-
Bug
-
Resolution: Unresolved
-
Normal
-
4.18
-
Important
-
No
-
3
-
MCO Sprint 266, MCO Sprint 267, MCO Sprint 268
-
3
-
Rejected
-
False
-
Description of problem:
When we enable OCL in the master and the worker pool at the same time, it may happen that one of the MOSB resources are not updated with the osImage value.
Version-Release number of selected component (if applicable):
4.18.0-0.nightly-2025-01-21-015441
How reproducible:
Intermittent
Steps to Reproduce:
1. Enable OCL at the same time in worker and master pool (or with a couple of seconds between creating the MOSC resources)
Actual results:
One of the MOSB resources may not be updated with the osImage and the image will never be applied.
Expected results:
The image should be built and applied without problems
Additional info:
More information in this slack conversation: https://redhat-internal.slack.com/archives/GH7G2MANS/p1737652042188709
This scenario may look like unlikely but we need to take into account that, when we upgrade clusters with OCL enabled in worker and master pools, both pools will always start updating at the same time.
