Merge branch 'opea-project:main' into main

.github/workflows/_helm-e2e.yaml

@@ -119,10 +119,20 @@ jobs:
             exit 0
           fi
           helm-charts/update_dependency.sh && helm dependency update ${{ env.CHART_FOLDER }}
+          for img in `helm template -n $NAMESPACE -f ${{ env.CHART_FOLDER }}/${value_file} $RELEASE_NAME ${{ env.CHART_FOLDER }} | grep 'image:' | grep 'opea/' | awk '{print $2}' | xargs`;
+          do
+            if ! .github/workflows/scripts/e2e/chart_test.sh check_local_opea_image "$img"; then
+            echo "skip_validate=true" >> $GITHUB_ENV
+            echo "should_cleanup=false" >> $GITHUB_ENV
+            exit 1
+            fi
+          done
           if ! helm install --create-namespace --namespace $NAMESPACE --wait \
               --timeout "$ROLLOUT_TIMEOUT_SECONDS" \
               --set GOOGLE_API_KEY=${{ env.GOOGLE_API_KEY}} \
               --set GOOGLE_CSE_ID=${{ env.GOOGLE_CSE_ID}} \
+              --set web-retriever.GOOGLE_API_KEY=${{ env.GOOGLE_API_KEY}} \
+              --set web-retriever.GOOGLE_CSE_ID=${{ env.GOOGLE_CSE_ID}} \
               --values ${{ env.CHART_FOLDER }}/${value_file} \
               $RELEASE_NAME ${{ env.CHART_FOLDER }} ; then
             echo "Failed to install chart ${{ inputs.workload }}"

.github/workflows/check-online-doc-build.yml

@@ -13,7 +13,7 @@ on:
 
 jobs:
   build:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
 
     - name: Checkout

.github/workflows/pr-chart-e2e.yaml

@@ -1,4 +1,4 @@
-# Copyright (C) 2024 Intel Corporation
+# Copyright (C) 2025 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
 
 name: E2E test with helm charts
@@ -50,8 +50,14 @@ jobs:
           cut -d'/' -f3 | sort -u )
           run_matrix="{\"include\":["
           for chart in ${e2e_charts}; do
-            for file in "$CHARTS_DIR/$chart"/ci-*values.yaml; do
+            for file in "$CHARTS_DIR/$chart"/*values.yaml; do
               if [ -f "$file" ]; then
+                if [[ "$file" == *"nv-values.yaml" ]] || [[ "$file" == *"hpa-values.yaml" ]] || [[ "$file" == *"/values.yaml" ]]; then
+                  continue
+                fi
+                if [[ "$file" == *"variant_"* ]] || [[ "$file" == *"milvus"* ]]; then
+                  continue
+                fi
                 filename=$(basename "$file" .yaml)
                 if [[ "$filename" == *"gaudi"* ]]; then
                   run_matrix="${run_matrix}{\"example\":\"${chart}\",\"hardware\":\"gaudi\", \"valuefile\":\"${filename}\"},"
@@ -62,8 +68,14 @@ jobs:
             done
           done
           for chart in ${common_charts}; do
-            for file in "$CHARTS_DIR/common/$chart"/ci-*values.yaml; do
+            for file in "$CHARTS_DIR/common/$chart"/*values.yaml; do
               if [ -f "$file" ]; then
+                if [[ "$file" == *"nv-values.yaml" ]] || [[ "$file" == *"hpa-values.yaml" ]] || [[ "$file" == *"/values.yaml" ]]; then
+                  continue
+                fi
+                if [[ "$file" == *"variant_"* ]] || [[ "$file" == *"milvus"* ]]; then
+                  continue
+                fi
                 filename=$(basename "$file" .yaml)
                 if [[ "$filename" == *"gaudi"* ]]; then
                   run_matrix="${run_matrix}{\"example\":\"${chart}\",\"hardware\":\"gaudi\", \"valuefile\":\"${filename}\",\"directory\":\"common\"},"

.github/workflows/pr-path-detection.yml

@@ -124,7 +124,7 @@ jobs:
                   fi
                 fi
               else
-                echo "$check_path does not exist"
+                echo "Invalid reference path from $refer_path, reference path: $(echo $png_line | cut -d ']' -f2)"
                 fail="TRUE"
               fi
             done

.github/workflows/releasedevhelm.yaml → .github/workflows/push-release-charts.yaml

