azure - 无法使用 Azure 文件预配器将卷装载到 Kubernetes 中的 pod
问题描述
我遇到的问题是，我无法使用 Azure 云中的 Azure File CSI 将卷安装到 Kubernetes 中的 pod。
我在 pod 中收到的错误消息是
Warning FailedMount 38s kubelet Unable to attach or mount volumes: unmounted volumes=[sensu-backend-etcd], unattached volumes=[default-token-42kfh sensu-backend-etcd sensu-asset-server-ca-cert]: timed out waiting for the condition
我的存储类如下所示：
items:
- allowVolumeExpansion: true
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: |
{"allowVolumeExpansion":true,"apiVersion":"storage.k8s.io/v1","kind":"StorageClass","metadata":{"annotations":{"storageclass.kubernetes.io/is-default-class":"true"},"name":"azure-csi-standard-lrs"},"mountOptions":["dir_mode=0640","file_mode=0640","uid=0","gid=0","mfsymlinks","cache=strict","nosharesock"],"parameters":{"location":"eastus","resourceGroup":"kubernetes-resource-group","shareName":"kubernetes","skuName":"Standard_LRS","storageAccount":"kubernetesrf"},"provisioner":"kubernetes.io/azure-file","reclaimPolicy":"Delete","volumeBindingMode":"Immediate"}
storageclass.kubernetes.io/is-default-class: "true"
creationTimestamp: "2020-12-21T19:16:19Z"
managedFields:
- apiVersion: storage.k8s.io/v1
fieldsType: FieldsV1
fieldsV1:
f:allowVolumeExpansion: {}
f:metadata:
f:annotations:
.: {}
f:kubectl.kubernetes.io/last-applied-configuration: {}
f:storageclass.kubernetes.io/is-default-class: {}
f:mountOptions: {}
f:parameters:
.: {}
f:location: {}
f:resourceGroup: {}
f:shareName: {}
f:skuName: {}
f:storageAccount: {}
f:provisioner: {}
f:reclaimPolicy: {}
f:volumeBindingMode: {}
manager: kubectl-client-side-apply
operation: Update
time: "2020-12-21T19:16:19Z"
name: azure-csi-standard-lrs
resourceVersion: "15914"
selfLink: /apis/storage.k8s.io/v1/storageclasses/azure-csi-standard-lrs
uid: 3de65d08-14e7-4d0b-a6fe-39ab9a714191
mountOptions:
- dir_mode=0640
- file_mode=0640
- uid=0
- gid=0
- mfsymlinks
- cache=strict
- nosharesock
parameters:
location: eastus
resourceGroup: kubernetes-resource-group
shareName: kubernetes
skuName: Standard_LRS
storageAccount: kubernetesrf
provisioner: kubernetes.io/azure-file
reclaimPolicy: Delete
volumeBindingMode: Immediate
kind: List
metadata:
resourceVersion: ""
selfLink: ""
我的 PV 和 PVC 是绑定的：
sensu-backend-etcd 10Gi RWX Retain Bound sensu-system/sensu-backend-etcd azure-csi-standard-lrs 4m31s
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
sensu-backend-etcd Bound sensu-backend-etcd 10Gi RWX azure-csi-standard-lrs 4m47s
在 kubelet 日志中，我得到以下信息：
Dez 21 19:26:37 kubernetes-3 kubelet[34828]: E1221 19:26:37.766476 34828 pod_workers.go:191] Error syncing pod bab5a69a-f8af-43f1-a3ae-642de8daa05d ("sensu-backend-0_sensu-system(bab5a69a-f8af-43f1-a3ae-642de8daa05d)"), skipping: unmounted volumes=[sensu-backend-etcd], unattached volumes=[sensu-backend-etcd sensu-asset-server-ca-cert default-token-42kfh]: timed out waiting for the condition
Dez 21 19:26:58 kubernetes-3 kubelet[34828]: I1221 19:26:58.002474 34828 reconciler.go:224] operationExecutor.VerifyControllerAttachedVolume started for volume "sensu-backend-etcd" (UniqueName: "kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd") pod "sensu-backend-0" (UID: "bab5a69a-f8af-43f1-a3ae-642de8daa05d")
Dez 21 19:26:58 kubernetes-3 kubelet[34828]: E1221 19:26:58.006699 34828 nestedpendingoperations.go:301] Operation for "{volumeName:kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd podName: nodeName:}" failed. No retries permitted until 2020-12-21 19:29:00.006639988 +0000 UTC m=+3608.682310977 (durationBeforeRetry 2m2s). Error: "Volume not attached according to node status for volume \"sensu-backend-etcd\" (UniqueName: \"kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd\") pod \"sensu-backend-0\" (UID: \"bab5a69a-f8af-43f1-a3ae-642de8daa05d\") "
Dez 21 19:28:51 kubernetes-3 kubelet[34828]: E1221 19:28:51.768309 34828 kubelet.go:1594] Unable to attach or mount volumes for pod "sensu-backend-0_sensu-system(bab5a69a-f8af-43f1-a3ae-642de8daa05d)": unmounted volumes=[sensu-backend-etcd], unattached volumes=[sensu-backend-etcd sensu-asset-server-ca-cert default-token-42kfh]: timed out waiting for the condition; skipping pod
