32(m))
+			}
+		}()
+	}
+
+	for _, de1 := range entries1 {
+		entries2, e := os.ReadDir(filepath.Join(checkpointsDir, de1.Name()))
+		if e != nil {
+			err = e
+		}
+
+		for _, de2 := range entries2 {
+			cdir := filepath.Join(checkpointsDir, de1.Name(), de2.Name())
+			adir := filepath.Join(archiveDir, de1.Name(), de2.Name())
+			work <- workItem{
+				adir: adir, cdir: cdir,
+				cluster: de1.Name(), host: de2.Name(),
+			}
+		}
+	}
+
+	close(work)
+	wg.Wait()
+
+	if err != nil {
+		return int(n), err
+	}
+
+	if errs > 0 {
+		return int(n), fmt.Errorf("%d errors happened while archiving (%d successes)", errs, n)
+	}
+	return int(n), nil
+}
+
+// Helper function for `ArchiveCheckpoints`.
+func archiveCheckpoints(dir string, archiveDir string, from int64, deleteInstead bool) (int, error) {
+	entries, err := os.ReadDir(dir)
+	if err != nil {
+		return 0, err
+	}
+
+	extension := Keys.Checkpoints.FileFormat
+	files, err := findFiles(entries, from, extension, false)
+	if err != nil {
+		return 0, err
+	}
+
+	if deleteInstead {
+		n := 0
+		for _, checkpoint := range files {
+			filename := filepath.Join(dir, checkpoint)
+			if err = os.Remove(filename); err != nil {
+				return n, err
+			}
+			n += 1
+		}
+		return n, nil
+	}
+
+	filename := filepath.Join(archiveDir, fmt.Sprintf("%d.zip", from))
+	f, err := os.OpenFile(filename, os.O_CREATE|os.O_WRONLY, CheckpointFilePerms)
+	if err != nil && os.IsNotExist(err) {
+		err = os.MkdirAll(archiveDir, CheckpointDirPerms)
+		if err == nil {
+			f, err = os.OpenFile(filename, os.O_CREATE|os.O_WRONLY, CheckpointFilePerms)
+		}
+	}
+	if err != nil {
+		return 0, err
+	}
+	defer f.Close()
+	bw := bufio.NewWriter(f)
+	defer bw.Flush()
+	zw := zip.NewWriter(bw)
+	defer zw.Close()
+
+	n := 0
+	for _, checkpoint := range files {
+		// Use closure to ensure file is closed immediately after use,
+		// avoiding file descriptor leak from defer in loop
+		err := func() error {
+			filename := filepath.Join(dir, checkpoint)
+			r, err := os.Open(filename)
+			if err != nil {
+				return err
+			}
+			defer r.Close()
+
+			w, err := zw.Create(checkpoint)
+			if err != nil {
+				return err
+			}
+
+			if _, err = io.Copy(w, r); err != nil {
+				return err
+			}
+
+			if err = os.Remove(filename); err != nil {
+				return err
+			}
+			return nil
+		}()
+		if err != nil {
+			return n, err
+		}
+		n += 1
+	}
+
+	return n, nil
+}

