Merge remote-tracking branch 'origin/genesis-release-test-p2' into genesis-release-test

Author: liustve

.github/workflows/node-k8s-test.yml

@@ -94,13 +94,17 @@ jobs:
             NODE_REMOTE_SAMPLE_APP_IMAGE, e2e-test/node-remote-sample-app-image
             RELEASE_TESTING_ECR_ACCOUNT, e2e-test/${{ github.event.repository.name }}/node-k8s-release-testing-account
 
+      # Temporarily use Java K8s Canary Cluster for Node.js e2e tests
       - name: Retrieve K8s EC2 Secrets
         if: ${{ env.CALLER_WORKFLOW_NAME != 'k8s-os-patching' }}
         uses: aws-actions/aws-secretsmanager-get-secrets@v1
         with:
+          # secret-ids: |
+          #   MAIN_SERVICE_ENDPOINT, e2e-test/${{ github.event.repository.name }}/node-k8s-master-node-endpoint
+          #   MASTER_NODE_SSH_KEY, e2e-test/${{ github.event.repository.name }}/node-k8s-ssh-key
           secret-ids: |
-            MAIN_SERVICE_ENDPOINT, e2e-test/${{ github.event.repository.name }}/node-k8s-master-node-endpoint
-            MASTER_NODE_SSH_KEY, e2e-test/${{ github.event.repository.name }}/node-k8s-ssh-key
+            MAIN_SERVICE_ENDPOINT, e2e-test/aws-application-signals-test-framework/java-k8s-master-node-endpoint
+            MASTER_NODE_SSH_KEY, e2e-test/aws-application-signals-test-framework/java-k8s-ssh-key
 
       - name: Retrieve K8s EC2 OS Patching Secrets
         if: ${{ env.CALLER_WORKFLOW_NAME == 'k8s-os-patching' }}

.github/workflows/python-ec2-genai-test.yml

@@ -4,24 +4,11 @@
 name: Python EC2 Gen AI Use Case
 on:
   workflow_dispatch: # be able to run the workflow on demand
-  push:
-    branches:
-      - genesis-release-test
   workflow_call:
     inputs:
       caller-workflow-name:
         required: true
         type: string
-      python-version:
-        description: "Currently support version 3.9, 3.10, 3.11, 3.12, 3.13"
-        required: false
-        type: string
-        default: '3.12'
-      cpu-architecture:
-        description: "Permitted values: x86_64 or arm64"
-        required: false
-        type: string
-        default: "x86_64"
       staging-wheel-name:
         required: false
         default: 'aws-opentelemetry-distro'
@@ -35,8 +22,7 @@ env:
   E2E_TEST_AWS_REGION: 'us-west-2'
   E2E_TEST_ACCOUNT_ID: ${{ secrets.APPLICATION_SIGNALS_E2E_TEST_ACCOUNT_ID }}
   E2E_TEST_ROLE_NAME: ${{ secrets.APPLICATION_SIGNALS_E2E_TEST_ROLE_NAME }}
-  ADOT_WHEEL_NAME: aws-opentelemetry-distro
-  PYTHON_VERSION: 3.12
+  ADOT_WHEEL_NAME: ${{ inputs.staging-wheel-name }}
   METRIC_NAMESPACE: genesis
   LOG_GROUP_NAME: test/genesis
   TEST_RESOURCES_FOLDER: ${GITHUB_WORKSPACE}
@@ -50,18 +36,18 @@ jobs:
       - uses: actions/checkout@v4
 
