CustomizedView - updates when node shutdown ungracefully

未关闭
#2,568 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

评估

难度
5/5
预计耗时
一周以上
新手友好度
25/100
Issue 类型
缺陷
描述清晰度
需要澄清
活跃度
停滞
技术栈
java

调研方向

首先跟踪 issue 中描述的参与者重新加入路径、LIVEINSTANCE 删除处理以及 Customized View 聚合。确定参与者在非正常关闭后如何重新填充其自定义状态,并验证过期的实例级状态不再被错误发布。

由索引模型根据 Issue 内容生成。

描述

bug
Describe the bug

Helix Customized View has following behavior

  • participant maintains the customized-state per-instance under /INSTANCE/CUSTOMIZED_VIEW//resource
  • Participant is responsible updating the state.
  • Helix controller aggregate per-instance customized-state and publishes at cluster-level.

So per instance state under given cluster, path - /INSTANCE/CUSTOMIZEDSTATES/ is maintained purely by instance.

Instance can go gracefully down, in which case they can update the state before going down but in case instance shutdown due to ungraceful shutdown, CUSTOMIZEDSTATE becomes STALE.

From Helix controller side, it is difficult to make out when LIVEINSTANCE ephmeral node is deleted whether: host temporarily disconnected, host gracefully shutdown or host ungracefully went down. Controller can't decide and so it doens't know when aggregating.

If customizedstate maintained session, then also you can't make out as temporary disconnect vs/vs ungraceful will be hard to do it.

So we need to find out if there are any hooks when participant re-joins the cluster, it populates the value.

To Reproduce

Steps to reproduce the behavior.

Expected behavior

A clear and concise description of what you expected to happen.

Additional context

Add any other context about the problem here.

主要语言
Java
星标
505
派生
258
平均合并
4 天 20 小时
30 天内合并 PR
3

环境准备

  • 没有 Dockerfile 或 Docker Compose 文件
  • 有 Pull Request 模板
  • 没有贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

apache/helix 的其他 Issue

查看 apache/helix 的全部 Issue

相似的 Issue

更多 Java Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。