internal/memorystore/avroCheckpoint.go

@@ -0,0 +1,477 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"bufio"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"os"
+	"path"
+	"sort"
+	"strconv"
+	"strings"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/linkedin/goavro/v2"
+)
+
+var NumAvroWorkers int = DefaultAvroWorkers
+var startUp bool = true
+
+func (as *AvroStore) ToCheckpoint(dir string, dumpAll bool) (int, error) {
+	levels := make([]*AvroLevel, 0)
+	selectors := make([][]string, 0)
+	as.root.lock.RLock()
+	// Cluster
+	for sel1, l1 := range as.root.children {
+		l1.lock.RLock()
+		// Node
+		for sel2, l2 := range l1.children {
+			l2.lock.RLock()
+			// Frequency
+			for sel3, l3 := range l2.children {
+				levels = append(levels, l3)
+				selectors = append(selectors, []string{sel1, sel2, sel3})
+			}
+			l2.lock.RUnlock()
+		}
+		l1.lock.RUnlock()
+	}
+	as.root.lock.RUnlock()
+
+	type workItem struct {
+		level    *AvroLevel
+		dir      string
+		selector []string
+	}
+
+	n, errs := int32(0), int32(0)
+
+	var wg sync.WaitGroup
+	wg.Add(NumAvroWorkers)
+	work := make(chan workItem, NumAvroWorkers*2)
+	for range NumAvroWorkers {
+		go func() {
+			defer wg.Done()
+
+			for workItem := range work {
+				from := getTimestamp(workItem.dir)
+
+				if err := workItem.level.toCheckpoint(workItem.dir, from, dumpAll); err != nil {
+					if err == ErrNoNewArchiveData {
+						continue
+					}
+
+					cclog.Errorf("error while checkpointing %#v: %s", workItem.selector, err.Error())
+					atomic.AddInt32(&errs, 1)
+				} else {
+					atomic.AddInt32(&n, 1)
+				}
+			}
+		}()
+	}
+
+	for i := range len(levels) {
+		dir := path.Join(dir, path.Join(selectors[i]...))
+		work <- workItem{
+			level:    levels[i],
+			dir:      dir,
+			selector: selectors[i],
+		}
+	}
+
+	close(work)
+	wg.Wait()
+
+	if errs > 0 {
+		return int(n), fmt.Errorf("%d errors happend while creating avro checkpoints (%d successes)", errs, n)
+	}
+
+	startUp = false
+
+	return int(n), nil
+}
+
+// getTimestamp returns the timestamp from the directory name
+func getTimestamp(dir string) int64 {
+	// Extract the resolution and timestamp from the directory name
+	// The existing avro file will be in epoch timestamp format
+	// iterate over all the files in the directory and find the maximum timestamp
+	// and return it
+
+	resolution := path.Base(dir)
+	dir = path.Dir(dir)
+
+	files, err := os.ReadDir(dir)
+	if err != nil {
+		return 0
+	}
+	var maxTS int64 = 0
+
+	if len(files) == 0 {
+		return 0
+	}
+
+	for _, file := range files {
+		if file.IsDir() {
+			continue
+		}
+		name := file.Name()
+
+		if len(name) < 5 || !strings.HasSuffix(name, ".avro") || !strings.HasPrefix(name, resolution+"_") {
+			continue
+		}
+
+		ts, err := strconv.ParseInt(name[strings.Index(name, "_")+1:len(name)-5], 10, 64)
+		if err != nil {
+			fmt.Printf("error while parsing timestamp: %s\n", err.Error())
+			continue
+		}
+
+		if ts > maxTS {
+			maxTS = ts
+		}
+	}
+
+	interval, _ := time.ParseDuration(Keys.Checkpoints.Interval)
+	updateTime := time.Unix(maxTS, 0).Add(interval).Add(time.Duration(CheckpointBufferMinutes-1) * time.Minute).Unix()
+
+	if startUp {
+		return 0
+	}
+
+	if updateTime < time.Now().Unix() {
+		return 0
+	}
+
+	return maxTS
+}
+
+func (l *AvroLevel) toCheckpoint(dir string, from int64, dumpAll bool) error {
+	l.lock.Lock()
+	defer l.lock.Unlock()
+
+	// fmt.Printf("Checkpointing directory: %s\n", dir)
+	// filepath contains the resolution
+	intRes, _ := strconv.Atoi(path.Base(dir))
+
+	// find smallest overall timestamp in l.data map and delete it from l.data
+	minTS := int64(1<<63 - 1)
+	for ts, dat := range l.data {
+		if ts < minTS && len(dat) != 0 {
+			minTS = ts
+		}
+	}
+
+	if from == 0 && minTS != int64(1<<63-1) {
+		from = minTS
+	}
+
+	if from == 0 {
+		return ErrNoNewArchiveData
+	}
+
+	var schema string
+	var codec *goavro.Codec
+	recordList := make([]map[string]any, 0)
+
+	var f *os.File
+
+	filePath := dir + fmt.Sprintf("_%d.avro", from)
+
+	var err error
+
+	fp_, err_ := os.Stat(filePath)
+	if errors.Is(err_, os.ErrNotExist) {
+		err = os.MkdirAll(path.Dir(dir), 0o755)
+		if err != nil {
+			return fmt.Errorf("failed to create directory: %v", err)
+		}
+	} else if fp_.Size() != 0 {
+		f, err = os.Open(filePath)
+		if err != nil {
+			return fmt.Errorf("failed to open existing avro file: %v", err)
+		}
+
+		br := bufio.NewReader(f)
+
+		reader, err := goavro.NewOCFReader(br)
+		if err != nil {
+			return fmt.Errorf("failed to create OCF reader: %v", err)
+		}
+		codec = reader.Codec()
+		schema = codec.Schema()
+
+		f.Close()
+	}
+
+	timeRef := time.Now().Add(time.Duration(-CheckpointBufferMinutes+1) * time.Minute).Unix()
+
+	if dumpAll {
+		timeRef = time.Now().Unix()
+	}
+
+	// Empty values
+	if len(l.data) == 0 {
+		// we checkpoint avro files every 60 seconds
+		repeat := 60 / intRes
+
+		for range repeat {
+			recordList = append(recordList, make(map[string]any))
+		}
+	}
+
+	readFlag := true
+
+	for ts := range l.data {
+		flag := false
+		if ts < timeRef {
+			data := l.data[ts]
+
+			schemaGen, err := generateSchema(data)
+			if err != nil {
+				return err
+			}
+
+			flag, schema, err = compareSchema(schema, schemaGen)
+			if err != nil {
+				return fmt.Errorf("failed to compare read and generated schema: %v", err)
+			}
+			if flag && readFlag && !errors.Is(err_, os.ErrNotExist) {
+
+				f.Close()
+
+				f, err = os.Open(filePath)
+				if err != nil {
+					return fmt.Errorf("failed to open Avro file: %v", err)
+				}
+
+				br := bufio.NewReader(f)
+
+				ocfReader, err := goavro.NewOCFReader(br)
+				if err != nil {
+					return fmt.Errorf("failed to create OCF reader while changing schema: %v", err)
+				}
+
+				for ocfReader.Scan() {
+					record, err := ocfReader.Read()
+					if err != nil {
+						return fmt.Errorf("failed to read record: %v", err)
+					}
+
+					recordList = append(recordList, record.(map[string]any))
+				}
+
+				f.Close()
+
+				err = os.Remove(filePath)
+				if err != nil {
+					return fmt.Errorf("failed to delete file: %v", err)
+				}
+
+				readFlag = false
+			}
+			codec, err = goavro.NewCodec(schema)
+			if err != nil {
+				return fmt.Errorf("failed to create codec after merged schema: %v", err)
+			}
+
+			recordList = append(recordList, generateRecord(data))
+			delete(l.data, ts)
+		}
+	}
+
+	if len(recordList) == 0 {
+		return ErrNoNewArchiveData
+	}
+
+	f, err = os.OpenFile(filePath, os.O_CREATE|os.O_APPEND|os.O_RDWR, 0o644)
+	if err != nil {
+		return fmt.Errorf("failed to append new avro file: %v", err)
+	}
+
+	// fmt.Printf("Codec : %#v\n", codec)
+
+	writer, err := goavro.NewOCFWriter(goavro.OCFConfig{
+		W:               f,
+		Codec:           codec,
+		CompressionName: goavro.CompressionDeflateLabel,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to create OCF writer: %v", err)
+	}
+
+	// Append the new record
+	if err := writer.Append(recordList); err != nil {
+		return fmt.Errorf("failed to append record: %v", err)
+	}
+
+	f.Close()
+
+	return nil
+}
+
+func compareSchema(schemaRead, schemaGen string) (bool, string, error) {
+	var genSchema, readSchema AvroSchema
+
+	if schemaRead == "" {
+		return false, schemaGen, nil
+	}
+
+	// Unmarshal the schema strings into AvroSchema structs
+	if err := json.Unmarshal([]byte(schemaGen), &genSchema); err != nil {
+		return false, "", fmt.Errorf("failed to parse generated schema: %v", err)
+	}
+	if err := json.Unmarshal([]byte(schemaRead), &readSchema); err != nil {
+		return false, "", fmt.Errorf("failed to parse read schema: %v", err)
+	}
+
+	sort.Slice(genSchema.Fields, func(i, j int) bool {
+		return genSchema.Fields[i].Name < genSchema.Fields[j].Name
+	})
+
+	sort.Slice(readSchema.Fields, func(i, j int) bool {
+		return readSchema.Fields[i].Name < readSchema.Fields[j].Name
+	})
+
+	// Check if schemas are identical
+	schemasEqual := true
+	if len(genSchema.Fields) <= len(readSchema.Fields) {
+
+		for i := range genSchema.Fields {
+			if genSchema.Fields[i].Name != readSchema.Fields[i].Name {
+				schemasEqual = false
+				break
+			}
+		}
+
+		// If schemas are identical, return the read schema
+		if schemasEqual {
+			return false, schemaRead, nil
+		}
+	}
+
+	// Create a map to hold unique fields from both schemas
+	fieldMap := make(map[string]AvroField)
+
+	// Add fields from the read schema
+	for _, field := range readSchema.Fields {
+		fieldMap[field.Name] = field
+	}
+
+	// Add or update fields from the generated schema
+	for _, field := range genSchema.Fields {
+		fieldMap[field.Name] = field
+	}
+
+	// Create a union schema by collecting fields from the map
+	var mergedFields []AvroField
+	for _, field := range fieldMap {
+		mergedFields = append(mergedFields, field)
+	}
+
+	// Sort fields by name for consistency
+	sort.Slice(mergedFields, func(i, j int) bool {
+		return mergedFields[i].Name < mergedFields[j].Name
+	})
+
+	// Create the merged schema
+	mergedSchema := AvroSchema{
+		Type:   "record",
+		Name:   genSchema.Name,
+		Fields: mergedFields,
+	}
+
+	// Check if schemas are identical
+	schemasEqual = len(mergedSchema.Fields) == len(readSchema.Fields)
+	if schemasEqual {
+		for i := range mergedSchema.Fields {
+			if mergedSchema.Fields[i].Name != readSchema.Fields[i].Name {
+				schemasEqual = false
+				break
+			}
+		}
+
+		if schemasEqual {
+			return false, schemaRead, nil
+		}
+	}
+
+	// Marshal the merged schema back to JSON
+	mergedSchemaJSON, err := json.Marshal(mergedSchema)
+	if err != nil {
+		return false, "", fmt.Errorf("failed to marshal merged schema: %v", err)
+	}
+
+	return true, string(mergedSchemaJSON), nil
+}
+
+func generateSchema(data map[string]schema.Float) (string, error) {
+	// Define the Avro schema structure
+	schema := map[string]any{
+		"type":   "record",
+		"name":   "DataRecord",
+		"fields": []map[string]any{},
+	}
+
+	fieldTracker := make(map[string]struct{})
+
+	for key := range data {
+		if _, exists := fieldTracker[key]; !exists {
+			key = correctKey(key)
+
+			field := map[string]any{
+				"name":    key,
+				"type":    "double",
+				"default": -1.0,
+			}
+			schema["fields"] = append(schema["fields"].([]map[string]any), field)
+			fieldTracker[key] = struct{}{}
+		}
+	}
+
+	schemaString, err := json.Marshal(schema)
+	if err != nil {
+		return "", fmt.Errorf("failed to marshal schema: %v", err)
+	}
+
+	return string(schemaString), nil
+}
+
+func generateRecord(data map[string]schema.Float) map[string]any {
+	record := make(map[string]any)
+
+	// Iterate through each map in data
+	for key, value := range data {
+		key = correctKey(key)
+
+		// Set the value in the record
+		// avro only accepts basic types
+		record[key] = value.Double()
+	}
+
+	return record
+}
+
+func correctKey(key string) string {
+	key = strings.ReplaceAll(key, "_", "_0x5F_")
+	key = strings.ReplaceAll(key, ":", "_0x3A_")
+	key = strings.ReplaceAll(key, ".", "_0x2E_")
+	return key
+}
+
+func ReplaceKey(key string) string {
+	key = strings.ReplaceAll(key, "_0x2E_", ".")
+	key = strings.ReplaceAll(key, "_0x3A_", ":")
+	key = strings.ReplaceAll(key, "_0x5F_", "_")
+	return key
+}

