Merge pull request 'Merge develop into main (v0.1.105: Worker healthcheck fix)' (#178) from develop into main

.gitea/workflows/deploy.yml

@@ -9,7 +9,7 @@ on:
 jobs:
   deploy-staging:
     name: Deploy Staging
-    runs-on: ubuntu-22.04
+    runs-on: saas
     if: github.ref_name == 'main' || github.ref_name == 'develop' || startsWith(github.ref_name, 'feature/')
 
     steps:
@@ -140,7 +140,7 @@ jobs:
 
   build-production-runtime-images:
     name: Build Production Runtime Images
-    runs-on: ubuntu-22.04
+    runs-on: saas
     if: startsWith(github.ref, 'refs/tags/v')
 
     steps:
@@ -249,7 +249,7 @@ jobs:
 
   deploy-production:
     name: Deploy Production
-    runs-on: ubuntu-22.04
+    runs-on: saas
     if: startsWith(github.ref, 'refs/tags/v')
     needs: build-production-runtime-images
 
@@ -290,7 +290,7 @@ jobs:
 
   production-e2e:
     name: Production Browser E2E
-    runs-on: ubuntu-22.04
+    runs-on: saas
     if: startsWith(github.ref, 'refs/tags/v')
     needs: deploy-production
 

infra/docker/compose.yml

@@ -120,12 +120,13 @@ services:
       - xiaoxia-net
 
     # 健康检查配置
+    # 注：celery inspect ping 依赖 broker 连接，在容器内不可靠，改用进程检查
     healthcheck:
-      test: ["CMD-SHELL", "celery -A app.celery_app inspect ping --timeout 5 || exit 1"]
+      test: ["CMD-SHELL", "pgrep -f 'celery.*worker' || exit 1"]
       interval: 30s
       timeout: 10s
       retries: 3
-      start_period: 40s
+      start_period: 30s
 
     logging: *default-logging