When Deployment is editing replicas and strategy simultaneously, it may get stuck and not continue to execute

[gukc007]

What happened?

When Deployment is editing replicas and strategy simultaneously, it may get stuck and not continue to execute

What did you expect to happen?

deployment continue upgrade or scaled

How can we reproduce it (as minimally and precisely as possible)?

1. created a deployment with a rolling upgrade and set a large number of replicas, such as 100.
2. After creating 100 pods, use a slow rolling upgrade like RollingUpdateStrategy: 1 max unavailable, 1 max surge
   to make two active replicasets appear at the same time.
3. While in the process of rolling upgrade (2 replicaSet), simultaneously edit the upgrade strategy of the deployment to recreate, and modify the number of replicas to 101. (kubectl apply -f xxx.yaml)
4. At this time, the deployment will be in a stuck state, and it will not be upgraded or scaled no matter how it is modified.

Anything else we need to know?

Deployment will get stuck because of these few lines of code.

https://github.com/kubernetes/kubernetes/blob/64355780d9d945a1135a004829ab9dcc38c3911d/pkg/controller/deployment/deployment_controller.go#L665C2-L671C3

Kubernetes version

v1.23.3

Cloud provider

OS version

Install tools

Container runtime (CRI) and version (if applicable)

Related plugins (CNI, CSI, ...) and versions (if applicable)

[k8s-ci-robot]

This issue is currently awaiting triage.

If a SIG or subproject determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[gukc007]

I don't know if this is a bug or a feature.

[mengjiao-liu]

/sig apps