-
Bug
-
Resolution: Won't Do
-
Normal
-
None
-
4.14
-
?
-
Important
-
No
-
ShiftStack Sprint 242, ShiftStack Sprint 243, ShiftStack Sprint 244
-
3
-
Rejected
-
False
-
-
N/A
-
Release Note Not Required
Description of problem:
When updating a non-existent subnet in CPMS a new machine is created and is stuck in Provisioning phase instead of moving to Failed state
Version-Release number of selected component (if applicable):
4.14.0-0.nightly-2023-07-31-181848
How reproducible:
Always
Steps to Reproduce:
1. Update subnet name $ oc edit controlplanemachineset/cluster -n openshift-machine-api spec.template.spec.machines_v1beta1_machine_openshift_io.spec.network.subnets.name 2. Verify a subnet can't be found e.g. $ oc logs machine-api-controllers-7f87b4cc58-r4l79 -n openshift-machine-api -c machine-controller E0801 14:39:15.883801 1 controller.go:329] "msg"="Reconciler error" "error"="filter matched no resources" "controller"="machine-controller" "name"="ostest-7zj5z-master-r7zm5-0" "namespace"="openshift-machine-api" "object"={"name":"ostest-7zj5z-master-r7zm5-0","namespace":"openshift-machine-api"} "reconcileID"="5cb99bab-96b0-4ef8-8842-0735373e2ba4" 3.Verify a new machine is created and is stuck in Provisioning state
Actual results:
Expected results:
Additional info:
- clones
-
OCPBUGS-17153 A machine phase should be failed when updating CPMS with non-existent subnet or non-existent availabilityZone or non-existent volume type
- Closed