internal/memorystore/avroHelper.go

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

internal/memorystore/avroStruct.go

@@ -0,0 +1,167 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"sync"
+
+	"github.com/ClusterCockpit/cc-lib/schema"
+)
+
+var (
+	LineProtocolMessages = make(chan *AvroStruct)
+	// SelectorDelimiter separates hierarchical selector components in metric names for Avro encoding
+	SelectorDelimiter = "_SEL_"
+)
+
+var CheckpointBufferMinutes = DefaultCheckpointBufferMin
+
+type AvroStruct struct {
+	MetricName string
+	Cluster    string
+	Node       string
+	Selector   []string
+	Value      schema.Float
+	Timestamp  int64
+}
+
+type AvroStore struct {
+	root AvroLevel
+}
+
+var avroStore AvroStore
+
+type AvroLevel struct {
+	children map[string]*AvroLevel
+	data     map[int64]map[string]schema.Float
+	lock     sync.RWMutex
+}
+
+type AvroField struct {
+	Name    string `json:"name"`
+	Type    any    `json:"type"`
+	Default any    `json:"default,omitempty"`
+}
+
+type AvroSchema struct {
+	Type   string      `json:"type"`
+	Name   string      `json:"name"`
+	Fields []AvroField `json:"fields"`
+}
+
+func (l *AvroLevel) findAvroLevelOrCreate(selector []string) *AvroLevel {
+	if len(selector) == 0 {
+		return l
+	}
+
+	// Allow concurrent reads:
+	l.lock.RLock()
+	var child *AvroLevel
+	var ok bool
+	if l.children == nil {
+		// Children map needs to be created...
+		l.lock.RUnlock()
+	} else {
+		child, ok := l.children[selector[0]]
+		l.lock.RUnlock()
+		if ok {
+			return child.findAvroLevelOrCreate(selector[1:])
+		}
+	}
+
+	// The level does not exist, take write lock for unique access:
+	l.lock.Lock()
+	// While this thread waited for the write lock, another thread
+	// could have created the child node.
+	if l.children != nil {
+		child, ok = l.children[selector[0]]
+		if ok {
+			l.lock.Unlock()
+			return child.findAvroLevelOrCreate(selector[1:])
+		}
+	}
+
+	child = &AvroLevel{
+		data:     make(map[int64]map[string]schema.Float, 0),
+		children: nil,
+	}
+
+	if l.children != nil {
+		l.children[selector[0]] = child
+	} else {
+		l.children = map[string]*AvroLevel{selector[0]: child}
+	}
+	l.lock.Unlock()
+	return child.findAvroLevelOrCreate(selector[1:])
+}
+
+func (l *AvroLevel) addMetric(metricName string, value schema.Float, timestamp int64, Freq int) {
+	l.lock.Lock()
+	defer l.lock.Unlock()
+
+	KeyCounter := int(CheckpointBufferMinutes * 60 / Freq)
+
+	// Create keys in advance for the given amount of time
+	if len(l.data) != KeyCounter {
+		if len(l.data) == 0 {
+			for i := range KeyCounter {
+				l.data[timestamp+int64(i*Freq)] = make(map[string]schema.Float, 0)
+			}
+		} else {
+			// Get the last timestamp
+			var lastTS int64
+			for ts := range l.data {
+				if ts > lastTS {
+					lastTS = ts
+				}
+			}
+			// Create keys for the next KeyCounter timestamps
+			l.data[lastTS+int64(Freq)] = make(map[string]schema.Float, 0)
+		}
+	}
+
+	closestTS := int64(0)
+	minDiff := int64(Freq) + 1 // Start with diff just outside the valid range
+	found := false
+
+	// Iterate over timestamps and choose the one which is within range.
+	// Since its epoch time, we check if the difference is less than 60 seconds.
+	for ts, dat := range l.data {
+		// Check if timestamp is within range
+		diff := timestamp - ts
+		if diff < -int64(Freq) || diff > int64(Freq) {
+			continue
+		}
+
+		// Metric already present at this timestamp — skip
+		if _, ok := dat[metricName]; ok {
+			continue
+		}
+
+		// Check if this is the closest timestamp so far
+		if Abs(diff) < minDiff {
+			minDiff = Abs(diff)
+			closestTS = ts
+			found = true
+		}
+	}
+
+	if found {
+		l.data[closestTS][metricName] = value
+	}
+}
+
+func GetAvroStore() *AvroStore {
+	return &avroStore
+}
+
+// Abs returns the absolute value of x.
+func Abs(x int64) int64 {
+	if x < 0 {
+		return -x
+	}
+	return x
+}

internal/memorystore/buffer.go

@@ -0,0 +1,190 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"errors"
+	"sync"
+
+	"github.com/ClusterCockpit/cc-lib/schema"
+)
+
+// BufferCap is the default buffer capacity.
+// buffer.data will only ever grow up to its capacity and a new link
+// in the buffer chain will be created if needed so that no copying
+// of data or reallocation needs to happen on writes.
+const BufferCap int = DefaultBufferCapacity
+
+var bufferPool sync.Pool = sync.Pool{
+	New: func() any {
+		return &buffer{
+			data: make([]schema.Float, 0, BufferCap),
+		}
+	},
+}
+
+var (
+	ErrNoData           error = errors.New("[METRICSTORE]> no data for this metric/level")
+	ErrDataDoesNotAlign error = errors.New("[METRICSTORE]> data from lower granularities does not align")
+)
+
+// Each metric on each level has it's own buffer.
+// This is where the actual values go.
+// If `cap(data)` is reached, a new buffer is created and
+// becomes the new head of a buffer list.
+type buffer struct {
+	prev      *buffer
+	next      *buffer
+	data      []schema.Float
+	frequency int64
+	start     int64
+	archived  bool
+	closed    bool
+}
+
+func newBuffer(ts, freq int64) *buffer {
+	b := bufferPool.Get().(*buffer)
+	b.frequency = freq
+	b.start = ts - (freq / 2)
+	b.prev = nil
+	b.next = nil
+	b.archived = false
+	b.closed = false
+	b.data = b.data[:0]
+	return b
+}
+
+// If a new buffer was created, the new head is returnd.
+// Otherwise, the existing buffer is returnd.
+// Normaly, only "newer" data should be written, but if the value would
+// end up in the same buffer anyways it is allowed.
+func (b *buffer) write(ts int64, value schema.Float) (*buffer, error) {
+	if ts < b.start {
+		return nil, errors.New("[METRICSTORE]> cannot write value to buffer from past")
+	}
+
+	// idx := int((ts - b.start + (b.frequency / 3)) / b.frequency)
+	idx := int((ts - b.start) / b.frequency)
+	if idx >= cap(b.data) {
+		newbuf := newBuffer(ts, b.frequency)
+		newbuf.prev = b
+		b.next = newbuf
+		b = newbuf
+		idx = 0
+	}
+
+	// Overwriting value or writing value from past
+	if idx < len(b.data) {
+		b.data[idx] = value
+		return b, nil
+	}
+
+	// Fill up unwritten slots with NaN
+	for i := len(b.data); i < idx; i++ {
+		b.data = append(b.data, schema.NaN)
+	}
+
+	b.data = append(b.data, value)
+	return b, nil
+}
+
+func (b *buffer) end() int64 {
+	return b.firstWrite() + int64(len(b.data))*b.frequency
+}
+
+func (b *buffer) firstWrite() int64 {
+	return b.start + (b.frequency / 2)
+}
+
+// Return all known values from `from` to `to`. Gaps of information are represented as NaN.
+// Simple linear interpolation is done between the two neighboring cells if possible.
+// If values at the start or end are missing, instead of NaN values, the second and thrid
+// return values contain the actual `from`/`to`.
+// This function goes back the buffer chain if `from` is older than the currents buffer start.
+// The loaded values are added to `data` and `data` is returned, possibly with a shorter length.
+// If `data` is not long enough to hold all values, this function will panic!
+func (b *buffer) read(from, to int64, data []schema.Float) ([]schema.Float, int64, int64, error) {
+	if from < b.firstWrite() {
+		if b.prev != nil {
+			return b.prev.read(from, to, data)
+		}
+		from = b.firstWrite()
+	}
+
+	i := 0
+	t := from
+	for ; t < to; t += b.frequency {
+		idx := int((t - b.start) / b.frequency)
+		if idx >= cap(b.data) {
+			if b.next == nil {
+				break
+			}
+			b = b.next
+			idx = 0
+		}
+
+		if idx >= len(b.data) {
+			if b.next == nil || to <= b.next.start {
+				break
+			}
+			data[i] += schema.NaN
+		} else if t < b.start {
+			data[i] += schema.NaN
+		} else {
+			data[i] += b.data[idx]
+		}
+		i++
+	}
+
+	return data[:i], from, t, nil
+}
+
+// Returns true if this buffer needs to be freed.
+func (b *buffer) free(t int64) (delme bool, n int) {
+	if b.prev != nil {
+		delme, m := b.prev.free(t)
+		n += m
+		if delme {
+			b.prev.next = nil
+			if cap(b.prev.data) == BufferCap {
+				bufferPool.Put(b.prev)
+			}
+			b.prev = nil
+		}
+	}
+
+	end := b.end()
+	if end < t {
+		return true, n + 1
+	}
+
+	return false, n
+}
+
+// Call `callback` on every buffer that contains data in the range from `from` to `to`.
+func (b *buffer) iterFromTo(from, to int64, callback func(b *buffer) error) error {
+	if b == nil {
+		return nil
+	}
+
+	if err := b.prev.iterFromTo(from, to, callback); err != nil {
+		return err
+	}
+
+	if from <= b.end() && b.start <= to {
+		return callback(b)
+	}
+
+	return nil
+}
+
+func (b *buffer) count() int64 {
+	res := int64(len(b.data))
+	if b.prev != nil {
+		res += b.prev.count()
+	}
+	return res
+}

