“K8s基础”的版本间的差异
跳到导航
跳到搜索
(→helm) |
(→base) |
||
第5行: | 第5行: | ||
[[Rancher基础]] | [[Rancher基础]] | ||
#使用glusterfs做存储,不懂glusterfs的请参考其他文章 | #使用glusterfs做存储,不懂glusterfs的请参考其他文章 | ||
+ | |||
+ | ==命名空间 == | ||
+ | |||
+ | |||
+ | [https://www.jianshu.com/p/66a963a646f7 k8s实战示例1:创建命名空间和简单pod] | ||
=[[Helm3]]= | =[[Helm3]]= |
2022年9月6日 (二) 06:46的版本
目录
base
#使用glusterfs做存储,不懂glusterfs的请参考其他文章
命名空间
Helm3
k8s-imagePullPolicy拉取策略
imagePullPolicy: Always 总是拉取 pull 以前好像是这个默认值 imagePullPolicy: IfNotPresent 默认值,本地有则使用本地镜像,不拉取 imagePullPolicy: Never 只使用本地镜像,从不拉取
k8s 进入pod shell
kubectl get pods #查看所有正在运行的pod NAME READY STATUS RESTARTS AGE nginx-56b8c64cb4-t97vb 1/1 Running 3 1d kubectl exec -it nginx-56b8c64cb4-t97vb -- /bin/bash #假如当前pod只有一个容器,运行以下命令即可 root@nginx-56b8c64cb4-t97vb:/# ps -ef UID PID PPID C STIME TTY TIME CMD root 1 0 0 14:45 ? 00:00:00 nginx: master process nginx -g daemon off; 假如一个pod里有多个容器,用--container or -c 参数。例如:假如这里有个Pod名为my-pod,这个Pod有两个容器,分别名为main-app 和 helper-app,下面的命令将打开到main-app的shell的容器里。 kubectl exec -it my-pod --container main-app -- /bin/bash
k8s删除pod
1、先删除pod2、再删除对应的deployment否则只是删除pod是不管用的,还会看到pod,因为deployment.yaml文件中定义了副本数量 实例如下: 删除pod [root@test2 ~]# kubectl get pod -n jenkins NAME READY STATUS RESTARTS AGE jenkins2-8698b5449c-grbdm 1/1 Running 0 8s [root@test2 ~]# kubectl delete pod jenkins2-8698b5449c-grbdm -n jenkins pod "jenkins2-8698b5449c-grbdm" deleted 查看pod仍然存储 [root@test2 ~]# kubectl get pod -n jenkins NAME READY STATUS RESTARTS AGE jenkins2-8698b5449c-dbqqb 1/1 Running 0 8s [root@test2 ~]# 删除deployment [root@test2 ~]# kubectl get deployment -n jenkins NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE jenkins2 1 1 1 1 17h [root@test2 ~]# kubectl delete deployment jenkins2 -n jenkins 再次查看pod消失 deployment.extensions "jenkins2" deleted [root@test2 ~]# kubectl get deployment -n jenkins No resources found. [root@test2 ~]# [root@test2 ~]# kubectl get pod -n jenkins No resources found.
[ k8s中正确删除一个pod k8s中正确删除一个pod ]
coredns状态为ImagePullBackOff问题
kubectl get pods --all-namespaces kube-system coredns-7f6cbbb7b8-knhl8 0/1 ImagePullBackOff 0 24h kube-system coredns-7f6cbbb7b8-lt4k5 0/1 ImagePullBackOff 0 24h 解决办法: 确定pod所使用的镜像 kubectl get pods coredns-7f6cbbb7b8-knhl8 -n kube-system -o yaml | grep image image: registry.aliyuncs.com/google_containers/coredns:v1.8.4 imagePullPolicy: IfNotPresent - image: registry.aliyuncs.com/google_containers/coredns:v1.8.4 imageID: "" message: Back-off pulling image "registry.aliyuncs.com/google_containers/coredns:v1.8.4" root@k8s-master:~# docker pull registry.aliyuncs.com/google_containers/coredns:v1.8.4 Error response from daemon: manifest for registry.aliyuncs.com/google_containers/coredns:v1.8.4 not found: manifest unknown: manifest unknown 拉个latest 再别名 docker pull registry.aliyuncs.com/google_containers/coredns root@k8s-master:~# docker images | grep coredns registry.aliyuncs.com/google_containers/coredns latest 8d147537fb7d 3 months ago 47.6MB 强制 改一下 tag root@k8s-master:~# docker tag 8d147537fb7d registry.aliyuncs.com/google_containers/coredns:v1.8.4 root@k8s-master:~# kubectl get pods --all-namespaces | grep coredns kube-system coredns-7f6cbbb7b8-knhl8 1/1 Running 0 24h kube-system coredns-7f6cbbb7b8-lt4k5 1/1 Running 0 24h root@k8s-master:~# root@k8s-master:~# kubectl get all -n kube-system NAME READY STATUS RESTARTS AGE pod/coredns-7f6cbbb7b8-knhl8 0/1 ImagePullBackOff 0 24h pod/coredns-7f6cbbb7b8-lt4k5 0/1 ImagePullBackOff 0 24h docker pull registry.aliyuncs.com/google_containers/coredns
k8s解决coredns 的ImagePullBackOff 和ErrImagePull 问题
https://stackoverflow.com/questions/53559291/kubernetes-coredns-in-crashloopbackoff
Kubernetes CoreDNS 状态是 CrashLoopBackOff 解决思路
CrashLoopBackOff的解决方法
记录一次修复k8s pod长时间处于CrashLoopBackOff状态问题
k8s启动Pod遇到CrashLoopBackOff的解决方法
k8s常见问题
pod无故重启
OOM