-
Bug
-
Resolution: Unresolved
-
Major
-
openshift-4.22
-
None
-
None
Description of problem:
The boot_counter setting logic has changed in the greenboot-rs implementation.
Version-Release number of selected component (if applicable):
4.22 on RHEL 9.8 or RHEL 10.2
How reproducible:
100% in CI in upgrade-then-fails scenarios using RHEL 9.8 or RHEL 10.2
Actual results:
Deployment is staged, machine reboot, (only if) greenboot healthcheck fails, boot_counter is set
Expected results:
Deployment is staged, boot_counter is set, machine is rebooted
Additional info:
- Make sure to revert theĀ Temporarily disable upgrade-fails-then-recovers scenarios due to greenboot behavior change commit
- Backward incompatible change in Greenboot along these lines in the code.
- New Greenboot Architecture document
- is caused by
-
USHIFT-6516 Support MicroShift V4.22 on RHEL 9.8
-
- In Progress
-
- links to