internal/memorystore/checkpoint.go

@@ -0,0 +1,761 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"bufio"
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io/fs"
+	"os"
+	"path"
+	"path/filepath"
+	"runtime"
+	"sort"
+	"strconv"
+	"strings"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/linkedin/goavro/v2"
+)
+
+const (
+	CheckpointFilePerms = 0o644
+	CheckpointDirPerms  = 0o755
+	GCTriggerInterval   = DefaultGCTriggerInterval
+)
+
+// Whenever changed, update MarshalJSON as well!
+type CheckpointMetrics struct {
+	Data      []schema.Float `json:"data"`
+	Frequency int64          `json:"frequency"`
+	Start     int64          `json:"start"`
+}
+
+type CheckpointFile struct {
+	Metrics  map[string]*CheckpointMetrics `json:"metrics"`
+	Children map[string]*CheckpointFile    `json:"children"`
+	From     int64                         `json:"from"`
+	To       int64                         `json:"to"`
+}
+
+var lastCheckpoint time.Time
+
+func Checkpointing(wg *sync.WaitGroup, ctx context.Context) {
+	lastCheckpoint = time.Now()
+
+	if Keys.Checkpoints.FileFormat == "json" {
+		ms := GetMemoryStore()
+
+		go func() {
+			defer wg.Done()
+			d, err := time.ParseDuration(Keys.Checkpoints.Interval)
+			if err != nil {
+				cclog.Fatal(err)
+			}
+			if d <= 0 {
+				return
+			}
+
+			ticker := time.NewTicker(d)
+			defer ticker.Stop()
+
+			for {
+				select {
+				case <-ctx.Done():
+					return
+				case <-ticker.C:
+					cclog.Infof("[METRICSTORE]> start checkpointing (starting at %s)...", lastCheckpoint.Format(time.RFC3339))
+					now := time.Now()
+					n, err := ms.ToCheckpoint(Keys.Checkpoints.RootDir,
+						lastCheckpoint.Unix(), now.Unix())
+					if err != nil {
+						cclog.Errorf("[METRICSTORE]> checkpointing failed: %s", err.Error())
+					} else {
+						cclog.Infof("[METRICSTORE]> done: %d checkpoint files created", n)
+						lastCheckpoint = now
+					}
+				}
+			}
+		}()
+	} else {
+		go func() {
+			defer wg.Done()
+
+			select {
+			case <-ctx.Done():
+				return
+			case <-time.After(time.Duration(CheckpointBufferMinutes) * time.Minute):
+				GetAvroStore().ToCheckpoint(Keys.Checkpoints.RootDir, false)
+			}
+
+			ticker := time.NewTicker(DefaultAvroCheckpointInterval)
+			defer ticker.Stop()
+
+			for {
+				select {
+				case <-ctx.Done():
+					return
+				case <-ticker.C:
+					GetAvroStore().ToCheckpoint(Keys.Checkpoints.RootDir, false)
+				}
+			}
+		}()
+	}
+}
+
+// As `Float` implements a custom MarshalJSON() function,
+// serializing an array of such types has more overhead
+// than one would assume (because of extra allocations, interfaces and so on).
+func (cm *CheckpointMetrics) MarshalJSON() ([]byte, error) {
+	buf := make([]byte, 0, 128+len(cm.Data)*8)
+	buf = append(buf, `{"frequency":`...)
+	buf = strconv.AppendInt(buf, cm.Frequency, 10)
+	buf = append(buf, `,"start":`...)
+	buf = strconv.AppendInt(buf, cm.Start, 10)
+	buf = append(buf, `,"data":[`...)
+	for i, x := range cm.Data {
+		if i != 0 {
+			buf = append(buf, ',')
+		}
+		if x.IsNaN() {
+			buf = append(buf, `null`...)
+		} else {
+			buf = strconv.AppendFloat(buf, float64(x), 'f', 1, 32)
+		}
+	}
+	buf = append(buf, `]}`...)
+	return buf, nil
+}
+
+// Metrics stored at the lowest 2 levels are not stored away (root and cluster)!
+// On a per-host basis a new JSON file is created. I have no idea if this will scale.
+// The good thing: Only a host at a time is locked, so this function can run
+// in parallel to writes/reads.
+func (m *MemoryStore) ToCheckpoint(dir string, from, to int64) (int, error) {
+	levels := make([]*Level, 0)
+	selectors := make([][]string, 0)
+	m.root.lock.RLock()
+	for sel1, l1 := range m.root.children {
+		l1.lock.RLock()
+		for sel2, l2 := range l1.children {
+			levels = append(levels, l2)
+			selectors = append(selectors, []string{sel1, sel2})
+		}
+		l1.lock.RUnlock()
+	}
+	m.root.lock.RUnlock()
+
+	type workItem struct {
+		level    *Level
+		dir      string
+		selector []string
+	}
+
+	n, errs := int32(0), int32(0)
+
+	var wg sync.WaitGroup
+	wg.Add(Keys.NumWorkers)
+	work := make(chan workItem, Keys.NumWorkers*2)
+	for worker := 0; worker < Keys.NumWorkers; worker++ {
+		go func() {
+			defer wg.Done()
+
+			for workItem := range work {
+				if err := workItem.level.toCheckpoint(workItem.dir, from, to, m); err != nil {
+					if err == ErrNoNewArchiveData {
+						continue
+					}
+
+					cclog.Errorf("[METRICSTORE]> error while checkpointing %#v: %s", workItem.selector, err.Error())
+					atomic.AddInt32(&errs, 1)
+				} else {
+					atomic.AddInt32(&n, 1)
+				}
+			}
+		}()
+	}
+
+	for i := 0; i < len(levels); i++ {
+		dir := path.Join(dir, path.Join(selectors[i]...))
+		work <- workItem{
+			level:    levels[i],
+			dir:      dir,
+			selector: selectors[i],
+		}
+	}
+
+	close(work)
+	wg.Wait()
+
+	if errs > 0 {
+		return int(n), fmt.Errorf("[METRICSTORE]> %d errors happened while creating checkpoints (%d successes)", errs, n)
+	}
+	return int(n), nil
+}
+
+func (l *Level) toCheckpointFile(from, to int64, m *MemoryStore) (*CheckpointFile, error) {
+	l.lock.RLock()
+	defer l.lock.RUnlock()
+
+	retval := &CheckpointFile{
+		From:     from,
+		To:       to,
+		Metrics:  make(map[string]*CheckpointMetrics),
+		Children: make(map[string]*CheckpointFile),
+	}
+
+	for metric, minfo := range m.Metrics {
+		b := l.metrics[minfo.offset]
+		if b == nil {
+			continue
+		}
+
+		allArchived := true
+		b.iterFromTo(from, to, func(b *buffer) error {
+			if !b.archived {
+				allArchived = false
+			}
+			return nil
+		})
+
+		if allArchived {
+			continue
+		}
+
+		data := make([]schema.Float, (to-from)/b.frequency+1)
+		data, start, end, err := b.read(from, to, data)
+		if err != nil {
+			return nil, err
+		}
+
+		for i := int((end - start) / b.frequency); i < len(data); i++ {
+			data[i] = schema.NaN
+		}
+
+		retval.Metrics[metric] = &CheckpointMetrics{
+			Frequency: b.frequency,
+			Start:     start,
+			Data:      data,
+		}
+	}
+
+	for name, child := range l.children {
+		val, err := child.toCheckpointFile(from, to, m)
+		if err != nil {
+			return nil, err
+		}
+
+		if val != nil {
+			retval.Children[name] = val
+		}
+	}
+
+	if len(retval.Children) == 0 && len(retval.Metrics) == 0 {
+		return nil, nil
+	}
+
+	return retval, nil
+}
+
+func (l *Level) toCheckpoint(dir string, from, to int64, m *MemoryStore) error {
+	cf, err := l.toCheckpointFile(from, to, m)
+	if err != nil {
+		return err
+	}
+
+	if cf == nil {
