Fix reconstruction of FC volumes #74023
Conversation
It should reconstruct all WWNs or WWIDs instead of just the first one. On-disk directory name format had to be changed to contain all WWNs/WWIDs.
k8s-ci-robot
added
the
release-note-none
k8s-ci-robot
added
kind/bug
k8s-ci-robot
added
size/L
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: jsafrane The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
the
approved
|
cc: @mtanino, PTAL. |
| @@ -131,20 +132,47 @@ func scsiHostRescan(io ioHandler) { | |||
| } | |||
| } | |||
|
|
|||
| // make a directory like /var/lib/kubelet/plugins/kubernetes.io/fc/target-lun-0 | |||
| // make a directory like /var/lib/kubelet/plugins/kubernetes.io/fc/target1-target2-lun-0 | |||
| func makePDNameInternal(host volume.VolumeHost, wwns []string, lun string, wwids []string) string { | |||
There was a problem hiding this comment.
How will this work for existing FC mounts? I assume this only fixes new mounts right? Existing mounts that did not store all information will still be broken?
There was a problem hiding this comment.
Yes, only the first WWN / WWID will be restored from mounts created by old kubelet. That's equal to behaviour before this PR, i.e. it won't re-introduce any regression (tested on my FC machine).
|
/retest |
|
/lgtm |
k8s-ci-robot
added
the
lgtm
Currently, Fibre Channel volume stores only the first WWN / WWID in global mount directory name. This is not enough to reconstruct a volume after kubelet restart.
This PR stores all WWNs / WWIDs in the global path name and parses it from there.
/kind bug
/sig storage
Does this PR introduce a user-facing change?:
/assign @rootfs @gnufied @tsmetana