etcd-io/etcd

Introduce more failpoints into etcd watch

Open

#18,681 创建于 2024年10月5日

在 GitHub 查看
 (11 评论) (0 反应) (0 负责人)Go (51,701 star) (10,352 fork)batch import
area/robustness-testingarea/testinghelp wanted

描述

What would you like to be added?

Spin-off from https://github.com/etcd-io/etcd/issues/17529

Similar to https://github.com/etcd-io/etcd/pull/17680/files we should add more failpoints into watch and simulate:

  • slow watch sync
  • slow watch resynchronization
  • slow watch progress notify

etc

Why is this needed?

Prevent and detect other issues like https://github.com/etcd-io/etcd/issues/17529

贡献者指南

技术栈
go
领域
backendtesting
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
3
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-2 days
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
active
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
Goetcd watch internalsfailpoint library
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
45
研究方向
Study the existing failpoint implementation in etcd, particularly PR #17680 which added failpoints to other components. Identify the watch related files such as watchstore, watcher, etc. Understand the failpoint library used and how to inject delays. Then implement similar failpoints for slow watch sync, resynchronization, and progress notify.