问题概述

查看kubelet或/var/log/messages日志一直包错,发现是孤儿pod,是由于其pod被删除后存储路径还保存在磁盘。

报错如下

[root@node5 ~]# journalctl -fu kubelet
-- Logs begin at Tue 2020-06-16 23:41:14 CST. --
Jun 19 17:25:12 node5 kubelet[4711]: E0619 17:25:12.038458 4711 kubelet_volumes.go:154] orphaned pod "27960f19-29f1-486a-9a9d-c6c9290d014a" found, but volume paths are still present on disk : There were a total of 2 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:14 node5 kubelet[4711]: E0619 17:25:14.071432 4711 kubelet_volumes.go:154] orphaned pod "27960f19-29f1-486a-9a9d-c6c9290d014a" found, but volume paths are still present on disk : There were a total of 2 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:16 node5 kubelet[4711]: E0619 17:25:16.037737 4711 kubelet_volumes.go:154] orphaned pod "27960f19-29f1-486a-9a9d-c6c9290d014a" found, but volume paths are still present on disk : There were a total of 2 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:18 node5 kubelet[4711]: E0619 17:25:18.070147 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:20 node5 kubelet[4711]: E0619 17:25:20.036447 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:22 node5 kubelet[4711]: E0619 17:25:22.069562 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:24 node5 kubelet[4711]: E0619 17:25:24.065490 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:26 node5 kubelet[4711]: E0619 17:25:26.073979 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:28 node5 kubelet[4711]: E0619 17:25:28.037987 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.
Jun 19 17:25:30 node5 kubelet[4711]: E0619 17:25:30.068136 4711 kubelet_volumes.go:154] orphaned pod "62e47eeb-8de7-4d4e-9e0f-28503d63be6a" found, but volume paths are still present on disk : There were a total of 1 errors similar to this. Turn up verbosity to see them.

解决办法

解决思路

  • 根据/var/log/messages或kubelet报错,查找孤儿pod
  • 通过mount命令查找孤儿pod的挂载
  • 卸载孤儿pod的存储挂载
  • 删除/var/lib/kubelet/pods下孤儿pod的存储路径

使用脚本处理,脚本内容如下

#!/bin/sh

orphanedPods=`cat /var/log/messages|grep 'orphaned pod'|awk -F '"' '{print $2}'|uniq`;
orphanedPodsNum=`echo $orphanedPods|awk -F ' ' '{print NF}'`;
echo -e "orphanedPods: $orphanedPodsNum \n$orphanedPods"; for i in $orphanedPods
do
echo "Deleting Orphaned pod id: $i";
rm -rf /var/lib/kubelet/pods/$i;
done

在报错的节点直接执行脚本即可

相关issue:

https://github.com/kubernetes/kubernetes/issues/60987

https://github.com/kubernetes/kubernetes/pull/68616

处理 K8S Orphaned pod found - but volume paths are still present on disk 孤儿pod的更多相关文章

  1. k8s集群启动了上万个容器(一个pod里放上百个容器,起百个pod就模拟出上万个容器)服务器超时,无法操作的解决办法

    问题说明: 一个POD里放了百个容器,然后让K8S集群部署上百个POD,得到可运行上万个容器的实验目的. 实验环境:3台DELL裸机服务器,16核+64G,硬盘容量忽略吧,上T了,肯定够. 1.一开始 ...

  2. Pod容器共享Volume

    同一个Pod中的多个容器能够共享Pod级别的存储卷Volume.Volume可以被定义为各种类型,多个容器各自进行挂载操作,将一个Volume挂载为容器内部需要的目录,如图 在下面的例子中,在Pod内 ...

  3. Docker 与 K8S学习笔记(二十)—— 使用Downward API向容器注入Pod信息

    Kubernetes在创建Pod时,会为Pod和容器设置一些额外的信息,比如Pod名称.Pod IP.Node IP.Label.Annotation.资源限制等,我们经常会在应用程序中使用到这些数据 ...

