Closed sgerme closed 2 years ago
What version of mayastor are you running?
I suspect you've hit a pool operator bug, which is causing the pool to simply appear as "unknown".
Could you please run kubectl-mayastor get pools
and check if the pool is also "unkown" there?
I am running on ARM and never succeed to run mayastor binaries :( I'll try on another one.
zsh: exec format error: kubectl-mayastor
Oh, we don't have a published arm version. You can try building it yourself from the control-plane repo from the 1 release branch, if that's what you're running:
nix-build -A utils.release.linux-musl.kubectl-plugin
@sgerme, to recreate the pool I think you can delete the CR and recreate it. If delete is stuck you may remove the finalizer - this is safe, we don't delete any replicas on CR delete, only the pool if it has no replicas known to the control-plane.
I have finally deleted everything and recreate the pools. I've downloaded backups. Thanks anyway :)
Is your feature request related to a problem? Please describe. Hi, I have a cluster running Talos 0.14 on bare metal using NVMe SSDs. After a crash of one of the nodes, I re-deployed it with the same name. but the pool is not recognized anymore.
Describe the solution you'd like a way to start a normal Linux, to recover all the data. When I try to mount the disk, it is not mountable since it does not appear as a valid format.
Describe alternatives you've considered Try to recreate the pool. But this pool appears as status "unknown"
Additional context two bare metal servers, with two NVMe SSD, one for the talos system, another dedicated to mayastor. I try to recover data from this second drive.
thanks for helping !