try another approach

chaptersix · chaptersix · commit e4e27bf92fb7 · 2025-12-16T13:32:04.000-06:00
diff --git a/.github/workflows/compose.yaml b/.github/workflows/compose.yaml
@@ -58,71 +58,39 @@ jobs:
 
       - name: Start compose stack
         working-directory: compose
-        run: docker compose -f ${{ matrix.compose-file }} up -d || true
+        run: docker compose -f ${{ matrix.compose-file }} up -d
 
-      - name: Check admin-tools completion
+      - name: Start validation container
         working-directory: compose
-        run: |
-          # Wait a moment for container to start and potentially fail
-          sleep 10
-          # Check if admin-tools container exists and get its exit code
-          if docker compose -f ${{ matrix.compose-file }} ps -a temporal-admin-tools --format json | jq -e '.[0]' > /dev/null 2>&1; then
-            EXIT_CODE=$(docker compose -f ${{ matrix.compose-file }} ps -a temporal-admin-tools --format json | jq -r '.[0].ExitCode // "null"')
-            if [ "$EXIT_CODE" != "0" ] && [ "$EXIT_CODE" != "null" ]; then
-              echo "Admin-tools failed with exit code: $EXIT_CODE"
-              docker compose -f ${{ matrix.compose-file }} logs temporal-admin-tools
-              exit 1
-            fi
-          fi
+        run: docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml up temporal-validate -d
 
-      - name: Wait for services to be healthy
+      - name: Wait for validation to complete
         working-directory: compose
         run: |
-          echo "Waiting up to 5 minutes for services to be healthy..."
-          timeout 300 bash -c 'until docker compose -f ${{ matrix.compose-file }} ps | grep -E "(healthy|Started)"; do sleep 5; done' || {
-            echo "Services failed to become healthy"
-            docker compose -f ${{ matrix.compose-file }} ps
-            docker compose -f ${{ matrix.compose-file }} logs
+          echo "Waiting for validation container to complete (max 5 minutes)..."
+          timeout 300 bash -c 'until docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml ps temporal-validate --format json | jq -e ".[0].State == \"exited\"" > /dev/null 2>&1; do sleep 2; done' || {
+            echo "Validation container did not complete in time"
+            docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml logs temporal-validate
             exit 1
           }
 
-      - name: Check admin-tools logs
-        working-directory: compose
-        run: |
-          echo "=== Admin-tools logs ==="
-          docker compose -f ${{ matrix.compose-file }} logs temporal-admin-tools 2>&1 | grep -v "level=warning" || true
-          echo "=== Checking for errors ==="
-          if docker compose -f ${{ matrix.compose-file }} logs temporal-admin-tools 2>&1 | grep -i "error" | grep -v "unable to describe namespace" | grep -v "level=warning"; then
-            echo "Found errors in admin-tools logs"
-            exit 1
-          fi
-          echo "No errors found in admin-tools logs"
+          # Check exit code
+          EXIT_CODE=$(docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml ps -a temporal-validate --format json | jq -r '.[0].ExitCode')
+          echo "Validation container exit code: $EXIT_CODE"
 
-      - name: Check namespace creation logs
-        working-directory: compose
-        run: |
-          echo "=== Namespace creation logs ==="
-          docker compose -f ${{ matrix.compose-file }} logs temporal-create-namespace 2>&1 | grep -v "level=warning" || true
-          echo "=== Checking for success message ==="
-          if ! docker compose -f ${{ matrix.compose-file }} logs temporal-create-namespace 2>&1 | grep -q "Namespace.*created"; then
-            echo "Namespace creation failed or did not complete"
+          if [ "$EXIT_CODE" != "0" ]; then
+            echo "Validation failed!"
+            docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml logs temporal-validate
             exit 1
           fi
-          echo "Namespace creation succeeded"
 
-      - name: Verify temporal server is healthy
-        working-directory: compose
-        run: |
-          docker compose -f ${{ matrix.compose-file }} exec -T temporal temporal operator cluster health || {
-            echo "Temporal server health check failed"
-            docker compose -f ${{ matrix.compose-file }} logs temporal
-            exit 1
-          }
+          echo "Validation passed!"
+          docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml logs temporal-validate
 
