知识卡片

普通读书笔记卡 · 2313.e.5.p.1.b

编排状态承接作业进度

理解等级:高级

管理器可比较来源条目与已建Job,创建尚未处理的工作,并利用Job状态记录运行与完成,减少自建状态存储。Job可以按策略重试失败进程,但资源不足、永久失败、超时或策略限制都可阻止成功;作业还可能重复启动,业务效果需幂等或去重。

这张卡片看不懂?

已有反馈 · 0 / 10 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • Chapter 11. Work Queue Systems · 查看原书依据

      the Kubernetes container orchestrator contains a number of features that make it significantly easier to implement. Namely, Kubernetes contains a Job object that allows for the reliable execution of the work queue. The Job can be configured to either run the worker container once or to run it

    参考来源

    - 位置:《Designing Distributed Systems, 2nd Edition》第11章《Chapter 11. Work Queue Systems》(来源项目 book-sources;源文件:extracted/Designing Distributed Systems, 2nd Edition/_epub-src/OEBPS/ch11.html) - 结论依据:原文说明Job与注解复用,但其“最终成功”须限定为策略和可恢复故障条件。 - 原始内容:the Kubernetes container orchestrator contains a number of features that make it significantly easier to implement. Namely, Kubernetes contains a Job object that allows for the reliable execution of the work queue. The Job can be configured to either run the worker container once or to run it