       - name: Set Get ADOT Wheel command environment variable
-        run: echo GET_ADOT_WHEEL_COMMAND="aws s3 cp s3://adot-main-build-staging-jar/${{ env.ADOT_WHEEL_NAME }} ./${{ env.ADOT_WHEEL_NAME }} && python${{ env.PYTHON_VERSION }} -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV 
-          # if [ "${{ github.event.repository.name }}" = "aws-otel-python-instrumentation" ]; then
-          #   # Reusing the adot-main-build-staging-jar bucket to store the python wheel file
-          #   echo GET_ADOT_WHEEL_COMMAND="aws s3 cp s3://adot-main-build-staging-jar/${{ env.ADOT_WHEEL_NAME }} ./${{ env.ADOT_WHEEL_NAME }} && python${{ env.PYTHON_VERSION }} -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
-          # elif [ "${{ env.OTEL_SOURCE }}" == "pypi" ]; then
-          #   echo GET_ADOT_WHEEL_COMMAND="python${{ env.PYTHON_VERSION }} -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
-          # else
-          #   latest_release_version=$(curl -sL https://github.com/aws-observability/aws-otel-python-instrumentation/releases/latest | grep -oP '/releases/tag/v\K[0-9]+\.[0-9]+\.[0-9]+' | head -n 1)
-          #   echo "The latest version is $latest_release_version"
-          #   echo GET_ADOT_WHEEL_COMMAND="wget -O ${{ env.ADOT_WHEEL_NAME }} https://github.com/aws-observability/aws-otel-python-instrumentation/releases/latest/download/aws_opentelemetry_distro-$latest_release_version-py3-none-any.whl \
-          #   && python${{ env.PYTHON_VERSION }} -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
-          # fi
+        run: |
+          if [ "${{ github.event.repository.name }}" = "aws-otel-python-instrumentation" ]; then
+            # Reusing the adot-main-build-staging-jar bucket to store the python wheel file
+            echo GET_ADOT_WHEEL_COMMAND="aws s3 cp s3://adot-main-build-staging-jar/${{ env.ADOT_WHEEL_NAME }} ./${{ env.ADOT_WHEEL_NAME }} && python3.12 -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
+          elif [ "${{ env.OTEL_SOURCE }}" == "pypi" ]; then
+            echo GET_ADOT_WHEEL_COMMAND="python3.12 -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
+          else
+            latest_release_version=$(curl -sL https://github.com/aws-observability/aws-otel-python-instrumentation/releases/latest | grep -oP '/releases/tag/v\K[0-9]+\.[0-9]+\.[0-9]+' | head -n 1)
+            echo "The latest version is $latest_release_version"
+            echo GET_ADOT_WHEEL_COMMAND="wget -O ${{ env.ADOT_WHEEL_NAME }} https://github.com/aws-observability/aws-otel-python-instrumentation/releases/latest/download/aws_opentelemetry_distro-$latest_release_version-py3-none-any.whl \
+            && python3.12 -m pip install ${{ env.ADOT_WHEEL_NAME }}" >> $GITHUB_ENV
+          fi
 
       - name: Initiate Gradlew Daemon
         uses: ./.github/workflows/actions/execute_and_retry
@@ -119,7 +105,6 @@ jobs:
             -var="service_zip_url=${{ env.SAMPLE_APP_ZIP }}" \
             -var="trace_id=${{ env.TRACE_ID_HEADER }}" \
             -var="get_adot_wheel_command=${{ env.GET_ADOT_WHEEL_COMMAND }}" \
-            -var="python_version=${{ env.PYTHON_VERSION }}"
 
       - name: Get deployment info
         working-directory: terraform/python/ec2/adot-genai
@@ -152,12 +137,6 @@ jobs:
           --instance-id ${{ env.INSTANCE_ID }}
           --trace-id ${{ env.XRAY_TRACE_ID }}'
 
-      - name: Wait for metrics to be published
-        if: (success() || failure()) && !cancelled()
-        run: |
-          echo "Waiting 60 seconds to ensure EMF metrics are published to CloudWatch"
-          sleep 60
-
       - name: Validate generated metrics
         if: (success() || failure()) && !cancelled()
         run: ./gradlew validator:run --args='-c python/ec2/adot-genai/metric-validation.yml

sample-apps/python/genai_service/ec2-requirements.txt

@@ -10,6 +10,6 @@ openlit
 botocore
 setuptools
 boto3
-aws_opentelemetry_distro_genai_beta
+aws_opentelemetry_distro
 fastapi
 uvicorn[standard]

