POD with Trident SAN or NAS RWO PV goes offline when moving to another node
Applies to
- Openshift/Kubernetes
- SAN/iSCSI PVs
- NAS RWO PVs
- Trident
Issue
The following error occurs in a replacement pod, when the original pod using the PV fails or is deleted on its node and Kuberenetes is attempting to start it up on one of the other nodes, while the pod is using an RWO based PVC, such as those using an iSCSI
ontap-san
backend, but also NFS RWO type PV:Multi-Attach error for volume "pvc-<UUID>" Volume is already exclusively attached to one node and can't be attached to another
However, after a while the only error left is:
Warning FailedMount <time_delay> kubelet MountVolume.MountDevice failed for volume "pvc-<UUID>" : rpc error: code = DeadlineExceeded desc = context deadline exceeded