       - name: Cleanup
         if: always()
         working-directory: compose
-        run: docker compose -f ${{ matrix.compose-file }} down -v
+        run: docker compose -f ${{ matrix.compose-file }} -f docker-compose-validate.yml down -v
 
   compose-tls-test:
     name: Test docker-compose-tls.yml
@@ -151,63 +119,40 @@ jobs:
 
       - name: Start TLS compose stack
         working-directory: compose
-        run: COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml up -d || true
+        run: COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml up -d
 
-      - name: Check admin-tools-setup completion
+      - name: Start validation container
         working-directory: compose
-        run: |
-          # Wait a moment for container to start and potentially fail
-          sleep 10
-          # Check if admin-tools-setup container exists and get its exit code
-          if COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml ps -a temporal-admin-tools-setup --format json | jq -e '.[0]' > /dev/null 2>&1; then
-            EXIT_CODE=$(COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml ps -a temporal-admin-tools-setup --format json | jq -r '.[0].ExitCode // "null"')
-            if [ "$EXIT_CODE" != "0" ] && [ "$EXIT_CODE" != "null" ]; then
-              echo "Admin-tools-setup failed with exit code: $EXIT_CODE"
-              COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs temporal-admin-tools-setup
-              exit 1
-            fi
-          fi
+        run: COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml up temporal-validate -d
 
-      - name: Wait for services to be healthy
+      - name: Wait for validation to complete
         working-directory: compose
         run: |
-          echo "Waiting up to 5 minutes for services to be healthy..."
-          timeout 300 bash -c 'until COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml ps | grep -E "(healthy|Started)"; do sleep 5; done' || {
-            echo "Services failed to become healthy"
-            COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml ps
-            COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs
+          echo "Waiting for validation container to complete (max 5 minutes)..."
+          timeout 300 bash -c 'until COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml ps temporal-validate --format json | jq -e ".[0].State == \"exited\"" > /dev/null 2>&1; do sleep 2; done' || {
+            echo "Validation container did not complete in time"
+            COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml logs temporal-validate
             exit 1
           }
 
-      - name: Check admin-tools logs
-        working-directory: compose
-        run: |
-          echo "=== Admin-tools-setup logs ==="
-          COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs temporal-admin-tools-setup 2>&1 | grep -v "level=warning" || true
-          echo "=== Checking for errors ==="
-          if COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs temporal-admin-tools-setup 2>&1 | grep -i "error" | grep -v "unable to describe namespace" | grep -v "level=warning"; then
-            echo "Found errors in admin-tools logs"
-            exit 1
-          fi
-          echo "No errors found in admin-tools logs"
+          # Check exit code
+          EXIT_CODE=$(COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml ps -a temporal-validate --format json | jq -r '.[0].ExitCode')
+          echo "Validation container exit code: $EXIT_CODE"
 
-      - name: Check namespace creation logs
-        working-directory: compose
-        run: |
-          echo "=== Namespace creation logs ==="
-          COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs temporal-create-namespace 2>&1 | grep -v "level=warning" || true
-          echo "=== Checking for success message ==="
-          if ! COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml logs temporal-create-namespace 2>&1 | grep -q "Namespace.*created"; then
-            echo "Namespace creation failed or did not complete"
+          if [ "$EXIT_CODE" != "0" ]; then
+            echo "Validation failed!"
+            COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml logs temporal-validate
             exit 1
           fi
-          echo "Namespace creation succeeded"
+
+          echo "Validation passed!"
+          COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml logs temporal-validate
 
       - name: Cleanup
         if: always()
         working-directory: compose
         run: |
-          COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml down -v
+          COMPOSE_PROJECT_NAME=tls_test docker compose -f docker-compose-tls.yml -f docker-compose-validate.yml down -v
           docker volume rm temporal_tls_pki || true
           rm -rf .pki
 
@@ -230,45 +175,34 @@ jobs:
 
       - name: Start multirole compose stack
         working-directory: compose
-        run: docker compose -f docker-compose-multirole.yaml up -d || true
+        run: docker compose -f docker-compose-multirole.yaml up -d
 