terraform/python/ec2/adot-genai/main.tf

@@ -37,15 +37,15 @@ data "aws_ami" "ami" {
   most_recent = true
   filter {
     name = "name"
-    values = ["al2023-ami-*-${var.cpu_architecture}"]
+    values = ["al2023-ami-*-x86_64"]
   }
   filter {
     name   = "state"
     values = ["available"]
   }
   filter {
     name   = "architecture"
-    values = [var.cpu_architecture]
+    values = ["x86_64"]
   }
   filter {
     name   = "virtualization-type"
@@ -55,7 +55,7 @@ data "aws_ami" "ami" {
 
 resource "aws_instance" "main_service_instance" {
   ami                                   = data.aws_ami.ami.id
-  instance_type                         = var.cpu_architecture == "x86_64" ? "t3.medium" : "t4g.medium"
+  instance_type                         = "t3.medium"
   key_name                              = local.ssh_key_name
   iam_instance_profile                  = "APP_SIGNALS_EC2_TEST_ROLE"
   vpc_security_group_ids                = [aws_default_vpc.default.default_security_group_id]
@@ -73,20 +73,20 @@ resource "aws_instance" "main_service_instance" {
   user_data = base64encode(<<-EOF
 #!/bin/bash
 yum update -y
-yum install -y python${var.python_version} python${var.python_version}-pip unzip bc
+yum install -y python3.12 python3.12-pip unzip bc
 
 mkdir -p /app
 cd /app
 aws s3 cp ${var.service_zip_url} genai-service.zip
 unzip genai-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
-python${var.python_version} -m pip install fastapi uvicorn[standard] --no-cache-dir
-python${var.python_version} -m pip install boto3 botocore setuptools --no-cache-dir
-python${var.python_version} -m pip install opentelemetry-api opentelemetry-sdk opentelemetry-semantic-conventions --no-cache-dir
-python${var.python_version} -m pip install langchain langchain-community langchain_aws --no-cache-dir
-python${var.python_version} -m pip install python-dotenv openlit --no-cache-dir
-python${var.python_version} -m pip install openinference-instrumentation-langchain --no-cache-dir
+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
+python3.12 -m pip install langchain langchain-community langchain_aws --no-cache-dir
+python3.12 -m pip install python-dotenv openlit --no-cache-dir
+python3.12 -m pip install openinference-instrumentation-langchain --no-cache-dir
 ${var.get_adot_wheel_command}
 
 export AWS_REGION=${var.aws_region}
@@ -97,7 +97,7 @@ export OTEL_EXPORTER_OTLP_LOGS_HEADERS="x-aws-log-group=test/genesis,x-aws-log-s
 export OTEL_RESOURCE_ATTRIBUTES="service.name=langchain-traceloop-app"
 export AGENT_OBSERVABILITY_ENABLED="true"
 
-nohup opentelemetry-instrument python${var.python_version} server.py > /var/log/langchain-service.log 2>&1 &
+nohup opentelemetry-instrument python3.12 server.py > /var/log/langchain-service.log 2>&1 &
 
 # Wait for service to be ready
 echo "Waiting for service to be ready..."

terraform/python/ec2/adot-genai/variables.tf

@@ -25,13 +25,9 @@ variable "service_zip_url" {
   description = "S3 URL for the service zip file"
 }
 
-variable "python_version" {
-  default = "3.12"
-}
 
-variable "cpu_architecture" {
-  default = "x86_64"
-}
+
+
 
 variable "user" {
   default = "ec2-user"

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

@@ -92,7 +92,6 @@ public void validate() throws Exception {
         maxRetryCount,
         () -> {
           String httpPath = validationConfig.getHttpPath();
-
           // Special handling for Genesis path - just check if any metrics exists in namespace
           // since ADOT will just capture any OTel Metrics emitted from the instrumentation library used
           // and convert them into EMF metrics, it's impossible to create a validation template for this.

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

@@ -148,7 +148,7 @@ private Map<String, Object> getTrace() throws Exception {
               validationConfig.getHttpPath()));
     }
 
-    if (validationConfig.getHttpPath().contains("ai-chat")) {
+    if (validationConfig.getHttpPath() != null && validationConfig.getHttpPath().contains("ai-chat")) {
       return this.getTraceById(Collections.singletonList(context.getTraceId()));
     }