driver name org.gluster.glusterfs not found in the list of registered CSI drivers

[muecahit94]

I can create a new PVC, it works without any issue:

kubectl describe pvc test2 -n test
Name:          test2
Namespace:     test
StorageClass:  glusterfs-csi
Status:        Bound
Volume:        pvc-46e094b7-2ddf-11e9-a318-002590927952
Labels:        cattle.io/creator=norman
Annotations:   field.cattle.io/creatorId=u-jtvds
               pv.kubernetes.io/bind-completed=yes
               pv.kubernetes.io/bound-by-controller=yes
               volume.beta.kubernetes.io/storage-provisioner=org.gluster.glusterfs
Capacity:      1Gi
Access Modes:  RWX
Events:        <none>

But when i want to use it on a container i get this issue:

Events:
  Type     Reason       Age                From                  Message
  ----     ------       ----               ----                  -------
  Warning  FailedMount  14m (x27 over 1h)  kubelet, myhostname  Unable to mount volumes for pod "test-56b7889987-wklxs_test(ec1d4cd0-2de6-11e9-877e-00259092799a)": timeout expired waiting for volumes to attach or mount for pod "test"/"test-56b7889987-wklxs". list of unmounted volumes=[test]. list of unattached volumes=[test default-token-6k4fj]
  Warning  FailedMount  3m (x43 over 1h)   kubelet, myhostname  MountVolume.MountDevice failed for volume "pvc-46e094b7-2ddf-11e9-a318-002590927952" : driver name org.gluster.glusterfs not found in the list of registered CSI drivers

I use the kubernetes cluster under rancher. Kubernets version is v1.12.3

I hope someone can help me.

[Madhu-1]

@muecahit94 current GCS support kube 1.13+ version

to deploy in kube 1.12 may be you need to update driver register path in gcs deployment from var/lib/kubelet/plugins_registry to var/lib/kubelet/plugins in https://github.com/gluster/gcs/blob/master/deploy/templates/gcs-manifests/gcs-fs-csi.yml.j2

[muecahit94]

there is plugins_registry at three places:

cat  templates/gcs-manifests/gcs-fs-csi.yml.j2 | grep var/lib/kubelet/plugins_registry
              value: /var/lib/kubelet/plugins_registry/org.gluster.glusterfs/csi.sock
            path: /var/lib/kubelet/plugins_registry/org.gluster.glusterfs
            path: /var/lib/kubelet/plugins_registry/

when I change this part:

        - name: registration-dir
          hostPath:
            path: /var/lib/kubelet/plugins/
            type: Directory

I got this error on csi-nodeplugin:

I0212 17:14:04.271438 1 main.go:111] Version: v1.0.1-0-g27703026
I0212 17:14:04.271552 1 main.go:118] Attempting to open a gRPC connection with: "/plugin/csi.sock"
I0212 17:14:04.271570 1 connection.go:69] Connecting to /plugin/csi.sock
I0212 17:14:04.272055 1 connection.go:96] Still trying, connection is CONNECTING
I0212 17:14:04.272383 1 connection.go:93] Connected
I0212 17:14:04.272401 1 main.go:126] Calling CSI driver to discover driver name.
I0212 17:14:04.272421 1 connection.go:137] GRPC call: /csi.v1.Identity/GetPluginInfo
I0212 17:14:04.272430 1 connection.go:138] GRPC request: {}
I0212 17:14:04.276837 1 connection.go:140] GRPC response: {"name":"org.gluster.glusterfs","vendor_version":"1.0.0"}
I0212 17:14:04.277597 1 connection.go:141] GRPC error: <nil>
I0212 17:14:04.277608 1 main.go:134] CSI driver name: "org.gluster.glusterfs"
I0212 17:14:04.277630 1 main.go:138] Loading kubeconfig.
I0212 17:14:04.277943 1 node_register.go:55] Calling CSI driver to discover node ID.
I0212 17:14:04.277976 1 connection.go:137] GRPC call: /csi.v1.Node/NodeGetInfo
I0212 17:14:04.277984 1 connection.go:138] GRPC request: {}
I0212 17:14:04.279040 1 connection.go:140] GRPC response: {"node_id":"hostname"}
I0212 17:14:04.279609 1 connection.go:141] GRPC error: <nil>
I0212 17:14:04.279616 1 node_register.go:63] CSI driver node ID: "hostname"
I0212 17:14:04.279696 1 node_register.go:86] Starting Registration Server at: /registration/org.gluster.glusterfs-reg.sock
I0212 17:14:04.279778 1 node_register.go:93] Registration Server started at: /registration/org.gluster.glusterfs-reg.sock
I0212 17:14:04.280374 1 main.go:84] Received GetInfo call: &InfoRequest{}
I0212 17:14:04.282779 1 main.go:94] Received NotifyRegistrationStatus call: &RegistrationStatus{PluginRegistered:false,Error:plugin registration failed with err: error updating CSI node info in the cluster: error adding CSI driver node info: driverNodeID must not be empty,}
E0212 17:14:04.282897 1 main.go:96] Registration process failed with error: plugin registration failed with err: error updating CSI node info in the cluster: error adding CSI driver node info: driverNodeID must not be empty, restarting registration container.

[muecahit94]

is there no way to deploy gcs on a kubernetes version 1.12.x ?

Pls give me some tips, what i could try out.