+		return ErrNoNewArchiveData
+	}
+
+	filepath := path.Join(dir, fmt.Sprintf("%d.json", from))
+	f, err := os.OpenFile(filepath, os.O_CREATE|os.O_WRONLY, CheckpointFilePerms)
+	if err != nil && os.IsNotExist(err) {
+		err = os.MkdirAll(dir, CheckpointDirPerms)
+		if err == nil {
+			f, err = os.OpenFile(filepath, os.O_CREATE|os.O_WRONLY, CheckpointFilePerms)
+		}
+	}
+	if err != nil {
+		return err
+	}
+	defer f.Close()
+
+	bw := bufio.NewWriter(f)
+	if err = json.NewEncoder(bw).Encode(cf); err != nil {
+		return err
+	}
+
+	return bw.Flush()
+}
+
+func (m *MemoryStore) FromCheckpoint(dir string, from int64, extension string) (int, error) {
+	var wg sync.WaitGroup
+	work := make(chan [2]string, Keys.NumWorkers)
+	n, errs := int32(0), int32(0)
+
+	wg.Add(Keys.NumWorkers)
+	for worker := 0; worker < Keys.NumWorkers; worker++ {
+		go func() {
+			defer wg.Done()
+			for host := range work {
+				lvl := m.root.findLevelOrCreate(host[:], len(m.Metrics))
+				nn, err := lvl.fromCheckpoint(m, filepath.Join(dir, host[0], host[1]), from, extension)
+				if err != nil {
+					cclog.Errorf("[METRICSTORE]> error while loading checkpoints for %s/%s: %s", host[0], host[1], err.Error())
+					atomic.AddInt32(&errs, 1)
+				}
+				atomic.AddInt32(&n, int32(nn))
+			}
+		}()
+	}
+
+	i := 0
+	clustersDir, err := os.ReadDir(dir)
+	for _, clusterDir := range clustersDir {
+		if !clusterDir.IsDir() {
+			err = errors.New("[METRICSTORE]> expected only directories at first level of checkpoints/ directory")
+			goto done
+		}
+
+		hostsDir, e := os.ReadDir(filepath.Join(dir, clusterDir.Name()))
+		if e != nil {
+			err = e
+			goto done
+		}
+
+		for _, hostDir := range hostsDir {
+			if !hostDir.IsDir() {
+				err = errors.New("[METRICSTORE]> expected only directories at second level of checkpoints/ directory")
+				goto done
+			}
+
+			i++
+			if i%Keys.NumWorkers == 0 && i > GCTriggerInterval {
+				// Forcing garbage collection runs here regulary during the loading of checkpoints
+				// will decrease the total heap size after loading everything back to memory is done.
+				// While loading data, the heap will grow fast, so the GC target size will double
+				// almost always. By forcing GCs here, we can keep it growing more slowly so that
+				// at the end, less memory is wasted.
+				runtime.GC()
+			}
+
+			work <- [2]string{clusterDir.Name(), hostDir.Name()}
+		}
+	}
+done:
+	close(work)
+	wg.Wait()
+
+	if err != nil {
+		return int(n), err
+	}
+
+	if errs > 0 {
+		return int(n), fmt.Errorf("[METRICSTORE]> %d errors happened while creating checkpoints (%d successes)", errs, n)
+	}
+	return int(n), nil
+}
+
+// Metrics stored at the lowest 2 levels are not loaded (root and cluster)!
+// This function can only be called once and before the very first write or read.
+// Different host's data is loaded to memory in parallel.
+func (m *MemoryStore) FromCheckpointFiles(dir string, from int64) (int, error) {
+	if _, err := os.Stat(dir); os.IsNotExist(err) {
+		// The directory does not exist, so create it using os.MkdirAll()
+		err := os.MkdirAll(dir, CheckpointDirPerms) // CheckpointDirPerms sets the permissions for the directory
+		if err != nil {
+			cclog.Fatalf("[METRICSTORE]> Error creating directory: %#v\n", err)
+		}
+		cclog.Debugf("[METRICSTORE]> %#v Directory created successfully", dir)
+	}
+
+	// Config read (replace with your actual config read)
+	fileFormat := Keys.Checkpoints.FileFormat
+	if fileFormat == "" {
+		fileFormat = "avro"
+	}
+
+	// Map to easily get the fallback format
+	oppositeFormat := map[string]string{
+		"json": "avro",
+		"avro": "json",
+	}
+
+	// First, attempt to load the specified format
+	if found, err := checkFilesWithExtension(dir, fileFormat); err != nil {
+		return 0, fmt.Errorf("[METRICSTORE]> error checking files with extension: %v", err)
+	} else if found {
+		cclog.Infof("[METRICSTORE]> Loading %s files because fileformat is %s", fileFormat, fileFormat)
+		return m.FromCheckpoint(dir, from, fileFormat)
+	}
+
+	// If not found, attempt the opposite format
+	altFormat := oppositeFormat[fileFormat]
+	if found, err := checkFilesWithExtension(dir, altFormat); err != nil {
+		return 0, fmt.Errorf("[METRICSTORE]> error checking files with extension: %v", err)
+	} else if found {
+		cclog.Infof("[METRICSTORE]> Loading %s files but fileformat is %s", altFormat, fileFormat)
+		return m.FromCheckpoint(dir, from, altFormat)
+	}
+
+	cclog.Print("[METRICSTORE]> No valid checkpoint files found in the directory")
+	return 0, nil
+}
+
+func checkFilesWithExtension(dir string, extension string) (bool, error) {
+	found := false
+
+	err := filepath.Walk(dir, func(path string, info os.FileInfo, err error) error {
+		if err != nil {
+			return fmt.Errorf("[METRICSTORE]> error accessing path %s: %v", path, err)
+		}
+		if !info.IsDir() && filepath.Ext(info.Name()) == "."+extension {
+			found = true
+			return nil
+		}
+		return nil
+	})
+	if err != nil {
+		return false, fmt.Errorf("[METRICSTORE]> error walking through directories: %s", err)
+	}
+
+	return found, nil
+}
+
+func (l *Level) loadAvroFile(m *MemoryStore, f *os.File, from int64) error {
+	br := bufio.NewReader(f)
+
+	fileName := f.Name()[strings.LastIndex(f.Name(), "/")+1:]
+	resolution, err := strconv.ParseInt(fileName[0:strings.Index(fileName, "_")], 10, 64)
+	if err != nil {
+		return fmt.Errorf("[METRICSTORE]> error while reading avro file (resolution parsing) : %s", err)
+	}
+
+	fromTimestamp, err := strconv.ParseInt(fileName[strings.Index(fileName, "_")+1:len(fileName)-5], 10, 64)
+
+	// Same logic according to lineprotocol
+	fromTimestamp -= (resolution / 2)
+
+	if err != nil {
+		return fmt.Errorf("[METRICSTORE]> error converting timestamp from the avro file : %s", err)
+	}
+
+	// fmt.Printf("File : %s with resolution : %d\n", fileName, resolution)
+
+	var recordCounter int64 = 0
+
+	// Create a new OCF reader from the buffered reader
+	ocfReader, err := goavro.NewOCFReader(br)
+	if err != nil {
+		return fmt.Errorf("[METRICSTORE]> error creating OCF reader: %w", err)
+	}
+
+	metricsData := make(map[string]schema.FloatArray)
+
+	for ocfReader.Scan() {
+		datum, err := ocfReader.Read()
+		if err != nil {
+			return fmt.Errorf("[METRICSTORE]> error while reading avro file : %s", err)
+		}
+
+		record, ok := datum.(map[string]any)
+		if !ok {
+			return fmt.Errorf("[METRICSTORE]> failed to assert datum as map[string]interface{}")
+		}
+
+		for key, value := range record {
