pod Lifecycle 及 pod 异常状态处理

https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/

1、Pod phase

value for pod phase

Pending	集群已接收请求，未调度到node节点。node资源,硬盘pressure IO cpu memory
Running	1/1 running 已经ready，程序已要以pod内正常运行
Succeeded	Pod terminated from node
ContainerStatusUnknown	无法获知pod状态，失联状态。
Failed	All containers in the Pod have terminated, and at least one container has terminated in failure. That is, the container either exited with non-zero status or was terminated by the system
CrashLoopBackOff	挂掉

Pod ContainerStatusUnknown 状态

pod失连无法受管控，直接强制删除 --force
delete pods xxxx --force -n xxxx 强制删除即可
2、还有一种情况，是pod所在机器cpu打满，pod是terminating状态。实际上pod也是已经失联
查看日志如下:
 kubelet, \  error killing pod: [failed to "KillContainer" for "xxx" with KillContainerError:
 
 使用--force 强制可以删除

CrashLoopBackOff状态

将pod删除重启 ，如果未启动查看日志
kubectl delete pods xxxx -n xxx
kubectl describe pods -n xxx

Pending状态

pending状态肯定是条件不满足调度，通过查看详细日志来定位
kubectl describe pod xxx -n xxx

2、Container States

kubernetes 会追踪pod 内的container状态

kubectl get pods -n xxxx | grep xxxx

waitting	,pulling the container image from a container image registry, or applying ​​Secret​​ data
running	container entered the running state
terminated	A container in the ​​Terminated​​ state began execution and then either ran to completion or failed for some reason

3、pod conditions

pod 有很多的状态，kubelet来控制这些 pod conditions

PodScheduled	the Pod has been scheduled to a node //已被调度到某个node节点
podhashnetwork	the Pod sandbox has been successfully created and networking configured //网络OK
containersReady	all containers in the Pod are ready  //pod 中的containers ready状态
Initialized	all ​​init containers​​ have completed successfully. // init containers  初始化已完成
ready	the Pod is able to serve requests and should be added to the load balancing pools of all matching Services   //ready for requests