Description of the problem:

If an additional local PV is exposed and listed in an "Available" state, one which is not an osd, the nodediscovery resource will be stuck  "WaitForPersistenVolumeBound" until it times out and fails.

This is a possible scenario where an end user has available PVs outside of the OSD cluster.

How reproducible:

So far seems 100%

Version:

• ocp 4.18
• quay.io/jordigilh/odf-node-recovery-controller-catalog:v1.1.0-rc.6

Steps to reproduce:

1. Have a 3 node OSD cluster
2. Add an additional disk to one of the OSD nodes - wait for it to get exposed by lso as an additional "Available" PV
3. Simulate an OSD disk failure by replacing one of the OSD disks
4. Apply node recovery

Actual results:

Noderecovery stuck (it may actually fix the osd cluster but it will get stuck)

Expected results:

Additional non-osd pv's do not affect the node recovery and it completes successfully.