-
Bug
-
Resolution: Done-Errata
-
Normal
-
4.15.0
-
+
-
No
-
MCO Sprint 247, MCO Sprint 248
-
2
-
False
-
-
N/A
-
Release Note Not Required
Description of problem:
When I test PR https://github.com/openshift/machine-config-operator/pull/4083, there is no machineset does not have any machine linked. $ oc get machineset/rioliu-1220c-bz2gp-worker-f -n openshift-machine-api NAME DESIRED CURRENT READY AVAILABLE AGE rioliu-1220c-bz2gp-worker-f 0 0 3h47m Many errors found in MCD log like below I1220 09:15:59.743704 1 machine_set_boot_image_controller.go:211] Error syncing machineset openshift-machine-api/rioliu-1220c-bz2gp-worker-f: failed to fetch architecture type of machineset rioliu-1220c-bz2gp-worker-f, err: could not find any machines linked to machineset, error: %!w(<nil>) the machineset patch is skipped in reconcile loop due to above error, boot image info cannot be patched even it does not have any machine provisioned.
Version-Release number of selected component (if applicable):
How reproducible:
Consistently
Steps to Reproduce:
https://github.com/openshift/machine-config-operator/pull/4083#issuecomment-1864226629
Actual results:
the machineset is skipped in reconcile loop due to above error, boot image info cannot be patched
Expected results:
the machineset should be updated even no linked machine found, because maybe it is scaled down to 0 replica
Additional info:
- is related to
-
MCO-589 Update boot images for GCP (tech preview)
- Closed
-
MCO-679 Add updated boot images support for GCP
- Closed
- links to
-
RHEA-2024:0041 OpenShift Container Platform 4.16.z bug fix update