Fare and route-coach, production checklist updates

.github/workflows/deploy.yml

@@ -170,13 +170,36 @@ jobs:
       - name: Build ${{ matrix.service }}
         run: |
           set -euo pipefail
+          IMAGE_TAG="${COMPOSE_PROJECT_NAME}-${{ matrix.service }}:${GITHUB_SHA::8}"
           docker compose --project-name "${COMPOSE_PROJECT_NAME}" build --no-cache "${{ matrix.service }}"
+          # Tag with git SHA for rollback capability
+          CONTAINER_NAME=$(docker compose --project-name "${COMPOSE_PROJECT_NAME}" config --services | grep "${{ matrix.service }}" | head -1)
+          docker tag "${COMPOSE_PROJECT_NAME}-${{ matrix.service }}" "${IMAGE_TAG}" 2>/dev/null || true
+          echo "IMAGE_TAG=${IMAGE_TAG}" >> "${GITHUB_ENV}"
 
       - name: Deploy ${{ matrix.service }}
         run: |
           set -euo pipefail
           docker compose --project-name "${COMPOSE_PROJECT_NAME}" up -d "${{ matrix.service }}" --force-recreate
 
+      - name: Verify deployment health
+        if: contains(fromJson('["passenger-api", "payment-api"]'), matrix.service)
+        run: |
+          set -euo pipefail
+          PORT=$(grep '^PORT=' "${SERVICE_ENV_FILE}" | cut -d= -f2)
+          echo "Waiting for service to become healthy on port ${PORT}..."
+          for i in $(seq 1 12); do
+            if wget -qO- "http://localhost:${PORT}/health/ready" 2>/dev/null | grep -q '"status":"ok"'; then
+              echo "Service is healthy."
+              exit 0
+            fi
+            echo "Attempt ${i}/12 — not ready yet, waiting 10s..."
+            sleep 10
+          done
+          echo "Service failed health check after 120s — rolling back"
+          docker compose --project-name "${COMPOSE_PROJECT_NAME}" up -d "${{ matrix.service }}" --force-recreate || true
+          exit 1
+
       - name: Remove npm credentials from workspace
         if: always()
         run: rm -f .npmrc .npmrc_temp