-      - name: Check setup completion
+      - name: Start validation container
         working-directory: compose
-        run: |
-          # Wait a moment for container to start and potentially fail
-          sleep 10
-          # Check if setup container exists and get its exit code
-          if docker compose -f docker-compose-multirole.yaml ps -a temporal-setup --format json | jq -e '.[0]' > /dev/null 2>&1; then
-            EXIT_CODE=$(docker compose -f docker-compose-multirole.yaml ps -a temporal-setup --format json | jq -r '.[0].ExitCode // "null"')
-            if [ "$EXIT_CODE" != "0" ] && [ "$EXIT_CODE" != "null" ]; then
-              echo "Setup failed with exit code: $EXIT_CODE"
-              docker compose -f docker-compose-multirole.yaml logs temporal-setup
-              exit 1
-            fi
-          fi
+        run: docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml up temporal-validate -d
 
-      - name: Wait for services to be healthy
+      - name: Wait for validation to complete
         working-directory: compose
         run: |
-          echo "Waiting up to 5 minutes for services to be healthy..."
-          timeout 300 bash -c 'until docker compose -f docker-compose-multirole.yaml ps | grep -E "(healthy|Started)"; do sleep 5; done' || {
-            echo "Services failed to become healthy"
-            docker compose -f docker-compose-multirole.yaml ps
-            docker compose -f docker-compose-multirole.yaml logs
+          echo "Waiting for validation container to complete (max 5 minutes)..."
+          timeout 300 bash -c 'until docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml ps temporal-validate --format json | jq -e ".[0].State == \"exited\"" > /dev/null 2>&1; do sleep 2; done' || {
+            echo "Validation container did not complete in time"
+            docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml logs temporal-validate
             exit 1
           }
 
-      - name: Check setup logs
-        working-directory: compose
-        run: |
-          echo "=== Setup logs ==="
-          docker compose -f docker-compose-multirole.yaml logs temporal-setup 2>&1 | grep -v "level=warning" || true
-          echo "=== Checking for errors ==="
-          if docker compose -f docker-compose-multirole.yaml logs temporal-setup 2>&1 | grep -i "error" | grep -v "unable to describe namespace" | grep -v "level=warning"; then
-            echo "Found errors in setup logs"
+          # Check exit code
+          EXIT_CODE=$(docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml ps -a temporal-validate --format json | jq -r '.[0].ExitCode')
+          echo "Validation container exit code: $EXIT_CODE"
+
+          if [ "$EXIT_CODE" != "0" ]; then
+            echo "Validation failed!"
+            docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml logs temporal-validate
             exit 1
           fi
-          echo "No errors found in setup logs"
+
+          echo "Validation passed!"
+          docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml logs temporal-validate
 
