Spinnaker: Scale (Manifest) stage timeout when scale replicaSet to 0 replicas

[vmagamedov]

Issue Summary:

This is the same issue as #3148, where only deployments were fixed in https://github.com/spinnaker/clouddriver/pull/2848. Please make the same fix also for a replicaSet kind.

I assume that this function needs a fix: https://github.com/spinnaker/clouddriver/blob/b73c5f1/clouddriver-kubernetes/src/main/groovy/com/netflix/spinnaker/clouddriver/kubernetes/v2/op/handler/KubernetesReplicaSetHandler.java#L97

Cloud Provider(s):

Kubernetes

Environment:

Kubernetes

Feature Area:

Scale (Manifest) stage, replicaSet kind

Additional Details:

Error details:

Exception ( Wait For Manifest To Stabilize )

WaitForManifestStableTask of stage Down API Canary timed out after 30 minutes 1 second. pausedDuration: 0 seconds, elapsedTime: 30 minutes 1 second,timeoutValue: 30 minutes

Scale (Manifest) stage succeeds and replicas count is equal to 0, but this stage waits for a 30 minutes and then fails the whole pipeline.

[ezimanyi]

FYI @lwander

[lwander]

@spinnakerbot assign-issue lwander

[vmagamedov]

Updated image in the spin-clouddriver deployment from 4.2.0-20181220153244 to 4.3.0-20190104022809 and can confirm that pipeline now works. Thanks!