The context for this question is using longhorn via #C01GKHKAG0K - but I think it really is a longhorn question. (I'll cross post) For reasons™️ I needed to make a copy of an image backed disk and convert it to a flat disk. Basically it went from a random image backed

longhorn-image-*

storage class to the default,

harvester-longhorn

. Both disks were/are RWX and block devices. I made the copy by creating a job that mounts the PVCs and uses

dd

to make a clone. The volume finishes and, while it says it's "ready" trying to swap out the image in the VM leaves mixed results. It seems like LH is working in the background to catch up on all the writes. If I wait for a while, it seems like those writes settle and then I can move forward with the patching the VM or whatever. Even though the job and the pod have completed the Longhorn UI shows the volume associated with the pod that ran. (even if I deleted them) I'm not sure if that's part of the hold up? Is there something easy to grok from

kubectl

for the actual volume status? Wait for it to detach? Something else I'm just not aware of?