+			metricsData[key] = append(metricsData[key], schema.ConvertToFloat(value.(float64)))
+		}
+
+		recordCounter += 1
+	}
+
+	to := (fromTimestamp + (recordCounter / (60 / resolution) * 60))
+	if to < from {
+		return nil
+	}
+
+	for key, floatArray := range metricsData {
+		metricName := ReplaceKey(key)
+
+		if strings.Contains(metricName, SelectorDelimiter) {
+			subString := strings.Split(metricName, SelectorDelimiter)
+
+			lvl := l
+
+			for i := 0; i < len(subString)-1; i++ {
+
+				sel := subString[i]
+
+				if lvl.children == nil {
+					lvl.children = make(map[string]*Level)
+				}
+
+				child, ok := lvl.children[sel]
+				if !ok {
+					child = &Level{
+						metrics:  make([]*buffer, len(m.Metrics)),
+						children: nil,
+					}
+					lvl.children[sel] = child
+				}
+				lvl = child
+			}
+
+			leafMetricName := subString[len(subString)-1]
+			err = lvl.createBuffer(m, leafMetricName, floatArray, fromTimestamp, resolution)
+			if err != nil {
+				return fmt.Errorf("[METRICSTORE]> error while creating buffers from avroReader : %s", err)
+			}
+		} else {
+			err = l.createBuffer(m, metricName, floatArray, fromTimestamp, resolution)
+			if err != nil {
+				return fmt.Errorf("[METRICSTORE]> error while creating buffers from avroReader : %s", err)
+			}
+		}
+
+	}
+
+	return nil
+}
+
+func (l *Level) createBuffer(m *MemoryStore, metricName string, floatArray schema.FloatArray, from int64, resolution int64) error {
+	n := len(floatArray)
+	b := &buffer{
+		frequency: resolution,
+		start:     from,
+		data:      floatArray[0:n:n],
+		prev:      nil,
+		next:      nil,
+		archived:  true,
+	}
+
+	minfo, ok := m.Metrics[metricName]
+	if !ok {
+		return nil
+	}
+
+	prev := l.metrics[minfo.offset]
+	if prev == nil {
+		l.metrics[minfo.offset] = b
+	} else {
+		if prev.start > b.start {
+			return fmt.Errorf("[METRICSTORE]> buffer start time %d is before previous buffer start %d", b.start, prev.start)
+		}
+
+		b.prev = prev
+		prev.next = b
+
+		missingCount := ((int(b.start) - int(prev.start)) - len(prev.data)*int(b.frequency))
+		if missingCount > 0 {
+			missingCount /= int(b.frequency)
+
+			for range missingCount {
+				prev.data = append(prev.data, schema.NaN)
+			}
+
+			prev.data = prev.data[0:len(prev.data):len(prev.data)]
+		}
+	}
+	l.metrics[minfo.offset] = b
+
+	return nil
+}
+
+func (l *Level) loadJSONFile(m *MemoryStore, f *os.File, from int64) error {
+	br := bufio.NewReader(f)
+	cf := &CheckpointFile{}
+	if err := json.NewDecoder(br).Decode(cf); err != nil {
+		return err
+	}
+
+	if cf.To != 0 && cf.To < from {
+		return nil
+	}
+
+	if err := l.loadFile(cf, m); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func (l *Level) loadFile(cf *CheckpointFile, m *MemoryStore) error {
+	for name, metric := range cf.Metrics {
+		n := len(metric.Data)
+		b := &buffer{
+			frequency: metric.Frequency,
+			start:     metric.Start,
+			data:      metric.Data[0:n:n],
+			prev:      nil,
+			next:      nil,
+			archived:  true,
+		}
+
+		minfo, ok := m.Metrics[name]
+		if !ok {
+			continue
+		}
+
+		prev := l.metrics[minfo.offset]
+		if prev == nil {
+			l.metrics[minfo.offset] = b
+		} else {
+			if prev.start > b.start {
+				return fmt.Errorf("[METRICSTORE]> buffer start time %d is before previous buffer start %d", b.start, prev.start)
+			}
+
+			b.prev = prev
+			prev.next = b
+		}
+		l.metrics[minfo.offset] = b
+	}
+
+	if len(cf.Children) > 0 && l.children == nil {
+		l.children = make(map[string]*Level)
+	}
+
+	for sel, childCf := range cf.Children {
+		child, ok := l.children[sel]
+		if !ok {
+			child = &Level{
+				metrics:  make([]*buffer, len(m.Metrics)),
+				children: nil,
+			}
+			l.children[sel] = child
+		}
+
+		if err := child.loadFile(childCf, m); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+
+func (l *Level) fromCheckpoint(m *MemoryStore, dir string, from int64, extension string) (int, error) {
+	direntries, err := os.ReadDir(dir)
+	if err != nil {
+		if os.IsNotExist(err) {
+			return 0, nil
+		}
+
+		return 0, err
+	}
+
+	allFiles := make([]fs.DirEntry, 0)
+	filesLoaded := 0
+	for _, e := range direntries {
+		if e.IsDir() {
+			child := &Level{
+				metrics:  make([]*buffer, len(m.Metrics)),
+				children: make(map[string]*Level),
+			}
+
+			files, err := child.fromCheckpoint(m, path.Join(dir, e.Name()), from, extension)
+			filesLoaded += files
+			if err != nil {
+				return filesLoaded, err
+			}
+
+			l.children[e.Name()] = child
+		} else if strings.HasSuffix(e.Name(), "."+extension) {
+			allFiles = append(allFiles, e)
+		} else {
+			continue
+		}
+	}
+
+	files, err := findFiles(allFiles, from, extension, true)
+	if err != nil {
+		return filesLoaded, err
+	}
+
+	loaders := map[string]func(*MemoryStore, *os.File, int64) error{
+		"json": l.loadJSONFile,
+		"avro": l.loadAvroFile,
+	}
+
+	loader := loaders[extension]
+
+	for _, filename := range files {
+		// Use a closure to ensure file is closed immediately after use
+		err := func() error {
+			f, err := os.Open(path.Join(dir, filename))
+			if err != nil {
+				return err
+			}
+			defer f.Close()
+
+			return loader(m, f, from)
+		}()
+		if err != nil {
+			return filesLoaded, err
+		}
+
+		filesLoaded += 1
+	}
+
+	return filesLoaded, nil
+}
+
+// This will probably get very slow over time!
+// A solution could be some sort of an index file in which all other files
+// and the timespan they contain is listed.
+func findFiles(direntries []fs.DirEntry, t int64, extension string, findMoreRecentFiles bool) ([]string, error) {
+	nums := map[string]int64{}
+	for _, e := range direntries {
+		if !strings.HasSuffix(e.Name(), "."+extension) {
+			continue
+		}
+
+		ts, err := strconv.ParseInt(e.Name()[strings.Index(e.Name(), "_")+1:len(e.Name())-5], 10, 64)
+		if err != nil {
+			return nil, err
+		}
+		nums[e.Name()] = ts
+	}
+
+	sort.Slice(direntries, func(i, j int) bool {
+		a, b := direntries[i], direntries[j]
+		return nums[a.Name()] < nums[b.Name()]
+	})
+
+	filenames := make([]string, 0)
+	for i := range direntries {
+		e := direntries[i]
+		ts1 := nums[e.Name()]
+
+		if findMoreRecentFiles && t <= ts1 {
+			filenames = append(filenames, e.Name())
+		}
+		if i == len(direntries)-1 {
+			continue
+		}
+
+		enext := direntries[i+1]
+		ts2 := nums[enext.Name()]
+
+		if findMoreRecentFiles {
+			if ts1 < t && t < ts2 {
+				filenames = append(filenames, e.Name())
+			}
+		} else {
+			if ts2 < t {
+				filenames = append(filenames, e.Name())
+			}
+		}
+	}
+
+	return filenames, nil
+}

