test

Author: liustve

.github/workflows/python-ec2-genesis-test.yml renamed to .github/workflows/python-ec2-genai-test.yml

@@ -21,9 +21,10 @@ env:
   SPANS_LOG_GROUP: aws/spans
   LOG_GROUP_NAME: test/genesis
   TEST_RESOURCES_FOLDER: ${GITHUB_WORKSPACE}
+  SAMPLE_APP_ZIP: s3://sigv4perfresults/langchain-service.zip
 
 jobs:
-  python-ec2-adot-langchain:
+  python-ec2-adot-genai:
     runs-on: ubuntu-latest
     timeout-minutes: 30
     steps:
@@ -68,20 +69,20 @@ jobs:
       - name: Initiate Terraform
         uses: ./.github/workflows/actions/execute_and_retry
         with:
-          command: "cd ${{ env.TEST_RESOURCES_FOLDER }}/terraform/python/ec2/langchain && terraform init && terraform validate"
+          command: "cd ${{ env.TEST_RESOURCES_FOLDER }}/terraform/python/ec2/adot-genai && terraform init && terraform validate"
           cleanup: "rm -rf .terraform && rm -rf .terraform.lock.hcl"
           max_retry: 6
 
       - name: Deploy LangChain service via terraform
-        working-directory: terraform/python/ec2/langchain
+        working-directory: terraform/python/ec2/adot-genai
         run: |
           terraform apply -auto-approve \
             -var="aws_region=${{ env.E2E_TEST_AWS_REGION }}" \
             -var="test_id=${{ env.TESTING_ID }}" \
-            -var="service_zip_url=s3://sigv4perfresults/langchain-service.zip"
+            -var="service_zip_url=${{ env.SAMPLE_APP_ZIP }}"
 
       - name: Get deployment info
-        working-directory: terraform/python/ec2/langchain
+        working-directory: terraform/python/ec2/adot-genai
         run: |
           echo "INSTANCE_IP=$(terraform output langchain_service_public_ip)" >> $GITHUB_ENV
           echo "INSTANCE_ID=$(terraform output langchain_service_instance_id)" >> $GITHUB_ENV
@@ -151,9 +152,9 @@ jobs:
       - name: Cleanup
         if: always()
         continue-on-error: true
-        working-directory: terraform/python/ec2/langchain
+        working-directory: terraform/python/ec2/adot-genai
         run: |
           terraform destroy -auto-approve \
             -var="aws_region=${{ env.E2E_TEST_AWS_REGION }}" \
             -var="test_id=${{ env.TESTING_ID }}" \
-            -var="service_zip_url=s3://sigv4perfresults/langchain-service.zip"
+            -var="service_zip_url=${{ env.SAMPLE_APP_ZIP }}"

sample-apps/python/langchain_service/ec2-requirements.txt renamed to sample-apps/python/genai_service/ec2-requirements.txt

@@ -82,18 +82,12 @@ async def root():
     return {
         "message": "LangChain Bedrock OpenInference API is running!",
         "endpoints": {
-            "/chat": "Single message chat endpoint",
-            "/batch": "Batch message processing endpoint",
-            "/sample": "Run sample prompts",
+            "/ai-chat": "Single message chat endpoint",
             "/hello": "Simple hello endpoint"
         }
     }
 
-@app.get("/hello")
-async def hello():
-    return {"message": "Hello from LangChain service!"}
-
-@app.post("/chat", response_model=ChatResponse)
+@app.post("/ai-chat", response_model=ChatResponse)
 async def chat(request: ChatRequest):
     """
     Chat endpoint that processes a single user message through AWS Bedrock
@@ -105,40 +99,6 @@ async def chat(request: ChatRequest):
     except Exception as e:
         raise HTTPException(status_code=500, detail=str(e))
 