       - name: Verify temporal services are running
         working-directory: compose
@@ -278,4 +212,4 @@ jobs:
       - name: Cleanup
         if: always()
         working-directory: compose
-        run: docker compose -f docker-compose-multirole.yaml down -v
+        run: docker compose -f docker-compose-multirole.yaml -f docker-compose-validate-multirole.yml down -v
diff --git a/compose/docker-compose-validate-multirole.yml b/compose/docker-compose-validate-multirole.yml
@@ -0,0 +1,29 @@
+version: '3.5'
+
+services:
+  temporal-validate:
+    image: temporalio/admin-tools:${TEMPORAL_ADMINTOOLS_VERSION}
+    container_name: temporal-validate
+    deploy:
+      restart_policy:
+        condition: on-failure
+        delay: 10s
+        max_attempts: 5
+    depends_on:
+      temporal-setup:
+        condition: service_completed_successfully
+    environment:
+      - TEMPORAL_ADDRESS=temporal-nginx:7233
+      - NAMESPACE=default
+      - MAX_WAIT_TIME=300
+    networks:
+      - temporal-network
+    volumes:
+      - ./scripts:/scripts
+    entrypoint: ["/bin/sh"]
+    command: /scripts/validate-temporal.sh
+
+networks:
+  temporal-network:
+    external: true
+    name: temporal-network
diff --git a/compose/docker-compose-validate.yml b/compose/docker-compose-validate.yml
@@ -0,0 +1,29 @@
+version: '3.5'
+
+services:
+  temporal-validate:
+    image: temporalio/admin-tools:${TEMPORAL_ADMINTOOLS_VERSION}
+    container_name: temporal-validate
+    deploy:
+      restart_policy:
+        condition: on-failure
+        delay: 10s
+        max_attempts: 5
+    depends_on:
+      temporal-create-namespace:
+        condition: service_completed_successfully
+    environment:
+      - TEMPORAL_ADDRESS=temporal:7233
+      - NAMESPACE=default
+      - MAX_WAIT_TIME=300
+    networks:
+      - temporal-network
+    volumes:
+      - ./scripts:/scripts
+    entrypoint: ["/bin/sh"]
+    command: /scripts/validate-temporal.sh
+
+networks:
+  temporal-network:
+    external: true
+    name: temporal-network
diff --git a/compose/scripts/validate-temporal.sh b/compose/scripts/validate-temporal.sh
@@ -0,0 +1,81 @@
+#!/bin/bash
+
+set -e
+
+# Configuration
+TEMPORAL_ADDRESS="${TEMPORAL_ADDRESS:-temporal:7233}"
+NAMESPACE="${NAMESPACE:-default}"
+MAX_WAIT_TIME="${MAX_WAIT_TIME:-300}"
+
+echo "=== Temporal Validation Script ==="
+echo "Temporal address: $TEMPORAL_ADDRESS"
+echo "Namespace: $NAMESPACE"
+echo "Max wait time: ${MAX_WAIT_TIME}s"
+echo
+
+# Wait a bit for Temporal to be ready
+echo "Waiting for Temporal server to be accessible..."
+sleep 15
+
+# Check cluster health
+echo "=== Checking cluster health ==="
+if temporal operator cluster health --address "$TEMPORAL_ADDRESS"; then
+    echo "✓ Cluster health check passed"
+else
+    echo "✗ Cluster health check failed"
+    exit 1
+fi
+
+# Describe namespace to verify it exists and is accessible
+echo
+echo "=== Verifying namespace '$NAMESPACE' ==="
+if temporal operator namespace describe --namespace "$NAMESPACE" --address "$TEMPORAL_ADDRESS" > /dev/null 2>&1; then
+    echo "✓ Namespace '$NAMESPACE' is accessible"
+else
+    echo "✗ Namespace '$NAMESPACE' not found or not accessible"
+    exit 1
+fi
+
+# Try to start a simple workflow to validate full functionality
+echo
+echo "=== Testing workflow execution ==="
+WORKFLOW_ID="validation-test-$(date +%s)"
+TASK_QUEUE="validation-queue"
+
+# Start a workflow (this will fail gracefully if no workers, but proves the system is working)
+if temporal workflow start \
+    --workflow-id "$WORKFLOW_ID" \
+    --type "NonExistentWorkflow" \
+    --task-queue "$TASK_QUEUE" \
+    --namespace "$NAMESPACE" \
+    --address "$TEMPORAL_ADDRESS" \
+    --workflow-execution-timeout 10s 2>&1 | grep -q "Started"; then
+    echo "✓ Successfully initiated workflow (proves Temporal is functional)"
+
+    # Clean up - terminate the workflow since it won't complete
+    temporal workflow terminate \
+        --workflow-id "$WORKFLOW_ID" \
+        --namespace "$NAMESPACE" \
+        --address "$TEMPORAL_ADDRESS" \
+        --reason "Validation complete" > /dev/null 2>&1 || true
+else
+    # Check if it's just missing workers (which is okay for validation)
+    if temporal workflow describe \
+        --workflow-id "$WORKFLOW_ID" \
+        --namespace "$NAMESPACE" \
+        --address "$TEMPORAL_ADDRESS" > /dev/null 2>&1; then
+        echo "✓ Workflow was accepted by server (functional)"
+        temporal workflow terminate \
+            --workflow-id "$WORKFLOW_ID" \
+            --namespace "$NAMESPACE" \
+            --address "$TEMPORAL_ADDRESS" \
+            --reason "Validation complete" > /dev/null 2>&1 || true
+    else
+        echo "✗ Failed to start workflow"
+        exit 1
+    fi
+fi
+
+echo
+echo "=== All validation checks passed! ==="
+exit 0
