K8s 错误 2 副本集正在为来自 github 操作的每个新部署创建

Question

我在 Helm 管理的 EKS 中看到了这个错误，其中为来自 github 操作的每个新部署生成了两个副本集，第二个副本集 pod 在终止之前存在 1000 秒。因为我们使用滚动更新作为我们的部署策略，所以 replicaset 仍然可以运行。 我还在所有微服务中发现了相同的行为。 我尝试使用 deoployment.yaml 文件在本地重现此问题，也尝试使用 helm chart 但没有发现任何错误。

kubectl describe pods eun-6f58xxxxdb-28ddd -n staging
Name:             eun-6f58xxxxdb-28ddd
Namespace:        staging
Priority:         0
Service Account:  default
Node:             ip-10-40.us-east-2.compute.internal/10.40
Start Time:       Tue, 21 Feb 2023 12:59:59 +0530
Labels:           app.kubernetes.io/component=loud
                  app.kubernetes.io/instance=eun
                  app.kubernetes.io/managed-by=Helm
                  app.kubernetes.io/name=eun
                  app.kubernetes.io/version=PR-xxx-e13xx1a
                  helm.sh/chart=cloud-0.0.47
                  pod-template-hash=6f58xxxxdb
Annotations:      kubernetes.io/psp: eks.privileged
Status:           Running
IP:               10.xx.xx.214
IPs:
  IP:           10.xx.xx.214
Controlled By:  ReplicaSet/eun-6f58xxxxdb
Containers:
  eun:
    Container ID:   docker://e3fed29f82836b7b4b2
    Image:          940xxxx.dkr.ecr.us-east-2.amazonaws.com/eun:PR-xxx-e13xx1a
    Image ID:       docker-pullable://940xxxx.dkr.ecr.us-west-2.amazonaws.com/eun@sha256:bff318
    Port:           10001/TCP
    Host Port:      0/TCP
    State:          Running
      Started:      Tue, 21 Feb 2023 13:00:07 +0530
    Ready:          True
    Restart Count:  0
    Limits:
      cpu:     600m
      memory:  628Mi
    Requests:
      cpu:      200m
      memory:   248Mi
    Liveness:   tcp-socket :10001 delay=20s timeout=1s period=60s #success=1 #failure=3
    Readiness:  http-get http://:10001/ delay=20s timeout=60s period=60s #success=1 #failure=3
    Environment Variables from:
      eunomia  Secret  Optional: false
    Environment:
      DD_ENTITY_ID:    (v1:metadata.uid)
      DD_AGENT_HOST:   (v1:status.hostIP)
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-qdssj (ro)
Readiness Gates:
  Type                                                         Status
  target-health.elbv2.k8s.aws/k8s-staging-eunomia-eddd6c8379   True 
Conditions:
  Type                                                         Status
  target-health.elbv2.k8s.aws/k8s-staging-eunomia-eddd6c8379   True 
  Initialized                                                  True 
  Ready                                                        True 
  ContainersReady                                              True 
  PodScheduled                                                 True 
Volumes:
  kube-api-access-qdssj:
    Type:                     Projected (a volume that contains injected data from multiple sources)
    TokenExpirationSeconds:   3607
    ConfigMapName:            kube-root-ca.crt
    ConfigMapOptional:        <nil>
    DownwardAPI:              true
QoS Class:                    Burstable
Node-Selectors:               <none>
Tolerations:                  node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                              node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Topology Spread Constraints:  kubernetes.io/hostname:ScheduleAnyway when max skew 1 is exceeded
Events:                       <none>

kubectl describe deployment enu -n staging
Name:                   enu
Namespace:              staging
CreationTimestamp:      Fri, 17 Dec 2021 20:14:58 +0530
Labels:                 app.kubernetes.io/component=cloud
                        app.kubernetes.io/instance=enu
                        app.kubernetes.io/managed-by=Helm
                        app.kubernetes.io/name=enu
                        app.kubernetes.io/version=PR-xxx-e13xx1a
                        argocd.argoproj.io/instance=stg-enu
                        helm.sh/chart=cloud-0.0.47
Annotations:            deployment.kubernetes.io/revision: 104
Selector:               app.kubernetes.io/component=cloud,app.kubernetes.io/instance=enu,app.kubernetes.io/managed-by=Helm,app.kubernetes.io/name=enu
Replicas:               1 desired | 1 updated | 1 total | 1 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app.kubernetes.io/component=cloud
           app.kubernetes.io/instance=enu
           app.kubernetes.io/managed-by=Helm
           app.kubernetes.io/name=enu
           app.kubernetes.io/version=PR-xxx-e13xx1a
           helm.sh/chart=cloud-0.0.47
  Containers:
   enu:
    Image:      940xxxx.dkr.ecr.us-east-2.amazonaws.com/enu:PR-xxx-e13xx1a
    Port:       10001/TCP
    Host Port:  0/TCP
    Limits:
      cpu:     600m
      memory:  628Mi
    Requests:
      cpu:      200m
      memory:   248Mi
    Liveness:   tcp-socket :10001 delay=20s timeout=1s period=60s #success=1 #failure=3
    Readiness:  http-get http://:10001/ delay=20s timeout=60s period=60s #success=1 #failure=3
    Environment Variables from:
      enu                   Secret  Optional: false
    Environment:                <none>
    Mounts:                     <none>
  Volumes:                      <none>
  Topology Spread Constraints:  kubernetes.io/hostname:ScheduleAnyway when max skew 1 is exceeded
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   enu-6f5897d6db (1/1 replicas created)
Events:          <none>