Another bug-fix release for the 2.10 branch in preparation for a bigger 2.11 release coming soon.
On the operator side, we improved the behaviour around creating storage pools, reporting proper errors if the pools cannot be configured.
On the component side, we fixed an interesting bug in the HA Controller that could accidentally initiate fail over of resources when a snapshot is taken. LINSTOR CSI improved the behaviour around incremental snapshots shipped to S3. DRBD contains the usual round of bug fixes.
Changed
- Storage pools are only registered once their backing volume group or ZFS pool exists on the node, avoiding pools stuck in the
errorstate. - Source devices are checked for existence before a storage pool is created from them, reporting any that are missing.
- Updated images:
- DRBD 9.3.3
- LINSTOR CSI 1.11.3
- LINSTOR Affinity Controller 1.4.2
- HA Controller 1.3.3
- Latest CSI sidecars