Bump brace-expansion from 1.1.12 to 1.1.13 in /src/rest-server

.azure-pipelines/cluster-update-all.yml

@@ -31,6 +31,14 @@ jobs:
     pool:
       vmImage: 'ubuntu-24.04'
     steps:
+    - checkout: self
+      submodules: false
+      persistCredentials : true
+    - powershell: |
+        $header = "AUTHORIZATION: bearer $(System.AccessToken)"
+        git -c http.extraheader="$header" submodule sync
+        git -c http.extraheader="$header" submodule update --init --force --depth=1
+      displayName: Checkout Submodule
     - script: python -m pip install pyyaml jinja2 paramiko etcd3 protobuf==3.20.3 kubernetes gitpython   
       displayName: Install python libs
     - script: |
@@ -48,17 +56,17 @@ jobs:
         destinationFolder: '$(Pipeline.Workspace)/config'
         cleanDestinationFolder: true
     - script: |
-        mv $(Pipeline.Workspace)/config/pylon-configuration /tmp/
         mv $(Pipeline.Workspace)/config/auth-configuration /tmp/
+        ls -l /tmp/auth-configuration
       displayName: Arrange Config Files
     - script: |
-        # Build all services # Skip "frameworkcontroller" "hivedscheduler" "openpai-runtime" "k8s-dashboard" "marketplace-db" "node-exporter" "prometheus"
-        $(Build.Repository.LocalPath)/build/pai_build.py build -c $(Pipeline.Workspace)/config/cluster-configuration -s base-image cleaning-image cluster-configuration device-plugin  job-exporter log-manager grafana alert-manager watchdog internal-storage postgresql database-controller fluentd pylon rest-server webportal
+        # Build all services # Skip"k8s-dashboard" "marketplace-db" "node-exporter" "prometheus"
+        $(Build.Repository.LocalPath)/build/pai_build.py build -c $(Pipeline.Workspace)/config/cluster-configuration -s base-image cleaning-image cluster-configuration device-plugin job-exporter log-manager grafana alert-manager frameworkcontroller hivedscheduler openpai-runtime watchdog internal-storage postgresql database-controller fluentd pylon rest-server webportal 
       displayName: 'Build all Services images'
       condition: or(eq(${{ parameters.rebuildImage }}, 'true'), eq(${{ parameters.repushImage }}, 'true'))
     - script: |
-        # Build all services # Skip "frameworkcontroller" "hivedscheduler" "openpai-runtime" "k8s-dashboard" "marketplace-db" "node-exporter" "prometheus"
-        $(Build.Repository.LocalPath)/build/pai_build.py push -c $(Pipeline.Workspace)/config/cluster-configuration -s base-image cleaning-image cluster-configuration device-plugin job-exporter log-manager grafana alert-manager watchdog internal-storage postgresql database-controller fluentd pylon rest-server webportal
+        # Push all services # Skip "k8s-dashboard" "marketplace-db" "node-exporter" "prometheus"
+        $(Build.Repository.LocalPath)/build/pai_build.py push -c $(Pipeline.Workspace)/config/cluster-configuration -s base-image cleaning-image cluster-configuration device-plugin job-exporter log-manager grafana alert-manager frameworkcontroller hivedscheduler openpai-runtime watchdog internal-storage postgresql database-controller fluentd pylon rest-server webportal
       displayName: 'Push all Services'
       condition: eq(${{ parameters.repushImage }}, 'true')
     - task: AzureCLI@2
@@ -97,7 +105,7 @@ jobs:
                   echo "Testing rest-server $(paiWebUrl)/rest-server/api/v2/info"
                   curl $(paiWebUrl)/rest-server/api/v2/info
                   echo "Checking virtual cluster status..."
-                  vc_info=$(curl -H "Authorization: Bearer $(paiWebToken)" -s $(paiWebUrl)/rest-server/api/v2/virtualclusters)
+                  vc_info=$(curl -H "Authorization: Bearer $(paiWebToken)" -s $(paiWebUrl)/rest-server/api/v2/virtual-clusters)
                   if [ $? -ne 0 ]; then
                     echo "Failed to access virtual cluster API"
                     exit 1

.azure-pipelines/cluster-update-changes.yml

@@ -12,12 +12,35 @@ variables:
   value: '$(Build.BuildId)'
 - group: 'pai-cicd-cluster'
 