  4. 二十二、Pod存储之volume

    Pod 的存储之volume ​容器磁盘上的文件的生命周期是短暂的,这就使得在容器中运行重要应用时会出现一些问题.首先,当容器崩溃时,kubelet 会重启它,但是容器中的文件将丢失--容器以干净的状 ...

  5. pod 的warning一定要注意消除,要不然你的pod配置有问题无法导入

    and the CHANGELOG for this version http://git.io/BaH8pQ. Downloading dependencies Using BlocksKit (2 ...

  6. [Linux] Boot分区满了的处理方法 The volume "boot" has only 0 bytes disk space remaining

    1.查看系统目前正在用的内核 abby@abby:~$ uname -r ..--generic 2.查看/boot保存的所有内核 abby@abby:~$ ls -lah /boot total 3 ...

  7. kubernetes 清理孤儿POD--转发

    孤儿pod的产生 节点OOM以后或者节点异常崩溃的情况下,pod未能被正常的清理而导致的孤儿进程. 提示如下 Orphaned pod found - but volume paths are sti ...

  8. kubernetes故障现场一之Orphaned pod

    系列目录 问题描述:周五写字楼整体停电,周一再来的时候发现很多pod的状态都是Terminating,经排查是因为测试环境kubernetes集群中的有些节点是PC机,停电后需要手动开机才能起来.起来 ...

  9. k8s 问题

    kubelet: Orphaned pod "4db449f0-4eaf-11e8-94ab-90b8d042b91a" found, but volume paths are s ...

随机推荐

  1. SpringBoot从入门到精通教程(七)

    今天,我们继续讲SpringBoot整合Redis ,也就缓存,它将与我们的Springboot整合 Redis 简介 Redis 是当前比较热门的NOSQL系统之一,它是一个开源的使用ANSI c语 ...

  2. 闲鱼上哪些商品抢手?Python 分析后告诉你

    1目 标 场 景 经常看到有朋友在闲鱼卖些小东西又或是自己擅长的一些技能,都能为他们带来不错的 睡后收入. 闲鱼上大量的商品,很难精准判断哪些受欢迎,哪些好卖:一个个录入数据去做数据分析,浪费时间的同 ...

  3. CI/CD自动化发版系统设计简介

    转载自:https://www.cnblogs.com/wellful/archive/2004/01/13/10604151.html 版本迭代是每一个互联网公司必须经历的,尤其是中小型公司,相信不 ...

  4. .net core 和 WPF 开发升讯威在线客服与营销系统:(插曲)一次端口攻击行为的分析与应对

    本系列文章详细介绍使用 .net core 和 WPF 开发 升讯威在线客服与营销系统 的过程.本产品已经成熟稳定并投入商用. 在线演示环境:https://kf.shengxunwei.com 注意 ...

  5. SQL语句中case,when,then的用法

    用法如下bai: 复制代码 SELECT s.s_id, s.s_name, s.s_sex, CASE WHENs.s_sex='1'THEN'男' WHENs.s_sex='2'THEN'女' E ...

  6. Java学习日报7.17

    控制台运行

  7. DBF 文件 ORACLE 数据库恢复

    DBF 文件 ORACLE 数据库恢复 清·魏源<庸易通义>:"至道问学之有知无行,分温故为存心,知新为致知,而敦厚为存心,崇礼为致知,此皆百密一疏." 起因 在我们的 ...

  8. 解决 unknown filesystem type ntfs U盘/移动硬盘挂载出错问题

    大内存U盘或者移动硬盘挂在再Linux 时,报错unknown filesystem type ntfs 1.安装ntfs-3g wget http://tuxera.com/opensource/n ...

  9. 浅入 .NET Core 中的内存和GC知识

    目录 托管代码 自动内存管理 参考资料: [1]https://docs.microsoft.com/zh-cn/dotnet/standard/managed-code [2]:https://do ...

  10. mysql修改sql_mode为宽松模式

    sql_mode ANSI TRADITIONAL STRICT_TRANS_TABLES sql_mode为空 最宽松的模式, 即使有错误既不会报错也不会有警告️ ANSI 宽松模式,对插入数据进行 ...