-
Bug
-
Resolution: Done-Errata
-
Critical
-
4.14, 4.15, 4.16
-
?
-
Critical
-
No
-
MCO Sprint 248, MCO Sprint 249
-
2
-
Rejected
-
False
-
-
Release Note Not Required
-
In Progress
This is a clone of issue OCPBUGS-28237. The following is the description of the original issue:
—
This is a clone of issue OCPBUGS-27508. The following is the description of the original issue:
—
Description of problem:
The MCO logic today allows users to not reboot when changing the registries.conf file (through ICSP/IDMS/ITMS objects), but the MCO will sometimes drain the node if the change is deemed "unsafe" (deleting a mirror, for example). This behaviour is very disruptive for some customers who with to make all image registries changes non-disruptive. We will address this long term with admin defined policies via the API properly, but we would like to have a backportable solution (as a support exception) for users to do so
Version-Release number of selected component (if applicable):
4.14->4.16
How reproducible:
Steps to Reproduce:
1. 2. 3.
Actual results:
Expected results:
Additional info:
- clones
-
OCPBUGS-28237 The MCO should allow users to skip image registry change disruption
- Verified
- is blocked by
-
OCPBUGS-28237 The MCO should allow users to skip image registry change disruption
- Verified
- links to
-
RHBA-2024:0735 OpenShift Container Platform 4.14.z bug fix update