+parameters:
+  - name: "rebuildImage"
+    type: boolean
+    default: true
+    displayName: "Rebuild images"
+  - name: "repushImage"
+    type: boolean
+    default: true
+    displayName: "Repush images"
+  - name: "redeployService"
+    type: boolean
+    default: true
+    displayName: "Redeploy services"  
+
 jobs:
   - job: Build
     displayName: 'Build and Deploy'
+    timeoutInMinutes: 120
     pool:
       vmImage: 'ubuntu-24.04'
     steps:
+    - checkout: self
+      submodules: false
+      persistCredentials : true
+    - powershell: |
+        $header = "AUTHORIZATION: bearer $(System.AccessToken)"
+        git -c http.extraheader="$header" submodule sync
+        git -c http.extraheader="$header" submodule update --init --force --depth=1
+      displayName: Checkout Submodule
     - script: |
         if [ "$(Build.Reason)" == "PullRequest" ]; then
           # Fetch the target branch
@@ -32,14 +55,7 @@ jobs:
         fi
 
         folders=$(echo "$changed_files" | grep '^src/' | awk -F'/' '{print $2}' | sort -u)
-
-        # Check if "hivedscheduler" is in the folder list
-        if echo "$folders" | grep -q "hivedscheduler"; then
-          folders=$(echo "$folders" | grep -v "cluster-configuration" | grep -v "rest-server")
-          # Add "cluster-configuration" to the head and "rest-server" to the end
-          folders="cluster-configuration $folders rest-server"
-        fi
-
+        folders=$(echo "$folders" | tr '\n' ' ')
         # Store the folder list in a pipeline variable
         echo "Changed folders: $folders"
         echo "##vso[task.setvariable variable=changed_folders]$folders"
@@ -72,27 +88,29 @@ jobs:
         destinationFolder: '$(Pipeline.Workspace)/config'
         cleanDestinationFolder: true
     - script: |
-        mv $(Pipeline.Workspace)/config/pylon-configuration /tmp/
         mv $(Pipeline.Workspace)/config/auth-configuration /tmp/
+        ls -l /tmp/auth-configuration
       displayName: Arrange Config Files
       condition: eq(variables['has_changed'], 'true')
     - script: |
         # Build the changed services
-        # Skip "openpai-runtime" due to the image built by other repo
-        changed_services=$(echo $(changed_folders) | tr ' ' '\n' | grep -v "openpai-runtime" | tr '\n' ' ')
+        echo "Building folders" $(changed_folders)
+        changed_services=$(echo $(changed_folders) | tr ' ' '\n' )
+        echo "Building: " $changed_services
         $(Build.Repository.LocalPath)/build/pai_build.py build -c $(Pipeline.Workspace)/config/cluster-configuration -s $changed_services
       displayName: 'Build Changed Services'
-      condition: eq(variables['has_changed'], 'true')
+      condition: and( eq( variables['has_changed'], 'true'), or(eq(${{ parameters.rebuildImage }}, 'true'), eq(${{ parameters.repushImage }}, 'true')) )
     - script: |
         # Push the changed services
-        # Skip "openpai-runtime" due to the image built by other repo
-        changed_services=$(echo $(changed_folders) | tr ' ' '\n' | grep -v "openpai-runtime" | tr '\n' ' ')
+
+        changed_services=$(echo $(changed_folders) | tr ' ' '\n' )
+        echo "Pushing: " $changed_services
         $(Build.Repository.LocalPath)/build/pai_build.py push -c $(Pipeline.Workspace)/config/cluster-configuration -s $changed_services
       displayName: 'Push Changed Services'
-      condition: eq(variables['has_changed'], 'true')
+      condition: and(eq(${{ parameters.repushImage }}, 'true'),  eq(variables['has_changed'], 'true') )
     - task: AzureCLI@2
       displayName: 'Azure CLI get credentials of aks and deploy the pai services'
-      condition: eq(variables['has_changed'], 'true')
+      condition: and(eq(${{ parameters.redeployService }}, 'true'), eq(variables['has_changed'], 'true'))
       inputs:
         azureSubscription: $(azureSubscriptionEndpoint)
         scriptType: bash
@@ -128,7 +146,7 @@ jobs:
                   echo "Testing rest-server $(paiWebUrl)/rest-server/api/v2/info"
                   curl $(paiWebUrl)/rest-server/api/v2/info
                   echo "Checking virtual cluster status..."
-                  vc_info=$(curl -H "Authorization: Bearer $(paiWebToken)" -s $(paiWebUrl)/rest-server/api/v2/virtualclusters)
+                  vc_info=$(curl -H "Authorization: Bearer $(paiWebToken)" -s $(paiWebUrl)/rest-server/api/v2/virtual-clusters)
                   if [ $? -ne 0 ]; then
                     echo "Failed to access virtual cluster API"
                     exit 1

