ci: 瞬态故障自动重试——runner 故障时自动 re-run,无需人工干预 #1572

Merged
xiaoxia merged 1 commits from ci/transient-fault-auto-retry into develop 2026-08-31 15:23:41 +08:00

1 Commits

Author SHA1 Message Date
xiaoxia 115dc462d7 ci: CI 瞬态故障自动重试机制
CI/CD Pipeline / Check push changed paths (pull_request) Has been skipped
CI/CD Pipeline / Dedup Check - skip PR tests when covered by push pipeline (pull_request) Successful in 1s
CI/CD Pipeline / Build Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Retag skipped Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / ACR Image Cleanup (pull_request) Has been skipped
CI/CD Pipeline / Check if frontend-only change (pull_request) Successful in 3m28s
CI/CD Pipeline / Frontend Lint (pull_request) Has been skipped
CI/CD Pipeline / Frontend Unit Tests (pull_request) Has been skipped
CI/CD Pipeline / PR Build Web Image (pull_request) Has been skipped
Preview Deploy / Deploy Preview Environment (pull_request) Successful in 4m19s
CI/CD Pipeline / Validate - Migration (alembic) (pull_request) Successful in 4m22s
PR Automation / Auto Merge on CI Green + Approved (pull_request) Successful in 4m26s
CI/CD Pipeline / PR Build API Image (pull_request) Successful in 1m20s
CI/CD Pipeline / PR Build Worker Image (pull_request) Successful in 1m21s
CI/CD Pipeline / Validate - Type Check (mypy) (pull_request) Successful in 5m11s
PR Automation / Auto Approve on CI Green (pull_request) Successful in 5m56s
CI/CD Pipeline / Validate - Code Quality (pull_request) Successful in 8m16s
AI Code Review / AI Code Review (pull_request) Successful in 9m3s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 6m32s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 15m44s
CI/CD Pipeline / Build Production API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Canary Release to Production (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
CI/CD Pipeline / CI Gate (pull_request) Successful in 3m16s
ACR Cleanup / ACR Image Cleanup (pull_request_target) Successful in 4m6s
Preview Cleanup / Cleanup Preview Environment (pull_request) Successful in 4m35s
问题:
  当 runner 出现瞬态故障(磁盘满/docker挂/网络抖动),分配到该 runner
  的 job 会在 checkout/setup 阶段失败。这种失败与代码无关,但没有自动
  重试机制,需要人工干预。

方案:
  ci_transient_retry.sh - 每 5 分钟扫描最近 30 分钟的 runs:
  - 检测特征:失败 job 的 checkout/setup 阶段失败 + 同 run 其他 job 成功
  - 自动 re-run 整个 workflow(Gitea API /rerun)
  - 每个 run 最多重试 1 次(状态文件跟踪,24h 自动清理)
  - 支持 --dry-run 模式
  - 日志记录到 /var/log/ci-auto-retry.log

判定逻辑:
  - 失败 job 第一个 step 是 checkout/setup 且 conclusion=failure → 瞬态
  - 同一 run 至少 2 个 job 成功(排除代码全挂)
  - 纯代码失败(lint/test/compile)→ 不重试

部署:
  - systemd timer: ci_transient_retry.{service,timer}
  - 或 crontab: */5 * * * *
  - Token 通过环境变量 / EnvironmentFile 注入,不硬编码
2026-08-31 14:57:34 +08:00