This message was deleted.
# longhorn-storage
a
l
Seeing anything interesting?
So tried using a storage class with
data-locality
set to
best-effort
executing the exact same mount cmd while
ssh'ed
into the worker whereon a PVC is tried mounted and failing with the err … results in the same err exactly
a
I saw this “Configured to mount disk /dev/longhorn/pvc-97861aaf-afc9-4f24-ab76-5bb43f19a5cc as LVM2_member but current format is ext4, things might break” before err. Did you set the fstype before or after upgrading?
l
So yeah found out yesterday. That the disk is added as a member of the lvm2 vg on the nodes. By mistake a configured udev rule was Having a filter that was too wide and therefore included disks added by Longhorn when processing pvc’s. I cleaned that up and everything works now. Thank you very much for the help! Super cool. Appreciated.
👍 1