Cannot create volumesnapshotclass in install.vxflexos

Question

[root@k8s-master-01 helm]# sh install.vxflexosKubernetes version v1.13.3Kubernetes master nodes: 192.168.18.10Kubernetes minion nodes: 192.168.18.11 192.168.18.12Verifying the SDC installation.Retrieved 1 mdm(s)MDM-ID 39f317742b58aa0f SDC ID 749f21df00000004 INSTALLATION ID 5082d8992e46577b IPs [0]-192.168.18.103 [1]-192.168.18.104Retrieved 1 mdm(s)MDM-ID 39f317742b58aa0f SDC ID 749f21e000000005 INSTALLATION ID 5082d8992e46577b IPs [0]-192.168.18.103 [1]-192.168.18.104Verifying the feature gates.fail: 0NAME: vxflexosLAST DEPLOYED: Sun Jun 23 17:19:59 2019NAMESPACE: vxflexosSTATUS: DEPLOYED RESOURCES:==> v1/ClusterRoleNAME AGEvxflexos-controller 4svxflexos-node 4s ==> v1/ClusterRoleBindingNAME AGEvxflexos-controller 4svxflexos-node 4s ==> v1/Pod(related)NAME READY STATUS RESTARTS AGEvxflexos-controller-0 0/4 ContainerCreating 0 4svxflexos-node-8g2zb 0/2 ContainerCreating 0 4svxflexos-node-swfsl 0/2 ContainerCreating 0 4s ==> v1/ServiceAccountNAME SECRETS AGEvxflexos-controller 1 4svxflexos-node 1 4s ==> v1/StorageClassNAME PROVISIONER AGEvxflexos (default) csi-vxflexos 4svxflexos-xfs csi-vxflexos 4s ==> v1alpha1/CSIDriverNAME AGEvxflexos 4s ==> v1beta1/StatefulSetNAME READY AGEvxflexos-controller 0/1 4s ==> v1beta2/DaemonSetNAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGEvxflexos-node 2 2 0 2 0 4s NAME READY STATUS RESTARTS AGEvxflexos-controller-0 0/4 ContainerCreating 0 10svxflexos-node-8g2zb 0/2 ContainerCreating 0 10svxflexos-node-swfsl 0/2 ContainerCreating 0 10sNAME READY STATUS RESTARTS AGEvxflexos-controller-0 0/4 ContainerCreating 0 15svxflexos-node-8g2zb 0/2 ContainerCreating 0 15svxflexos-node-swfsl 0/2 ContainerCreating 0 15sNAME READY STATUS RESTARTS AGEvxflexos-controller-0 0/4 ContainerCreating 0 20svxflexos-node-8g2zb 0/2 ContainerCreating 0 20svxflexos-node-swfsl 0/2 ContainerCreating 0 20sCSIDrivers:NAME AGEvxflexos 20sCSINodeInfos:No resources found.StorageClasses:NAME PROVISIONER AGEvxflexos (default) csi-vxflexos 20svxflexos-xfs csi-vxflexos 20serror: the server doesn't have a resource type 'volumesnapshotclass'installing volumesnapshotclasserror: unable to recognize 'volumesnapshotclass.yaml': no matches for kind 'VolumeSnapshotClass' in version 'snapshot.storage.k8s.io/v1alpha1'VolumeSnapshotClasses:error: the server doesn't have a resource type 'volumesnapshotclass'

tssgery · Accepted Answer

It would be good to get the logs from the controller, via:   kubectl logs -n vxflexos vxflexos-controller-0 driver

ZhangFan4 · Answer

Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 23m default-scheduler Successfully assigned vxflexos/vxflexos-controller-0 to k8s-node-02 Normal Pulled 23m kubelet, k8s-node-02 Container image 'quay.azk8s.cn/k8scsi/csi-attacher:v1.0.1' already present on machine Normal Created 23m kubelet, k8s-node-02 Created container Normal Started 23m kubelet, k8s-node-02 Started container Normal Pulled 23m kubelet, k8s-node-02 Container image 'quay.azk8s.cn/k8scsi/csi-provisioner:v1.0.1' already present on machine Normal Created 23m kubelet, k8s-node-02 Created container Normal Started 23m kubelet, k8s-node-02 Started container Normal Pulling 23m kubelet, k8s-node-02 pulling image 'dockerhub.azk8s.cn/dellemc/csi-vxflexos:v1.1.0' Normal Created 23m kubelet, k8s-node-02 Created container Normal Started 23m kubelet, k8s-node-02 Started container Normal Pulled 23m kubelet, k8s-node-02 Successfully pulled image 'dockerhub.azk8s.cn/dellemc/csi-vxflexos:v1.1.0' Warning BackOff 23m (x3 over 23m) kubelet, k8s-node-02 Back-off restarting failed container Normal Pulling 22m (x3 over 23m) kubelet, k8s-node-02 pulling image 'quay.azk8s.cn/k8scsi/csi-snapshotter:v1.0.1' Normal Created 22m (x3 over 23m) kubelet, k8s-node-02 Created container Normal Started 22m (x3 over 23m) kubelet, k8s-node-02 Started container Normal Pulled 22m (x3 over 23m) kubelet, k8s-node-02 Successfully pulled image 'quay.azk8s.cn/k8scsi/csi-snapshotter:v1.0.1' Warning BackOff 13m (x18 over 21m) kubelet, k8s-node-02 Back-off restarting failed container Warning BackOff 3m6s (x80 over 21m) kubelet, k8s-node-02 Back-off restarting failed container

ZhangFan4 · Answer

after docker search k8scsi, it created volumesnapshotsclass, but a new issue:

Nodes are ready, but controller not work

[root@k8s-master-01 helm]# kubectl get pods -n vxflexos
NAME READY STATUS RESTARTS AGE
vxflexos-controller-0 2/4 CrashLoopBackOff 22 20m
vxflexos-node-5qfbl 2/2 Running 0 20m
vxflexos-node-c7gtx 2/2 Running 0 20m
vxflexos-node-tt8rx 2/2 Running 0 20m

ZhangFan4 · Answer

Thanks tssgery! I found have to wait a long time to download image, and get it work now.

tssgery · Answer

Back to the original problem of not being able to create the volumesnapshotclass...

The workaround, if you see this, is to re-run the install.vxflexos script. The installation is idempotent and will pick up where left off and create the class.

This will be fixed in a future release of the CSI Driver for VxFlex OS

Containers

Was this post helpful?