.github/CODEOWNERS

@@ -0,0 +1,2 @@
+. @microsoft/ltpadmin
+.github @microsoft/ltpadmin

.github/workflows/build-all.yaml

@@ -0,0 +1,121 @@
+name: Build All Services
+
+permissions:
+  contents: read
+
+on:
+  pull_request:
+    types: [opened, reopened, closed]
+    branches: ["release/*"]
+  release:
+    types: [published]
+  workflow_dispatch:
+    inputs:
+      branch:
+        description: 'The branch name or tag to run the workflow on'
+        required: true
+        default: 'dev'
+        type: string
+
+env:
+  TAG: ${{ github.run_number }}
+
+jobs:
+  build:
+    name: Build All
+    runs-on: [self-hosted, paicicd]
+    timeout-minutes: 120
+    environment: auto-test
+    if: github.event_name != 'pull_request' || ( github.event.action == 'opened' || github.event.action == 'reopened' || github.event.pull_request.merged == true)
+    container:
+      image: ubuntu:latest
+      volumes:
+        - /var/run/docker.sock:/var/run/docker.sock
+    steps:
+      - name: Install git
+        run: |
+          DEBIAN_FRONTEND=noninteractive apt update
+          DEBIAN_FRONTEND=noninteractive apt install -y git
+
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          submodules: false
+          ref: ${{ github.event.inputs.branch || github.ref }}
+
+      - name: Get All Services
+        id: all
+        run: |
+          services=$(ls -1d src/* | awk -F'/' '{print $2}' | tr '\n' ' ')
+          skipped_services="base-image cleaning-image dev-box marketplace-db marketplace-restserver marketplace-webportal utilization-reporter"
+          for skip in $skipped_services; do
+            services=$(echo $services | sed "s/\b$skip\b//g")
+          done
+          echo "All services: $services"
+          echo "services=$services" >> $GITHUB_OUTPUT
+
+      - name: Install Package
+        if: steps.all.outputs.services != ''
+        run: |
+          DEBIAN_FRONTEND=noninteractive apt install -y python3 python-is-python3 pip git unzip ca-certificates curl apt-transport-https lsb-release gnupg parallel
+          curl -sL https://aka.ms/InstallAzureCLIDeb | bash
+          curl -fsSL https://get.docker.com | sh
+
+      - name: Install python libs
+        if: steps.all.outputs.services != ''
+        run: python -m pip install --break-system-packages pyyaml jinja2 paramiko etcd3 protobuf==3.20.3 kubernetes gitpython
+
+      - name: Decode and unzip config file
+        if: steps.all.outputs.services != ''
+        run: |
+          echo "${{ secrets.CONFIG_FILE_B64 }}" | base64 -d > config.zip
+          mkdir -p $GITHUB_WORKSPACE/config
+          unzip -o config.zip -d $GITHUB_WORKSPACE/config
+          ls -l $GITHUB_WORKSPACE/config
+
+      - name: Arrange Config Files
+        if: steps.all.outputs.services != ''
+        run: |
+          rm -rf /tmp/auth-configuration
+          mv $GITHUB_WORKSPACE/config/auth-configuration /tmp/
+          ls -l /tmp/auth-configuration
+
+      - name: Log in to GHCR
+        run: echo "${{ secrets.GITHUB_TOKEN }}" | docker login ghcr.io -u ${{ github.actor }} --password-stdin
+
+      - name: Build Images of Services
+        if: steps.all.outputs.services != ''
+        run: |
+          all_services="${{ steps.all.outputs.services }}"
+          echo "Building: $all_services"
+          echo "--------------------------------"
+          failed_services=""
+          for service in $all_services; do
+            if echo "$service" | grep -q "alert-manager"; then
+              echo "alert-manager is in the changed services"
+              # Build specific images in alert-manager
+              echo "Building specific alert-manager images"
+              $GITHUB_WORKSPACE/build/pai_build.py build \
+                -c $GITHUB_WORKSPACE/config/cluster-configuration \
+                -s alert-manager \
+                -i abnormal-detector,alert-handler,alert-parser,cert-expiration-checker,cluster-utilization,job-data-recorder,job-status-change-notification,node-failure-detection,node-issue-classifier,nvidia-gpu-low-perf-fixer,redis-monitoring
+            fi
+            echo "Building service: $service"
+            if python3 $GITHUB_WORKSPACE/build/pai_build.py build \
+              -c $GITHUB_WORKSPACE/config/cluster-configuration \
+              -s $service; then
+              echo "✓ Successfully built: $service"
+            else
+              echo "✗ Failed to build: $service"
+              failed_services="$failed_services $service"
+            fi
+          done
+
+          if [ -n "$failed_services" ]; then
+            echo "::error::Failed to build services:$failed_services"
+            echo "FAILED_SERVICES=$failed_services" 
+            exit 1
+          else
+            echo "All services built successfully"
+          fi