【解决】There is no way to force delete Namespaces with invalid finalizers

我的问题：删除k8s命名空间时一直处于卡主状态，无法删除。

[root@k8s147 ]# kubectl delete ns kubernetes-dashboard
namespace "kubernetes-dashboard" deleted
^C
[root@k8s147 ]# kubectl delete ns kubernetes-dashboard --force --grace-period=0
warning: Immediate deletion does not wait for confirmation that the running resource has been terminated. The resource may continue to run on the cluster indefinitely.
namespace "kubernetes-dashboard" force deleted
^C
[root@k8s147 ]# kubectl get ns
NAME                   STATUS        AGE
default                Active        4h58m
kube-node-lease        Active        4h58m
kube-public            Active        4h58m
kube-system            Active        4h58m
kubernetes-dashboard   Terminating   4h49m
monitoring             Active        3h28m

出现这种情况一般都是因为删除资源时没有按照资源绑定关系，无序删除导致。

比如这里命名空间无法删除的情况，这是因为删除命名空间时，该命名空间下还存在其他资源导致。

但是我们往往一个yaml文件会定义很多资源，删除时也会同样删除这些资源，这时候就有可能会发生此问题，这个很不友好，但也只能等待k8s官方优化了。

又比如删除PV时，如果你的PV已经绑定一个PVC，这时你强制删除PV也是会发生此问题，一直处于卡主的状态，正常删除顺序应该是：先删除POD ->再删除其绑定的PVC ->然后再删除该PV。

解决办法：本地开启proxy入口，然后通过PUT请求删除指定资源。

# 终端1
[root@k8s147 ]# kubectl proxy
Starting to serve on 127.0.0.1:8001

# 终端2
[root@k8s147 ~]# kubectl get ns kubernetes-dashboard -o json | jq '.spec.finalizers=[]' | curl -X PUT http://localhost:8001/api/v1/namespaces/kubernetes-dashboard/finalize -H "Content-Type: application/json" --data @-
{
  "kind": "Namespace",
  "apiVersion": "v1",
  "metadata": {
    "name": "kubernetes-dashboard",
    "selfLink": "/api/v1/namespaces/kubernetes-dashboard/finalize",
    "uid": "34c3b499-6327-4a7a-8d8a-7648127027b9",
    "resourceVersion": "55380",
    "creationTimestamp": "2020-04-26T01:51:19Z",
    "deletionTimestamp": "2020-04-26T06:29:30Z"
  },
  "spec": {

  },
  "status": {
    "phase": "Terminating",
    "conditions": [
      {
        "type": "NamespaceDeletionDiscoveryFailure",
        "status": "True",
        "lastTransitionTime": "2020-04-26T06:29:40Z",
        "reason": "DiscoveryFailed",
        "message": "Discovery failed for some groups, 1 failing: unable to retrieve the complete list of server APIs: metrics.k8s.io/v1beta1: the server is currently unable to handle the request"
      },
      {
        "type": "NamespaceDeletionGroupVersionParsingFailure",
        "status": "False",
        "lastTransitionTime": "2020-04-26T06:29:40Z",
        "reason": "ParsedGroupVersions",
        "message": "All legacy kube types successfully parsed"
      },
      {
        "type": "NamespaceDeletionContentFailure",
        "status": "False",
        "lastTransitionTime": "2020-04-26T06:29:40Z",
        "reason": "ContentDeleted",
        "message": "All content successfully deleted, may be waiting on finalization"
      },
      {
        "type": "NamespaceContentRemaining",
        "status": "False",
        "lastTransitionTime": "2020-04-26T06:29:51Z",
        "reason": "ContentRemoved",
        "message": "All content successfully removed"
      },
      {
        "type": "NamespaceFinalizersRemaining",
        "status": "False",
        "lastTransitionTime": "2020-04-26T06:29:40Z",
        "reason": "ContentHasNoFinalizers",
        "message": "All content-preserving finalizers finished"
      }
    ]
  }
}

查看命名空间，发现之前处于Terminating状态命名空间已被删除。

[root@k8s147 ~]#kubectl get ns
NAME              STATUS   AGE
default           Active   6h14m
kube-node-lease   Active   6h14m
kube-public       Active   6h14m
kube-system       Active   6h14m
monitoring        Active   4h44m

此解决方法参考自官方问题库（https://github.com/kubernetes/kubernetes/issues/77086）：

作者：Leozhanggg

出处：https://www.cnblogs.com/leozhanggg/p/12780653.html

本文版权归作者和博客园共有，欢迎转载，但未经作者同意必须保留此段声明，且在文章页面明显位置给出原文连接，否则保留追究法律责任的权利。