@@ -4,18 +4,16 @@
 name: Update the helm charts 0-latest
 
 on:
-  pull_request:
+  push:
     branches:
       - main
-    types: [closed]
     paths:
       - helm-charts/**
 env:
   CHARTS_DIR: "helm-charts"
 
 jobs:
-  run-on-merge:
-    if: github.event.pull_request.merged == true
+  release-on-push:
     runs-on: ubuntu-latest
 
     steps:
@@ -32,7 +30,7 @@ jobs:
             echo "Version check failed, only update 0-latest"
             exit 1
           fi
-          base_commit=${{ github.event.pull_request.base.sha }}
+          base_commit=$(git rev-parse HEAD~1) # push event
           merged_commit=$(git log -1 --format='%H')
           # Update components
           common_charts=$(git diff --name-only ${base_commit} ${merged_commit} | \
@@ -51,8 +49,8 @@ jobs:
           popd
           # Update Examples
           e2e_charts=$(git diff --name-only ${base_commit} ${merged_commit} | \
-          grep -v "common" | \
-          grep -vE 'README.md|*.sh' | \
+          grep "^$CHARTS_DIR" | \
+          grep -vE 'README.md|common|*.sh' | \
           cut -d'/' -f2 | sort -u )
           echo "Charts to be updated: $e2e_charts"
           echo "${{ secrets.ACTION_TOKEN }}" | helm registry login ghcr.io -u opea --password-stdin

.github/workflows/scripts/e2e/chart_test.sh

@@ -73,6 +73,27 @@ function dump_all_pod_logs() {
     done
 }
 
+function check_local_opea_image() {
+    origimg=$1
+    img=$(echo $origimg | xargs)
+    host=$(echo $img | awk 'BEGIN {FS="opea/"} {print $1}')
+    if [ -z "$host" ]; then
+        echo "Skip none CI local image $origimg"
+        exit 0
+    fi
+    image=$(echo $img | awk 'BEGIN {FS="opea/"} {print $2}')
+    image_name=$(echo $image | awk 'BEGIN {FS=":"} {print $1}')
+    image_tag=$(echo $image | awk 'BEGIN {FS=":"} {print $2}')
+    image_tag=${image_tag:-latest}
+    resp=$(curl -s -f -X GET "${host}v2/opea/${image_name}/manifests/${image_tag}")
+    if [ -z "$resp" ]; then
+        echo "Failed to get image manifest $origimg"
+        exit 1
+    fi
+    DATE=$(echo "$resp" | jq -r '.history[0].v1Compatibility' | jq -r '.created')
+    echo "Found image $origimg created at $DATE"
+}
+
 if [ $# -eq 0 ]; then
     echo "Usage: $0 <function_name>"
     exit 1
@@ -88,6 +109,9 @@ case "$1" in
     dump_all_pod_logs)
         dump_all_pod_logs $2
         ;;
+    check_local_opea_image)
+	check_local_opea_image $2
+	;;
     *)
         echo "Unknown function: $1"
         ;;

.pre-commit-config.yaml

@@ -4,7 +4,7 @@ ci:
 
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.5.0
+    rev: v5.0.0
     hooks:
       - id: end-of-file-fixer
         files: (.*\.(py|md|rst|yaml|yml|json|ts|js|html|svelte|sh))$
@@ -82,29 +82,29 @@ repos:
           - [email protected]
 
   - repo: https://github.com/psf/black.git
-    rev: 24.3.0
+    rev: 24.10.0
     hooks:
       - id: black
         files: (.*\.py)$
 
   - repo: https://github.com/asottile/blacken-docs
-    rev: 1.16.0
+    rev: 1.19.1
     hooks:
       - id: blacken-docs
         args: [--line-length=120, --skip-errors]
         additional_dependencies:
           - black==24.3.0
 
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.2.6
+    rev: v2.3.0
     hooks:
       - id: codespell
         args: [-w]
         additional_dependencies:
           - tomli
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.3.5
+    rev: v0.8.6
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix, --no-cache]

README.md

@@ -41,15 +41,15 @@ The following steps are optional. They're only required if you want to run the w
 
 4. [Install Intel Gaudi device plugin for Kubernetes](https://docs.habana.ai/en/latest/Installation_Guide/Additional_Installation/Kubernetes_Installation/index.html).
 
-   Alternatively, Intel provides a base operator to manage the Gaudi software stack. Refer to [this file](kubernetes-addons/Intel-Gaudi-Base-Operator/README.md) for details.
+Alternatively, Intel provides a [base operator](https://docs.habana.ai/en/latest/Installation_Guide/Additional_Installation/Kubernetes_Installation/Kubernetes_Operator.html) to manage the Gaudi software stack.
 
 ## Usages
 
 ### Use GenAI Microservices Connector (GMC) to deploy and adjust GenAIExamples
 
 Follow [GMC README](microservices-connector/README.md)
 to install GMC into your kubernetes cluster. [GenAIExamples](https://github.com/opea-project/GenAIExamples) contains several sample GenAI example use case pipelines such as ChatQnA, DocSum, etc.
-Once you have deployed GMC in your Kubernetes cluster, you can deploy any of the example pipelines by following its Readme file (e.g. [Docsum](https://github.com/opea-project/GenAIExamples/blob/main/DocSum/kubernetes/intel/README_gmc.md)).
+Once you have deployed GMC in your Kubernetes cluster, you can deploy any of the example pipelines by following its Readme file (e.g. [Docsum](https://github.com/opea-project/GenAIExamples/blob/main/DocSum/kubernetes/gmc/README.md)).
 
