Hi,
We are testing Portworx Essential on a 5 node K8s cluster.
While using dedicated (unique) storage (i.e., unmounted disks) on each worker node, the portworx installation works.
However, while using shared (common) storage (i.e., unmounted disks) attached to all the worker nodes, then the portworx installation fails.
Can someone kindly help us confirm whether a single disk attached to all the K8s worker nodes works for Portworx or not ?
Note that for KVDB, we are using separate disk ( on 3 nodes with label).
All the pods get created and READY except pods with the name “px-cluster”. While describing pod, we get following the error message.
[root@controlplane30 ~]# PX_POD=$(kubectl get pods -l name=portworx -n px -o jsonpath='{.items[4].metadata.name}'); kubectl exec $PX_POD -n px -- /opt/pwx/bin/pxctl status
Defaulted container "portworx" out of: portworx, csi-node-driver-registrar
Status: PX initialization failed. Check alerts for more details.
List of last known failures:
Type ID Resource Severity Count LastSeen FirstSeen Description
NODE ClusterManagerFailure 6a495edc-3293-420f-bc7f-ebc40b65a3db ALARM 1 Feb 7 09:29:10 UTC 2023 Feb 7 09:29:10 UTC 2023 Cluster Manager Failure on Node[10.10.19.122]: Storage failed initialization
NODE NodeInitFailure 6a495edc-3293-420f-bc7f-ebc40b65a3db ALARM 1 Feb 7 09:29:10 UTC 2023 Feb 7 09:29:10 UTC 2023 Failed in initializing drives on the node 10.10.19.122: scan found drives [/dev/sdb] from previous installation: found a drive /dev/sdb which is initialize with node id f3d47d51-d732-487d-96c9-dd9a4dc98eb1. Make sure the drives are cleaned up before reinstalling on this node.
command terminated with exit code 1
We uninstalled, cleaned the disk or used new disk and reinstalled but same error reappears!