-@app.post("/batch", response_model=BatchChatResponse)
-async def batch_chat(request: BatchChatRequest):
-    """
-    Batch endpoint that processes multiple messages
-    """
-    try:
-        responses = []
-        for message in request.messages:
-            result = await chain.ainvoke({"input": message})
-            responses.append({
-                "message": message,
-                "response": result["text"]
-            })
-        return BatchChatResponse(responses=responses)
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-@app.get("/sample", response_model=BatchChatResponse)
-async def run_samples():
-    """
-    Run the predefined sample prompts
-    """
-    try:
-        responses = []
-        for prompt in SAMPLE_PROMPTS:
-            result = await chain.ainvoke({"input": prompt})
-            responses.append({
-                "message": prompt,
-                "response": result["text"]
-            })
-        return BatchChatResponse(responses=responses)
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
 @app.get("/health")
 async def health():
     """Health check endpoint"""

sample-apps/python/langchain_service/server.py renamed to sample-apps/python/genai_service/server.py

@@ -2,7 +2,7 @@
 
 # Configuration
 SERVER_URL="${SERVER_URL:-http://localhost:8000}"
-ENDPOINT="${SERVER_URL}/chat"
+ENDPOINT="${SERVER_URL}/ai-chat"
 DELAY_SECONDS="${DELAY_SECONDS:-3600}"  # Default 1 hour (3600 seconds) between requests
 NUM_REQUESTS="${NUM_REQUESTS:-0}"    # 0 means infinite
 TIMEOUT="${TIMEOUT:-30}"              # Request timeout in seconds

sample-apps/python/langchain_service/Dockerfile

@@ -95,6 +95,7 @@ cd /app
 aws s3 cp ${var.service_zip_url} langchain-service.zip
 unzip langchain-service.zip
 
+# Having issues installing dependencies from ec2-requirements.txt as these dependencies are quite large and cause timeouts/memory issues on EC2, manually installing instead
 python3.12 -m pip install fastapi uvicorn[standard] --no-cache-dir
 python3.12 -m pip install boto3 botocore setuptools --no-cache-dir
 python3.12 -m pip install opentelemetry-api opentelemetry-sdk opentelemetry-semantic-conventions --no-cache-dir

sample-apps/traffic-generator/langchain/generate_traffic.sh renamed to sample-apps/traffic-generator/genai/generate_traffic.sh

@@ -22,7 +22,7 @@ variable "test_id" {
 }
 
 variable "service_zip_url" {
-  default = "s3://sigv4perfresults/langchain-service.zip"
+  description = "S3 URL for the service zip file"
 }
 
 variable "language_version" {

terraform/python/ec2/langchain/main.tf renamed to terraform/python/ec2/adot-genai/main.tf

@@ -148,10 +148,10 @@ private Map<String, Object> getTrace() throws Exception {
               validationConfig.getHttpPath()));
     }
 
-    if (validationConfig.getHttpPath().contains("chat")) {
+    if (validationConfig.getHttpPath().contains("ai-chat")) {
       return this.getTraceById(Collections.singletonList(context.getTraceId()));
     }
-    
+
     log.info("Trace Filter: {}", traceFilter);
     List<TraceSummary> retrieveTraceLists = xrayService.searchTraces(traceFilter);
     List<String> traceIdLists = Collections.singletonList(retrieveTraceLists.get(0).getId());

terraform/python/ec2/langchain/variables.tf renamed to terraform/python/ec2/adot-genai/variables.tf

@@ -3,7 +3,7 @@
   "trace_id": "^{{traceId}}$",
   "http": {
     "request": {
-      "url": "^.*/chat$",
+      "url": "^.*/ai-chat$",
       "method": "^POST$"
     }
   },
@@ -12,7 +12,7 @@
   },
   "annotations": {
     "aws.local.service": "^{{serviceName}}$",
-    "aws.local.operation": "^POST /chat$"
+    "aws.local.operation": "^POST /ai-chat$"
   },
   "metadata": {
     "service.name": "^{{serviceName}}$"  

validator/src/main/java/com/amazon/aoc/validators/TraceValidator.java

@@ -1,6 +1,6 @@
 -
   validationType: "cw-log"
-  httpPath: "chat"
+  httpPath: "ai-chat"
   httpMethod: "post"
   callingType: "http"
   expectedLogStructureTemplate: "PYTHON_EC2_ADOT_GENAI_LOG"