BUG REPORT:

Kubernetes version:

$ kc version
Client Version: version.Info{Major:"1", Minor:"5", GitVersion:"v1.5.2", GitCommit:"08e099554f3c31f6e6f07b448ab3ed78d0520507", GitTreeState:"clean", BuildDate:"2017-01-12T04:57:25Z", GoVersion:"go1.7.4", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"4", GitVersion:"v1.4.5", GitCommit:"5a0a696437ad35c133c0c8493f7e9d22b0f9b81b", GitTreeState:"clean", BuildDate:"2016-10-29T01:32:42Z", GoVersion:"go1.6.3", Compiler:"gc", Platform:"linux/amd64"}

Environment:

• Cloud provider or hardware configuration: AWS
• OS:

$ cat /etc/redhat-release
CentOS Linux release 7.3.1611 (Core)

• Kernel (e.g. uname -a):

$ uname -a
Linux ip-10-0-0-150.us-west-2.compute.internal 3.10.0-514.10.2.el7.x86_64 #1 SMP Fri Mar 3 00:04:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

• Install tools:
• Others:

What happened:
From #24137 (comment)

I had a deployment with label A and I went ahead and tried to replace it by introducing a change to its yaml (basically changing it to use label B), the deployment object was replaced successfully but the old pod and the old replica set were still lingering.

After I've deleted the deployment, a replicate set was still lingering there (which was no longer connected to a deployment, i.e., an orphan ReplicaSet).

What you expected to happen:

I was expecting all pods and replica sets to be deleted after deleting the deployment.

How to reproduce it (as minimally and precisely as possible):
Basically we need a deployment, any sample deployment yaml that you can create with kubectl create -f(which will naturally create its respective ReplicaSet and Pod (s) ). The trick is to edit the yaml of the existing deployment and change its "label". Once the label is modified, run "kubectl replace -f <yaml_file>" and then check the result as follows.

Run "kubectl get pods", you will see two pods in the list, there will be a new pod with the new label and the one with the old label lingers:

$ kubectl get pod mytest-console-1569190182-b4xhs -o=json | grep -A1 -i label
        "labels": {`
            "app": "mytest-console",`
$ kubectl get pod mytest-console-3610439534-x6eaz -o=json | grep -A1 -i label
        "labels": {`
            "app": "mytest-job",`
$ kubectl get deployments | grep mytest
mytest-console              1         1         1            1           8m
$ kubectl get rs | grep mytest
mytest-console-1569190182              1         1         1         8m
mytest-console-3610439534              1         1         1         1m

Note that we have two ReplicaSets here but only one deployment.

$ kubectl delete deployment mytest-console
deployment "mytest-console" deleted
$ kubectl get rs | grep mytest
mytest-console-1569190182              1         1         1         9m

Even after deleting the deployment we can still see the replica set there, i.e., an orphan ReplicaSet.

Anything else we need to know: