Description of problem: After deploying OpenShift 4.2.9 into Azure we have a constant stream of UpdateError warnings on the master machines: ~~~~ 17s Warning FailedUpdate machine/master-0 UpdateError: failed to update machine "master-0": found attempt to change immutable state 4m45s Warning FailedUpdate machine/master-1 UpdateError: failed to update machine "master-1": found attempt to change immutable state 2m19s Warning FailedUpdate machine/master-2 UpdateError: failed to update machine "master-2": found attempt to change immutable state ~~~~ Theser errors appear since the initial installation, nothing have changed on the machineconfig objects since running the installer. Errors in machine-controller: ~~~~ I1205 14:54:04.370797 1 controller.go:133] Reconciling Machine "master-2" I1205 14:54:04.370869 1 controller.go:304] Machine "master-2" in namespace "openshift-machine-api" doesn't specify "cluster.k8s.io/cluster-name" label, assuming nil cluster I1205 14:54:04.370885 1 actuator.go:196] Checking if machine master-2 exists I1205 14:54:04.577767 1 reconciler.go:360] Found vm for machine master-2 I1205 14:54:04.577797 1 reconciler.go:380] Machine XXXXXXXX-60c0-4ff0-XXXX-fd6209978249 is running I1205 14:54:04.577806 1 controller.go:244] Reconciling machine "master-2" triggers idempotent update I1205 14:54:04.577814 1 actuator.go:161] Updating machine master-2 I1205 14:54:04.815151 1 machine_scope.go:179] master-2: status unchanged E1205 14:54:04.831998 1 actuator.go:83] Machine error: failed to update machine "master-2": found attempt to change immutable state E1205 14:54:04.832020 1 controller.go:246] Error updating machine "openshift-machine-api/master-2": requeue in: 20s I1205 14:54:04.832036 1 controller.go:337] Actuator returned requeue-after error: requeue in: 20s ~~~~ These errors are filling the event stream with red warning events. Version-Release number of selected component (if applicable): 4.2.9 How reproducible: Have not been able to reproduce this in-house Steps to Reproduce: 1.n/a 2. 3. Actual results: Machine events are spammed with the messages 'UpdateError: failed to update machine "master-2": found attempt to change immutable state' for master machines. Expected results: These events are either not fed to the machine events or are suppressed. (If these events are 'normal' for master machines) Additional info:
https://github.com/openshift/machine-config-operator/blob/master/docs/FAQ.md#q-how-does-this-relate-to-machine-api
Verified in 4.4.0-0.nightly-2020-02-18-211831 The error is not occurring.