别人教怎么写，我们教怎么不出事——生产环境事故复盘与工程化实战https://ganhuo.dev/https://ganhuo.dev/blog/api-timeout-investigation/https://ganhuo.dev/blog/api-timeout-investigation/监控报了几十次接口超时，从调用方日志、服务方日志、MySQL 慢 SQL、JVM Full GC 四个方向逐一排查，最后落在一个没有结论的推测上Mon, 29 Jun 2026 00:00:00 GMThttps://ganhuo.dev/blog/cloudflare-pages-blog/https://ganhuo.dev/blog/cloudflare-pages-blog/从零开始，使用 Astro + Cloudflare Pages 搭建一个免费、快速的个人技术博客站点Sun, 01 Dec 2024 00:00:00 GMThttps://ganhuo.dev/blog/dragonfly-p2p-cache-avalanche/https://ganhuo.dev/blog/dragonfly-p2p-cache-avalanche/交换机故障导致 100 多台物理机断网，containerd 环境下 Dragonfly P2P 缓存清空，2000 个容器挂了一小时的复盘Mon, 29 Jun 2026 00:00:00 GMThttps://ganhuo.dev/blog/ebpf-probe-24-servers-crash/https://ganhuo.dev/blog/ebpf-probe-24-servers-crash/CentOS 4.18 内核的 eBPF 上下文清理 bug 导致 24 台物理机内核崩溃、900 个 Pod 被重建的完整复盘Mon, 29 Jun 2026 00:00:00 GMThttps://ganhuo.dev/blog/etcd-revision-explosion-outage/https://ganhuo.dev/blog/etcd-revision-explosion-outage/一个 --auto-compaction-retention 参数没配，etcd 4000 万条历史版本撑爆 8GB 存储，602 个坐席打不了电话的完整复盘Mon, 29 Jun 2026 00:00:00 GMThttps://ganhuo.dev/blog/mysql-master-slave-avalanche/https://ganhuo.dev/blog/mysql-master-slave-avalanche/上游 MySQL 主从切换引发接口超时，共享线程池 + 默认 JVM + WMB 自动重试三颗雷连环引爆，话务服务雪崩 9 小时的完整复盘Mon, 29 Jun 2026 00:00:00 GMThttps://ganhuo.dev/blog/solo-spring-scaffold-open-source/https://ganhuo.dev/blog/solo-spring-scaffold-open-source/厌倦 Initializr 空壳粘贴？solo-spring-scaffold 一键生成 Spring Boot 3 五模块骨架，H2 即跑，JWT/Docker/测试插件按需装，代码全归你。Sun, 05 Jul 2026 00:00:00 GMThttps://ganhuo.dev/blog/spring-boot-best-practices/https://ganhuo.dev/blog/spring-boot-best-practices/Spring Boot 项目开发中的规范、配置和最佳实践Fri, 15 Nov 2024 00:00:00 GMT