internal/memorystore/config.go

@@ -0,0 +1,117 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"fmt"
+	"time"
+)
+
+const (
+	DefaultMaxWorkers             = 10
+	DefaultBufferCapacity         = 512
+	DefaultGCTriggerInterval      = 100
+	DefaultAvroWorkers            = 4
+	DefaultCheckpointBufferMin    = 3
+	DefaultAvroCheckpointInterval = time.Minute
+)
+
+var InternalCCMSFlag bool = false
+
+type MetricStoreConfig struct {
+	// Number of concurrent workers for checkpoint and archive operations.
+	// If not set or 0, defaults to min(runtime.NumCPU()/2+1, 10)
+	NumWorkers  int `json:"num-workers"`
+	Checkpoints struct {
+		FileFormat string `json:"file-format"`
+		Interval   string `json:"interval"`
+		RootDir    string `json:"directory"`
+		Restore    string `json:"restore"`
+	} `json:"checkpoints"`
+	Debug struct {
+		DumpToFile string `json:"dump-to-file"`
+		EnableGops bool   `json:"gops"`
+	} `json:"debug"`
+	RetentionInMemory string `json:"retention-in-memory"`
+	Archive           struct {
+		Interval      string `json:"interval"`
+		RootDir       string `json:"directory"`
+		DeleteInstead bool   `json:"delete-instead"`
+	} `json:"archive"`
+	Subscriptions []struct {
+		// Channel name
+		SubscribeTo string `json:"subscribe-to"`
+
+		// Allow lines without a cluster tag, use this as default, optional
+		ClusterTag string `json:"cluster-tag"`
+	} `json:"subscriptions"`
+}
+
+var Keys MetricStoreConfig
+
+// AggregationStrategy for aggregation over multiple values at different cpus/sockets/..., not time!
+type AggregationStrategy int
+
+const (
+	NoAggregation AggregationStrategy = iota
+	SumAggregation
+	AvgAggregation
+)
+
+func AssignAggregationStrategy(str string) (AggregationStrategy, error) {
+	switch str {
+	case "":
+		return NoAggregation, nil
+	case "sum":
+		return SumAggregation, nil
+	case "avg":
+		return AvgAggregation, nil
+	default:
+		return NoAggregation, fmt.Errorf("[METRICSTORE]> unknown aggregation strategy: %s", str)
+	}
+}
+
+type MetricConfig struct {
+	// Interval in seconds at which measurements are stored
+	Frequency int64
+
+	// Can be 'sum', 'avg' or null. Describes how to aggregate metrics from the same timestep over the hierarchy.
+	Aggregation AggregationStrategy
+
+	// Private, used internally...
+	offset int
+}
+
+var Metrics map[string]MetricConfig
+
+func GetMetricFrequency(metricName string) (int64, error) {
+	if metric, ok := Metrics[metricName]; ok {
+		return metric.Frequency, nil
+	}
+	return 0, fmt.Errorf("[METRICSTORE]> metric %s not found", metricName)
+}
+
+// AddMetric adds logic to add metrics. Redundant metrics should be updated with max frequency.
+// use metric.Name to check if the metric already exists.
+// if not, add it to the Metrics map.
+func AddMetric(name string, metric MetricConfig) error {
+	if Metrics == nil {
+		Metrics = make(map[string]MetricConfig, 0)
+	}
+
+	if existingMetric, ok := Metrics[name]; ok {
+		if existingMetric.Frequency != metric.Frequency {
+			if existingMetric.Frequency < metric.Frequency {
+				existingMetric.Frequency = metric.Frequency
+				Metrics[name] = existingMetric
+			}
+		}
+	} else {
+		Metrics[name] = metric
+	}
+
+	return nil
+}

internal/memorystore/configSchema.go

@@ -0,0 +1,95 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+const configSchema = `{
+    "type": "object",
+    "description": "Configuration specific to built-in metric-store.",
+    "properties": {
+        "checkpoints": {
+            "description": "Configuration for checkpointing the metrics within metric-store",
+            "type": "object",
+            "properties": {
+                "file-format": {
+                    "description": "Specify the type of checkpoint file. There are 2 variants: 'avro' and 'json'. If nothing is specified, 'avro' is default.",
+                    "type": "string"
+                },
+                "interval": {
+                    "description": "Interval at which the metrics should be checkpointed.",
+                    "type": "string"
+                },
+                "directory": {
+                    "description": "Specify the parent directy in which the checkpointed files should be placed.",
+                    "type": "string"
+                },
+                "restore": {
+                    "description": "When cc-backend starts up, look for checkpointed files that are less than X hours old and load metrics from these selected checkpoint files.",
+                    "type": "string"
+                }
+            }
+        },
+        "archive": {
+            "description": "Configuration for archiving the already checkpointed files.",
+            "type": "object",
+            "properties": {
+                "interval": {
+                    "description": "Interval at which the checkpointed files should be archived.",
+                    "type": "string"
+                },
+                "directory": {
+                    "description": "Specify the parent directy in which the archived files should be placed.",
+                    "type": "string"
+                }
+            }
+        },
+        "retention-in-memory": {
+            "description": "Keep the metrics within memory for given time interval. Retention for X hours, then the metrics would be freed.",
+            "type": "string"
+        },
+        "nats": {
+            "description": "Configuration for accepting published data through NATS.",
+            "type": "array",
+            "items": {
+                "type": "object",
+                "properties": {
+                    "address": {
+                        "description": "Address of the NATS server.",
+                        "type": "string"
+                    },
+                    "username": {
+                        "description": "Optional: If configured with username/password method.",
+                        "type": "string"
+                    },
+                    "password": {
+                        "description": "Optional: If configured with username/password method.",
+                        "type": "string"
+                    },
+                    "creds-file-path": {
+                        "description": "Optional: If configured with Credential File method. Path to your NATS cred file.",
+                        "type": "string"
+                    },
+                    "subscriptions": {
+                        "description": "Array of various subscriptions. Allows to subscibe to different subjects and publishers.",
+                        "type": "array",
+                        "items": {
+                            "type": "object",
+                            "properties": {
+                                "subscribe-to": {
+                                    "description": "Channel name",
+                                    "type": "string"
+                                },
+                                "cluster-tag": {
+                                    "description": "Optional: Allow lines without a cluster tag, use this as default",
+                                    "type": "string"
+                                }
+                            }
+                        }
+                    }
+                }
+            }
+        }
+    }
+}`

