/lgtm

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: juanvallejo, soltysh

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Automatic merge from submit-queue (batch tested with PRs 63011, 68089, 67944, 68132). If you want to cherry-pick this change to another branch, please follow the instructions here: https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md.

@soltysh can we get some clarification here, it seems that 'kubectl run ' will remain, but the deprecation warning for 'kubectl run' seems to indicate that it's all going to be deprecated.

I think that this is because the default for kubectl run is Deployment based, not run-pod, but I think that is leading to some confusion (e.g. https://stackoverflow.com/questions/52890718/kubectl-run-is-deprecated-looking-for-alternative)

I think we should make it clear that a user can run:

kubectl run --generator=run-pod/v1 ... to get the behavior they likely want rather than looking for a new command.

@brendandburns the main gist is, we want the kubectl run to mimic docker run in k8s world. For creating resources we already have a bunch of kubectl create sub-commands. Additionally, the vast majority of different options and their influence on the actual resource produces by kubectl run is another reason we're going into simple and straightforward direction. I've replied in the SO thread. I'll also post that on my personal twitter, appreciate retweets for greater audience.

I've also opened #70288 to improve the deprecation warning.

This change breaks the example in the most recent Kubernetes Up & Running in the Service Discovery section. The following command is used:

$ kubectl run alpaca-prod \
 --image=gcr.io/kuar-demo/kuard-amd64:blue \
 --replicas=3 \
 --port=8080 \
 --labels="ver=1,app=alpaca,env=prod"

Using the --generator=run-pod/v1 flag idea like @brendanburns suggested doesn't work either:

$ kubectl run alpaca-prod --generator=run-pod/v1 --image=gcr.io/kuar-demo/kuard-amd64:blue --port=8080 --labels="app=foo,ver=1,env=prod" --replicas=3
Flag --generator has been deprecated, has no effect and will be removed in the future.
Flag --replicas has been deprecated, has no effect and will be removed in the future.
pod/foo-prod created

To continue along with the Kubernetes Up & Running example use this procedure:

1. Delete the alpaca-prod pod

   kubectl delete pod alpaca-prod
   

2. Read a blog post learn about the situation

3. Use this bash heredoc

   some_yaml=$(mktemp)
   cat <<EOF > $some_yaml
   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: alpaca-prod
     labels:
       app: alpaca
       ver: 1
       env: prod
   spec:
     replicas: 3
     selector:
       matchLabels:
         app: alpaca
     template:
       metadata:
         labels:
           app: alpaca
       spec:
         containers:
         - name: alpaca-prod
           image: gcr.io/kuar-demo/kuard-amd64:blue
           ports:
           - containerPort: 8080
   EOF
   kubectl apply -f $some_yaml
   error: unable to decode "/var/folders/_r/52tr69g54qvbxtg8fx7548tm0000gq/T/tmp.L5mckdlN": resource.metadataOnlyObject.ObjectMeta: v1.ObjectMeta.Name: Labels: ReadString: expects " or n, but found 1, error found in #10 byte of ...|d","ver":1},"name":"|..., bigger context ...|ta":{"labels":{"app":"alpaca","env":"prod","ver":1},"name":"alpaca-prod"},"spec":{"replicas":3,"sele|...
   

4. Install a command-line tool to check if the YAML is right

   brew install yj
   yj < $some_yaml
   {"apiVersion":"apps/v1","kind":"Deployment","metadata":{"name":"alpaca-prod","labels":{"app":"alpaca","ver":1,"env":"prod"}},"spec":{"replicas":3,"selector":{"matchLabels":{"app":"alpaca"}},"template":{"metadata":{"labels":{"app":"alpaca"}},"spec":{"containers":[{"name":"alpaca-prod","image":"gcr.io/kuar-demo/kuard-amd64:blue","ports":[{"containerPort":8080}]}]}}}}
   

5. Read the error message more closely and then use this YAML instead

   some_yaml=$(mktemp)
   cat <<EOF > $some_yaml
   apiVersion: apps/v1
   kind: Deployment
   metadata:
    name: alpaca-prod
    labels:
      app: alpaca
      ver: "1"
      env: prod
   spec:
    replicas: 3
    selector:
      matchLabels:
        app: alpaca
    template:
      metadata:
        labels:
          app: alpaca
      spec:
        containers:
        - name: alpaca-prod
          image: gcr.io/kuar-demo/kuard-amd64:blue
          ports:
          - containerPort: 8080
   EOF
   kubectl apply -f $some_yaml
   deployment.apps/alpaca-prod created
   

   • Labels need to be strings

6. Expose the alpaca-prod service

   kubectl expose deployment alpaca-prod
   

7. Continue with the Kubernetes Up & Running example

   kubectl expose deployment alpaca-prod
   kubectl get services -o wide
   ALPACA_POD=$(kubectl get pods -l app=alpaca \
    -o jsonpath='{.items[0].metadata.name}')
   kubectl port-forward $ALPACA_POD 48858:8080
   

8. Connect to the app

   open http://localhost:48858