 ### Use helm charts to deploy
 

helm-charts/agentqna/Chart.yaml

@@ -8,7 +8,11 @@ type: application
 dependencies:
   - name: agent
     version: 0-latest
-    alias: worker
+    alias: ragagent
+    repository: "file://../common/agent"
+  - name: agent
+    version: 0-latest
+    alias: sqlagent
     repository: "file://../common/agent"
   - name: agent
     version: 0-latest
@@ -18,6 +22,10 @@ dependencies:
     version: 0-latest
     repository: "file://../common/tgi"
     condition: tgi.enabled
+  - name: vllm
+    version: 0-latest
+    repository: "file://../common/vllm"
+    condition: vllm.enabled
   - name: tei
     version: 0-latest
     repository: "file://../common/tei"

helm-charts/agentqna/gaudi-values.yaml

@@ -5,23 +5,7 @@
 # by overriding their subchart values
 
 tgi:
-  accelDevice: "gaudi"
-  image:
-    repository: ghcr.io/huggingface/tgi-gaudi
-    tag: "2.0.6"
-  resources:
-    limits:
-      habana.ai/gaudi: 4
-  MAX_INPUT_LENGTH: "4096"
-  MAX_TOTAL_TOKENS: "8192"
-  CUDA_GRAPHS: ""
-  OMPI_MCA_btl_vader_single_copy_mechanism: "none"
-  PT_HPU_ENABLE_LAZY_COLLECTIVES: "true"
-  ENABLE_HPU_GRAPH: "true"
-  LIMIT_HPU_GRAPH: "true"
-  USE_FLASH_ATTENTION: "true"
-  FLASH_ATTENTION_RECOMPUTE: "true"
-  extraCmdArgs: ["--sharded","true","--num-shard","4"]
+  enabled: true
   livenessProbe:
     initialDelaySeconds: 5
     periodSeconds: 5
@@ -35,3 +19,12 @@ tgi:
     periodSeconds: 5
     timeoutSeconds: 1
     failureThreshold: 120
+supervisor:
+  llm_endpoint_url: http://{{ .Release.Name }}-tgi
+  llm_engine: tgi
+ragagent:
+  llm_endpoint_url: http://{{ .Release.Name }}-tgi
+  llm_engine: tgi
+sqlagent:
+  llm_endpoint_url: http://{{ .Release.Name }}-tgi
+  llm_engine: tgi

helm-charts/agentqna/templates/tests/test-pod.yaml

@@ -17,14 +17,54 @@ spec:
       command: ['bash', '-c']
       args:
         - |
+          # Ingest data
+          cd /mnt/tools
+          pip install requests tqdm
+          ./ingest_data.sh
+          # Test ragagent
+          max_retry=10;
+          for ((i=1; i<=max_retry; i++)); do
+            curl http://{{ include "agentqna.fullname" (index .Subcharts "ragagent") }}:{{ .Values.ragagent.service.port }}/v1/chat/completions -sS --fail-with-body \
+            -X POST \
+            -d '{"messages": "Tell me about Michael Jackson song Thriller"}' \
+            -H 'Content-Type: application/json' && break;
+            curlcode=$?
+            if [[ $curlcode -eq 7 ]]; then sleep 10; else echo "curl failed with code $curlcode"; exit 1; fi;
+          done;
+          if [ $i -gt $max_retry ]; then echo "test failed with maximum retry"; exit 1; fi
+          # Test sqlagent
+          max_retry=10;
+          for ((i=1; i<=max_retry; i++)); do
+            curl http://{{ include "agentqna.fullname" (index .Subcharts "sqlagent") }}:{{ .Values.sqlagent.service.port }}/v1/chat/completions -sS --fail-with-body \
+            -X POST \
+            -d '{"messages": "How many employees are there in the company?"}' \
+            -H 'Content-Type: application/json' && break;
+            curlcode=$?
+            if [[ $curlcode -eq 7 ]]; then sleep 10; else echo "curl failed with code $curlcode"; exit 1; fi;
+          done;
+          if [ $i -gt $max_retry ]; then echo "test failed with maximum retry"; exit 1; fi
+          # Test supervisor
           max_retry=20;
           for ((i=1; i<=max_retry; i++)); do
             curl http://{{ include "agentqna.fullname" (index .Subcharts "supervisor") }}:{{ .Values.supervisor.service.port }}/v1/chat/completions -sS --fail-with-body \
             -X POST \
-            -d '{"query": "Tell me three songs of Michael Jackson."}' \
+            -d '{"messages": "How many albums does Iron Maiden have?"}' \
             -H 'Content-Type: application/json' && break;
             curlcode=$?
             if [[ $curlcode -eq 7 ]]; then sleep 10; else echo "curl failed with code $curlcode"; exit 1; fi;
           done;
           if [ $i -gt $max_retry ]; then echo "test failed with maximum retry"; exit 1; fi
+      volumeMounts:
+        - mountPath: /mnt/tools
+          name: test
+  volumes:
+    - name: test
+      hostPath:
+        path: /mnt/tools
+        type: Directory
+  {{- with .Values.nodeSelector }}
+  nodeSelector:
+    {{- toYaml . | nindent 8 }}
+  {{- end }}
+
   restartPolicy: Never