internal/memorystore/debug.go

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

internal/memorystore/healthcheck.go

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

internal/memorystore/level.go

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

internal/memorystore/lineprotocol.go

@@ -0,0 +1,258 @@
+// Copyright (C) NHR@FAU, University Erlangen-Nuremberg.
+// All rights reserved. This file is part of cc-backend.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package memorystore
+
+import (
+	"context"
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/ClusterCockpit/cc-backend/pkg/nats"
+	cclog "github.com/ClusterCockpit/cc-lib/ccLogger"
+	"github.com/ClusterCockpit/cc-lib/schema"
+	"github.com/influxdata/line-protocol/v2/lineprotocol"
+)
+
+func ReceiveNats(ms *MemoryStore,
+	workers int,
+	ctx context.Context,
+) error {
+	nc := nats.GetClient()
+
+	if nc == nil {
+		cclog.Warn("NATS client not initialized")
+		return nil
+	}
+
+	var wg sync.WaitGroup
+
+	msgs := make(chan []byte, workers*2)
+
+	for _, sc := range Keys.Subscriptions {
+		clusterTag := sc.ClusterTag
+		if workers > 1 {
+			wg.Add(workers)
+
+			for range workers {
+				go func() {
+					for m := range msgs {
+						dec := lineprotocol.NewDecoderWithBytes(m)
+						if err := DecodeLine(dec, ms, clusterTag); err != nil {
+							cclog.Errorf("error: %s", err.Error())
+						}
+					}
+
+					wg.Done()
+				}()
+			}
+
+			nc.Subscribe(sc.SubscribeTo, func(subject string, data []byte) {
+				msgs <- data
+			})
+		} else {
+			nc.Subscribe(sc.SubscribeTo, func(subject string, data []byte) {
+				dec := lineprotocol.NewDecoderWithBytes(data)
+				if err := DecodeLine(dec, ms, clusterTag); err != nil {
+					cclog.Errorf("error: %s", err.Error())
+				}
+			})
+		}
+		cclog.Infof("NATS subscription to '%s' established", sc.SubscribeTo)
+	}
+
+	close(msgs)
+	wg.Wait()
+
+	return nil
+}
+
+// Place `prefix` in front of `buf` but if possible,
+// do that inplace in `buf`.
+func reorder(buf, prefix []byte) []byte {
+	n := len(prefix)
+	m := len(buf)
+	if cap(buf) < m+n {
+		return append(prefix[:n:n], buf...)
+	} else {
+		buf = buf[:n+m]
+		for i := m - 1; i >= 0; i-- {
+			buf[i+n] = buf[i]
+		}
+		for i := range n {
+			buf[i] = prefix[i]
+		}
+		return buf
+	}
+}
+
+// Decode lines using dec and make write calls to the MemoryStore.
+// If a line is missing its cluster tag, use clusterDefault as default.
+func DecodeLine(dec *lineprotocol.Decoder,
+	ms *MemoryStore,
+	clusterDefault string,
+) error {
+	// Reduce allocations in loop:
+	t := time.Now()
+	metric, metricBuf := Metric{}, make([]byte, 0, 16)
+	selector := make([]string, 0, 4)
+	typeBuf, subTypeBuf := make([]byte, 0, 16), make([]byte, 0)
+
+	// Optimize for the case where all lines in a "batch" are about the same
+	// cluster and host. By using `WriteToLevel` (level = host), we do not need
+	// to take the root- and cluster-level lock as often.
+	var lvl *Level = nil
+	prevCluster, prevHost := "", ""
+
+	var ok bool
+	for dec.Next() {
+		rawmeasurement, err := dec.Measurement()
+		if err != nil {
+			return err
+		}
+
+		// Needs to be copied because another call to dec.* would
+		// invalidate the returned slice.
+		metricBuf = append(metricBuf[:0], rawmeasurement...)
+
+		// The go compiler optimizes map[string(byteslice)] lookups:
+		metric.MetricConfig, ok = ms.Metrics[string(rawmeasurement)]
+		if !ok {
+			continue
+		}
+
+		typeBuf, subTypeBuf := typeBuf[:0], subTypeBuf[:0]
+		cluster, host := clusterDefault, ""
+		for {
+			key, val, err := dec.NextTag()
+			if err != nil {
+				return err
+			}
+			if key == nil {
+				break
+			}
+
+			// The go compiler optimizes string([]byte{...}) == "...":
+			switch string(key) {
+			case "cluster":
+				if string(val) == prevCluster {
+					cluster = prevCluster
+				} else {
+					cluster = string(val)
+					lvl = nil
+				}
+			case "hostname", "host":
+				if string(val) == prevHost {
+					host = prevHost
+				} else {
+					host = string(val)
+					lvl = nil
+				}
+			case "type":
+				if string(val) == "node" {
+					break
+				}
+
+				// We cannot be sure that the "type" tag comes before the "type-id" tag:
+				if len(typeBuf) == 0 {
+					typeBuf = append(typeBuf, val...)
+				} else {
+					typeBuf = reorder(typeBuf, val)
+				}
+			case "type-id":
+				typeBuf = append(typeBuf, val...)
+			case "subtype":
+				// We cannot be sure that the "subtype" tag comes before the "stype-id" tag:
+				if len(subTypeBuf) == 0 {
+					subTypeBuf = append(subTypeBuf, val...)
+				} else {
+					subTypeBuf = reorder(subTypeBuf, val)
+					// subTypeBuf = reorder(typeBuf, val)
+				}
+			case "stype-id":
+				subTypeBuf = append(subTypeBuf, val...)
+			default:
+			}
+		}
+
+		// If the cluster or host changed, the lvl was set to nil
+		if lvl == nil {
+			selector = selector[:2]
+			selector[0], selector[1] = cluster, host
+			lvl = ms.GetLevel(selector)
+			prevCluster, prevHost = cluster, host
+		}
+
+		// subtypes:
+		selector = selector[:0]
+		if len(typeBuf) > 0 {
+			selector = append(selector, string(typeBuf)) // <- Allocation :(
+			if len(subTypeBuf) > 0 {
+				selector = append(selector, string(subTypeBuf))
+			}
+		}
+
+		for {
+			key, val, err := dec.NextField()
+			if err != nil {
+				return err
+			}
+
+			if key == nil {
+				break
+			}
+
+			if string(key) != "value" {
+				return fmt.Errorf("host %s: unknown field: '%s' (value: %#v)", host, string(key), val)
+			}
+
+			if val.Kind() == lineprotocol.Float {
+				metric.Value = schema.Float(val.FloatV())
+			} else if val.Kind() == lineprotocol.Int {
+				metric.Value = schema.Float(val.IntV())
+			} else if val.Kind() == lineprotocol.Uint {
+				metric.Value = schema.Float(val.UintV())
+			} else {
+				return fmt.Errorf("host %s: unsupported value type in message: %s", host, val.Kind().String())
+			}
+		}
+
+		if t, err = dec.Time(lineprotocol.Second, t); err != nil {
+			t = time.Now()
+			if t, err = dec.Time(lineprotocol.Millisecond, t); err != nil {
+				t = time.Now()
+				if t, err = dec.Time(lineprotocol.Microsecond, t); err != nil {
+					t = time.Now()
+					if t, err = dec.Time(lineprotocol.Nanosecond, t); err != nil {
+						return fmt.Errorf("host %s: timestamp : %#v with error : %#v", host, t, err.Error())
+					}
+				}
+			}
+		}
+
+		if err != nil {
+			return fmt.Errorf("host %s: timestamp : %#v with error : %#v", host, t, err.Error())
+		}
+
+		time := t.Unix()
+
+		if Keys.Checkpoints.FileFormat != "json" {
+			LineProtocolMessages <- &AvroStruct{
+				MetricName: string(metricBuf),
+				Cluster:    cluster,
+				Node:       host,
+				Selector:   append([]string{}, selector...),
+				Value:      metric.Value,
+				Timestamp:  time,
+			}
+		}
+
+		if err := ms.WriteToLevel(lvl, selector, time, []Metric{metric}); err != nil {
+			return err
+		}
+	}
+	return nil
+}