Dez 21 19:28:51 kubernetes-3 kubelet[34828]: E1221 19:28:51.768335 34828 pod_workers.go:191] Error syncing pod bab5a69a-f8af-43f1-a3ae-642de8daa05d ("sensu-backend-0_sensu-system(bab5a69a-f8af-43f1-a3ae-642de8daa05d)"), skipping: unmounted volumes=[sensu-backend-etcd], unattached volumes=[sensu-backend-etcd sensu-asset-server-ca-cert default-token-42kfh]: timed out waiting for the condition
Dez 21 19:29:00 kubernetes-3 kubelet[34828]: I1221 19:29:00.103881 34828 reconciler.go:224] operationExecutor.VerifyControllerAttachedVolume started for volume "sensu-backend-etcd" (UniqueName: "kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd") pod "sensu-backend-0" (UID: "bab5a69a-f8af-43f1-a3ae-642de8daa05d")
Dez 21 19:29:00 kubernetes-3 kubelet[34828]: E1221 19:29:00.108069 34828 nestedpendingoperations.go:301] Operation for "{volumeName:kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd podName: nodeName:}" failed. No retries permitted until 2020-12-21 19:31:02.108044076 +0000 UTC m=+3730.783715065 (durationBeforeRetry 2m2s). Error: "Volume not attached according to node status for volume \"sensu-backend-etcd\" (UniqueName: \"kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd\") pod \"sensu-backend-0\" (UID: \"bab5a69a-f8af-43f1-a3ae-642de8daa05d\") "
Dez 21 19:31:02 kubernetes-3 kubelet[34828]: I1221 19:31:02.169246 34828 reconciler.go:224] operationExecutor.VerifyControllerAttachedVolume started for volume "sensu-backend-etcd" (UniqueName: "kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd") pod "sensu-backend-0" (UID: "bab5a69a-f8af-43f1-a3ae-642de8daa05d")
Dez 21 19:31:02 kubernetes-3 kubelet[34828]: E1221 19:31:02.172474 34828 nestedpendingoperations.go:301] Operation for "{volumeName:kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd podName: nodeName:}" failed. No retries permitted until 2020-12-21 19:33:04.172432877 +0000 UTC m=+3852.848103766 (durationBeforeRetry 2m2s). Error: "Volume not attached according to node status for volume \"sensu-backend-etcd\" (UniqueName: \"kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd\") pod \"sensu-backend-0\" (UID: \"bab5a69a-f8af-43f1-a3ae-642de8daa05d\") "
Dez 21 19:31:09 kubernetes-3 kubelet[34828]: E1221 19:31:09.766084 34828 kubelet.go:1594] Unable to attach or mount volumes for pod "sensu-backend-0_sensu-system(bab5a69a-f8af-43f1-a3ae-642de8daa05d)": unmounted volumes=[sensu-backend-etcd], unattached volumes=[default-token-42kfh sensu-backend-etcd sensu-asset-server-ca-cert]: timed out waiting for the condition; skipping pod
在 kube-controller-manager pod 中，我得到：
E1221 20:21:34.069309 1 csi_attacher.go:500] kubernetes.io/csi: attachdetacher.WaitForDetach timeout after 2m0s [volume=sensu-backend-etcd; attachment.ID=csi-9a83de4bef35f5d01e10e3a7d598204c459cac705371256e818e3a35b4b29e4e]
E1221 20:21:34.069453 1 nestedpendingoperations.go:301] Operation for "{volumeName:kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd podName: nodeName:kubernetes-3}" failed. No retries permitted until 2020-12-21 20:21:34.569430175 +0000 UTC m=+6862.322990347 (durationBeforeRetry 500ms). Error: "AttachVolume.Attach failed for volume \"sensu-backend-etcd\" (UniqueName: \"kubernetes.io/csi/file.csi.azure.com^sensu-backend-etcd\") from node \"kubernetes-3\" : attachdetachment timeout for volume sensu-backend-etcd"
I1221 20:21:34.069757 1 event.go:291] "Event occurred" object="sensu-system/sensu-backend-0" kind="Pod" apiVersion="v1" type="Warning" reason="FailedAttachVolume" message="AttachVolume.Attach failed for volume \"sensu-backend-etcd\" : attachdetachment timeout for volume sensu-backend-etcd"
谁知道这个错误以及如何减轻它？
提前致谢。
最好的问候， rforberger
解决方案
我修好了它。
我切换到 disk.csi.azure.com 配置程序，我不得不使用卷名作为 Azure 的资源链接，例如
volumeHandle: /subscriptions/XXXXXXXXXXXXXXXXXXXXXX/resourcegroups/kubernetes-resource-group/providers/Microsoft.Compute/disks/sensu-backend-etcd
在 PV 中。
此外，我在 PV 中有一些挂载选项，这些选项不适用于 Azure 磁盘配置程序。
推荐阅读
- c++ - 是否可以针对 openmp 优化我的代码？
- submit - 在我的 APP 中修改了提交和最后一个 powerApps 的文件夹的问题
- node.js - 仅在第一次尝试时执行 MongoDB bulkWrite replaceOne 时出现不可变 _id 错误
- c# - 为什么在 C# 的这种情况下局部变量的名称会发生冲突？
- python - 无法安装 django Mac 终端
- c++ - 为什么我的代码在初始化变量时与添加像 0 这样的简单值时给出不同的答案？
- mongodb - 如何使用数据库引用加入 mongodb
- python - 是否有另一种方法可以在没有 for 循环的情况下做同样的事情？
- r - 带有 mutate 的 for 循环仅输出最后一个结果
- java - Serenity Report 未显示功能文件中提到的所有场景大纲，尽管为所有场景大纲截取了所有屏幕截图