Merge pull request 'fix: Worker health check 扫描全部 /proc 进程修复 unhealthy 误判' (#1825) from fix/worker-healthcheck-ci-e2e into develop

2026-09-09 19:09:14 +08:00
parent da2302b7ad 54defa4429
commit 1bd0f0c1ae
3 changed files with 4 additions and 4 deletions
@@ -128,9 +128,9 @@ 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
+    # 注：容器内无 pgrep/ps，扫描 /proc 所有进程的 cmdline 查找 celery 进程
    healthcheck:
-      test: ["CMD-SHELL", "for pid in /proc/[0-9]*/cmdline; do if grep -ql celery \"$pid\" 2>/dev/null; then exit 0; fi; done; exit 1"]
+      test: ["CMD-SHELL", "grep -lq celery /proc/[0-9]*/cmdline 2>/dev/null || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
@@ -155,7 +155,7 @@ docker run -d \
  --restart unless-stopped \
  --cpus 2 \
  --memory 2g \
-  --health-cmd "sh -c \"for pid in /proc/[0-9]*/cmdline; do if grep -ql celery \"$pid\" 2>/dev/null; then exit 0; fi; done; exit 1\"" \
+  --health-cmd "sh -c \"grep -lq celery /proc/[0-9]*/cmdline 2>/dev/null || exit 1\"" \
  --health-interval 30s \
  --health-timeout 10s \
  --health-retries 3 \
@@ -116,7 +116,7 @@ docker run -d \
  -v "$GENERATED_DIR:/app/generated" \
  --restart unless-stopped \
  --label com.centurylinklabs.watchtower.enable=true \
-  --health-cmd "sh -c \"for pid in /proc/[0-9]*/cmdline; do if grep -ql celery \"$pid\" 2>/dev/null; then exit 0; fi; done; exit 1\"" \
+  --health-cmd "sh -c \"grep -lq celery /proc/[0-9]*/cmdline 2>/dev/null || exit 1\"" \
  --health-interval 30s \
  --health-timeout 10s \
  --health-retries 3 \