Add test for sqlagent and ragagent

Signed-off-by: Dolpher Du <dolpher.du@intel.com>

helm-charts/agentqna/Chart.yaml

@@ -8,11 +8,11 @@ type: application
 dependencies:
   - name: agent
     version: 0-latest
-    alias: rag-agent
+    alias: ragagent
     repository: "file://../common/agent"
   - name: agent
     version: 0-latest
-    alias: sql-agent
+    alias: sqlagent
     repository: "file://../common/agent"
   - name: agent
     version: 0-latest

helm-charts/agentqna/gaudi-values.yaml

@@ -6,23 +6,6 @@
 
 tgi:
   enabled: true
-  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"]
   livenessProbe:
     initialDelaySeconds: 5
     periodSeconds: 5
@@ -36,3 +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 $max_retry ]; then echo "test failed with maximum retry"; exit 1; fi
+          # Test supervisor
           max_retry=20;
           for ((i=1; i<=max_retry; i++)); do
             curl http://{{ include "agentqna.fullname" (index .Subcharts "supervisor") }}:{{ .Values.supervisor.service.port }}/v1/chat/completions -sS --fail-with-body \
             -X POST \
-            -d '{"messages": "Tell me three songs of Michael Jackson."}' \
+            -d '{"messages": "How many albums does Iron Maiden have?"}' \
             -H 'Content-Type: application/json' && break;
             curlcode=$?
             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

helm-charts/agentqna/values.yaml

@@ -57,7 +57,7 @@ docretriever:
     # Overrides the image tag whose default is the chart appVersion.
     tag: "latest"
 
-sql-agent:
+sqlagent:
   DBPath: "/mnt/tools"
   db_name: "Chinook"
   db_path: "sqlite:////home/user/chinook-db/Chinook_Sqlite.sqlite"
@@ -74,7 +74,7 @@ sql-agent:
   stream: "false"
   require_human_feedback: "false"
 
-rag-agent:
+ragagent:
   toolPath: "/mnt/tools"
   service:
     port: 9095
@@ -119,24 +119,39 @@ crag:
 # Override values in specific subcharts
 tgi:
   enabled: false
+  accelDevice: "gaudi"
+  image:
+    repository: ghcr.io/huggingface/tgi-gaudi
+    tag: "2.0.6"
+  resources:
+    limits:
+      habana.ai/gaudi: 4
   LLM_MODEL_ID: "meta-llama/Meta-Llama-3.1-70B-Instruct"
-  MAX_INPUT_LENGTH: "4096"
-  MAX_TOTAL_TOKENS: "8192"
+  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"]
+
 vllm:
   enabled: false
   accelDevice: "gaudi"
   image:
     repository: opea/vllm-gaudi
     tag: "latest"
+  resources:
+    limits:
+      habana.ai/gaudi: 4
   LLM_MODEL_ID: "meta-llama/Meta-Llama-3.1-70B-Instruct"
   OMPI_MCA_btl_vader_single_copy_mechanism: none
   PT_HPU_ENABLE_LAZY_COLLECTIVES: true
   VLLM_SKIP_WARMUP: true
   extraCmdArgs: ["--tensor-parallel-size", "4", "--max-seq_len-to-capture", "16384"]
-  resources:
-    limits:
-      habana.ai/gaudi: 4
-
 
 global:
   http_proxy: ""

helm-charts/agentqna/vllm-gaudi-values.yaml

@@ -8,10 +8,9 @@ vllm:
   enabled: true
   image:
     repository: opea/vllm-gaudi
-    # tag: "good"
 supervisor:
-  llm_endpoint_url: http://agentqna-vllm
-rag-agent:
-  llm_endpoint_url: http://agentqna-vllm
-sql-agent:
-  llm_endpoint_url: http://agentqna-vllm
+  llm_endpoint_url: http://{{ .Release.Name }}-vllm
+ragagent:
+  llm_endpoint_url: http://{{ .Release.Name }}-vllm
+sqlagent:
+  llm_endpoint_url: http://{{ .Release.Name }}-vllm

helm-charts/common/agent/templates/configmap.yaml

@@ -18,7 +18,7 @@ data:
   tools: {{ .Values.tools | quote }}
   {{- end }}
   {{- if .Values.llm_endpoint_url }}
-  llm_endpoint_url: {{ .Values.llm_endpoint_url | quote }}
+  llm_endpoint_url: {{ tpl .Values.llm_endpoint_url . | quote }}
   {{- else }}
   llm_endpoint_url: "http://{{ .Release.Name }}-tgi"
   {{- end }}

helm-charts/common/agent/templates/tests/test-pod.yaml

@@ -17,6 +17,10 @@ spec:
       command: ['bash', '-c']
       args:
         - |
+          if [ {{ include "agent.fullname" . }} != "agent" ]; then
+            echo "Skip test."
+            exit 0
+          fi
           max_retry=20;
           for ((i=1; i<=max_retry; i++)); do
           curl http://{{ include "agent.fullname" . }}:{{ .Values.service.port }}/v1/chat/completions -sS --fail-with-body \