momo's Blog.

Kubernetes 部署时提示 'Failed create pod sandbox'

pod
字数统计: 434阅读时长: 2 min
2021/06/15 Share

前言

在部署容器的时候, Pod提示 

1
Failed create pod sandbox: rpc error: code = Unknown desc = failed to create containerd task: OCI runtime create failed: container_linux.go:346: starting container process caused "process_linux.go:319: getting the final child's pid from pipe caused \"read init-p: connection reset by peer\"": unknown

原因

在宿主机上, 我们查看/var/log/message 发现有许多OOM的日志

1
2
3
4
5
6
7
8
9
10
11
12
Jun 14 19:56:26 10-11-10-85 kernel: Task in /kubepods/pod7325612f-d0a8-4bda-880c-510bddc9dce8/f03200d29dc96ea14ee5a5f65ae490569c8cfe32b5delled as a result of limit of /kubepods/pod7325612f-d0a8-4bda-880c-510bddc9dce8
Jun 14 19:56:26 10-11-10-85 kernel: memory: usage 0kB, limit 0kB, failcnt 377
Jun 14 19:56:26 10-11-10-85 kernel: memory+swap: usage 0kB, limit 9007199254740988kB, failcnt 0
Jun 14 19:56:26 10-11-10-85 kernel: kmem: usage 0kB, limit 9007199254740988kB, failcnt 0
Jun 14 19:56:26 10-11-10-85 kernel: Memory cgroup stats for /kubepods/pod7325612f-d0a8-4bda-880c-510bddc9dce8: cache:0KB rss:0KB rss_huge:le:0KB dirty:0KB writeback:0KB swap:0KB inactive_anon:0KB active_anon:0KB inactive_file:0KB active_file:0KB unevictable:0KB
Jun 14 19:56:26 10-11-10-85 kernel: Memory cgroup stats for /kubepods/pod7325612f-d0a8-4bda-880c-510bddc9dce8/f03200d29dc96ea14ee5a5f65ae4508487023c483: cache:0KB rss:0KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:0KB writeback:0KB swap:0KB inactive_anon:0KB active_anon:0KBve_file:0KB unevictable:0KB
Jun 14 19:56:26 10-11-10-85 kernel: Tasks state (memory values in pages):
Jun 14 19:56:26 10-11-10-85 kernel: [  pid  ]   uid  tgid total_vm      rss pgtables_bytes swapents oom_score_adj name
Jun 14 19:56:26 10-11-10-85 kernel: [ 294465]     0 294465     5735      995    73728        0          -998 exe
Jun 14 19:56:26 10-11-10-85 kernel: Memory cgroup out of memory: Kill process 294465 (exe) score 1013000 or sacrifice child
Jun 14 19:56:26 10-11-10-85 kernel: Killed process 294465 (exe) total-vm:22940kB, anon-rss:1508kB, file-rss:2472kB, shmem-rss:0kB
Jun 14 19:56:26 10-11-10-85 kernel: oom_reaper: reaped process 294465 (exe), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB

我的YAML是这样写的

1
2
3
4
resources:
   limits:
   cpu: "250m"
   memory: "300m"

实际上上面的写法写错了, memory 的单位只能是 E, P, T, G, M, K, Ei, Pi, Ti, Gi, Mi, Ki.

所以我们需要改成

1
2
3
4
resources:
   limits:
   cpu: "250m"
   memory: "300Mi"

参考

  1. manage-resources-containers
  2. Quantity

原文作者：Momo

原文链接：https://mo.xmomo521.top/2021/06/15/Kubernetes-部署时提示-Failed-create-pod-sandbox/

发表日期：June 15th 2021, 11:00:32 am

更新日期：December 21st 2021, 3:32:58 pm

版权声明：本文采用知识共享署名-非商业性使用 4.0 国际许可协议进行许可

CATALOG
  1. 1. 前言
  2. 2. 原因
  3. 3. 参考
Total : 133
2024
2023
2022
2021
2020
2019
Terraform CD TLS cfssl Python django drf JWT kubernetes CA 安全 Kibana JS Vue Devops Jenkins Gitlab DevOps Linux swarm pod ipvs ca k8s CPU CFS Linux时区 CST monitor PromQL Ajax logging Channels 聊天室 RPC Docker 学习笔记 Thanos Process ansible cryptography docker django-rest-framework Git 流程图 js 前端 Kubernetes prometheus-operator logstash filebeat Prometheus-operator eslint network axios vue cd xterm Docker网络 Prom联邦集群 secret CI ElementUi backup acme.sh Vuex Daemonset Taints log 动态路由 Swarm 软路由 标准运维 蓝鲸智云 ElementUI 监控 nginx ERROR
Ansible Atlantis Argocd Cloud-init Alpine Django CA Go etcd Elasticsearch Helm 日志 前端 Jenkins Kubernetes Jsonnet Docker kubernetes Linux Nginx NAS Grafana PostgreSQL Prometheus Python TerraForm Redis Terraform python alertmanager alpine docker Git gitlab-runner k3s ELK Mysql SSH vscode calico Python/Django LVM mysql Mongo TLS logging myself 蓝鲸 OpenWrt 源码分析 TCPDUMP Chrome Containerd iptables 技巧 nginx