Open jkroepke opened 2 days ago
Workaround:
dnf install xfsprogs
Run mkfs.xfs -f /dev/disk/azure/scsi1/lun0
(Dangerous command, it wipes all data on that disk)
related to this issue: https://github.com/kubernetes-sigs/azuredisk-csi-driver/issues/2588, just provide me the cluster info (by email) thus I could help you downgrade csi driver version to fix this issue, thanks.
btw, xfs disk format works on azure linux 3.0 since 3.0 has 6.2 linux kernel: https://learn.microsoft.com/en-us/azure/azure-linux/how-to-enable-azure-linux-3
just provide me the cluster info (by email) thus I could help you downgrade csi driver version to fix this issue, thanks.
It's not really helping use, since the issue appear on fresh AKS cluster. Once PVCs are created, new PVCs are very rarly.
Does it help to use the older patch version of 1.30 on creation time? Is the issues fixed in 1.31 AKS?
Describe the bug We are running AKS 1.30 with with Azure Linux
2.0.20241006
.We are using a custom Storage Class with following configuration:
After a pod is started with an volume, the following error is visible via k get events
and on the
dmesg
Additional infos
It seems like https://docs.redpanda.com/current/deploy/deployment-option/self-hosted/kubernetes/aks-guide/ this error is well known. It looks like xfsutils is newer than the xfs kernel module.