Old ReplicaSets are not being removed.

[axelluguercio]

I am noticing that when I start a rollout, the analysis ends successfully, the canary pod becomes stable, but the pod from the previous revision is still running.

Version: v.1.6.6

Rollout spec:

apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
annotations:
meta.helm.sh/release-name: xxxxxx
meta.helm.sh/release-namespace: xxxxxx
rollout.argoproj.io/revision: '2'
rollout.argoproj.io/workload-generation: '3'
creationTimestamp: '2024-05-10T19:15:26Z'
generation: 2
labels:
app.kubernetes.io/managed-by: Helm
chart: canary
release: xxxxxx
status:
HPAReplicas: 2
availableReplicas: 2
blueGreen: {}
canary:
weights:
canary:
podTemplateHash: 7fcb899664
weight: 0
stable:
podTemplateHash: 7fcb899664
weight: 100
conditions:
- lastTransitionTime: '2024-05-10T19:15:33Z'
lastUpdateTime: '2024-05-10T19:15:33Z'
message: Rollout has minimum availability
reason: AvailableReason
status: 'True'
type: Available
- lastTransitionTime: '2024-05-10T19:20:29Z'
lastUpdateTime: '2024-05-10T19:20:29Z'
message: Rollout is not healthy
reason: RolloutHealthy
status: 'False'
type: Healthy
- lastTransitionTime: '2024-05-10T19:26:34Z'
lastUpdateTime: '2024-05-10T19:26:34Z'
message: Rollout is paused
reason: RolloutPaused
status: 'False'
type: Paused
- lastTransitionTime: '2024-05-10T19:26:34Z'
lastUpdateTime: '2024-05-10T19:26:34Z'
message: RolloutCompleted
reason: RolloutCompleted
status: 'True'
type: Completed
- lastTransitionTime: '2024-05-10T19:36:34Z'
lastUpdateTime: '2024-05-10T19:36:34Z'
message: >-
ReplicaSet "deployment-xxxxx-7fcb899664" has timed out
progressing.
reason: ProgressDeadlineExceeded
status: 'False'
type: Progressing
currentPodHash: 7fcb899664
currentStepHash: 5cfd46867b
currentStepIndex: 9
message: >-
ProgressDeadlineExceeded: ReplicaSet
"deployment-xxxx-7fcb899664" has timed out progressing.
observedGeneration: '2'
phase: Degraded
readyReplicas: 2
replicas: 2
selector: app=xxxxxx
stableRS: 7fcb899664
updatedReplicas: 1
workloadObservedGeneration: '3'
spec:
analysis:
successfulRunHistoryLimit: 5
unsuccessfulRunHistoryLimit: 5
replicas: 1
revisionHistoryLimit: 5
rollbackWindow:
revisions: 5
selector:
matchLabels:
app: xxxxxxx
strategy:
canary:
analysis:
analysisRunMetadata: {}
args:
- name: service
value: xxxxx
- name: interval
value: 6m
templates:
- clusterScope: true
templateName: error-rate
- clusterScope: true
templateName: latency
canaryMetadata:
labels:
role: canary
version: canary
scaleDownDelaySeconds: 0
stableMetadata:
labels:
role: stable
version: stable
steps:
- setWeight: 20
- pause:
duration: 1m
- setWeight: 40
- pause:
duration: 1m
- setWeight: 60
- pause:
duration: 2m
- setWeight: 80
- pause:
duration: 2m
- setWeight: 100
trafficRouting:
istio:
destinationRule:
canarySubsetName: canary
name: destinationrule-xxxx
stableSubsetName: stable
virtualService:
name: virtualservice-xxxxx
workloadRef:
apiVersion: apps/v1
kind: Deployment
name: deployment-xxxxx
scaleDown: onsuccess

Name: deployment-xxxxxxxxx
Namespace: xxxxxxxxxxx
Status: ✖ Degraded
Message: ProgressDeadlineExceeded: ReplicaSet "deployment-xxxxxxxxx-7fcb899664" has timed out progressing.
Strategy: Canary
Step: 9/9
SetWeight: 100
ActualWeight: 100
Images: xxxxxxxxx/xxxxxxxx/xxxxxxxx:1.0.0-2024.05.10-12.49.09 (stable)
xxxxxxxxx/xxxxxxxxx/xxxxxxxxxx:1.0.0-2024.05.10-14.04.44
Replicas:
Desired: 1
Current: 2
Updated: 1
Ready: 2
Available: 2

NAME KIND STATUS AGE INFO
⟳ deployment-xxxxxxx Rollout ✖ Degraded 62m
├──# revision:2
│ ├──⧉ deployment-xxxxxx-7fcb899664 ReplicaSet ✔ Healthy 57m stable
│ │ └──□ deployment-xxxxx-7fcb899664-5hsdq Pod ✔ Running 57m ready:2/2
│ └──α deployment-xxxxxxxx-7fcb899664-2 AnalysisRun ✔ Successful 57m ✔ 14
└──# revision:1
└──⧉ deployment-xxxxx-6d86c795f6 ReplicaSet ✔ Healthy 62m
└──□ deployment-bbe-xxxxxx-6d86c795f6-m2ffl Pod ✔ Running 62m ready:2/2

Please any light on this?

[kostis-codefresh]

Can you post the original yaml with correct formatting please? Not the live object from the Kubernetes cluster.