Support patch with selector

[mbrancato]

What would you like to be added:
I came across a use-case where I would like to patch multiple things at once. Without looking I did:

$ kubectl patch nodes ...
error: resource(s) were provided, but no name, label selector, or --all flag specified

So seeing that error, I quickly remedied it by adding a selector as the error says I should do:

$ kubectl patch nodes --selector='mylabel' ...
Error: unknown flag: --selector

Well, interesting.

Why is this needed:
This is obviously possible in a two step process with get nodes and a loop, but it would seem nice to be able to patch multiple things using a selector.

[eddiezane]

This error is coming from the generic resource builder here and here.

I'm not sure if a conscious decision was made to not support label selectors or --all with patch. If so we might want to change that error to avoid confusion.

@mbrancato to help drill down the action here what's your use case for patching multiple nodes?

/triage needs-information

[mbrancato]

@eddiezane I wanted to remove labels on multiple nodes, but I guess this applies to anyone wanting to relabel an existing resource.

[eddiezane]

@mbrancato the kubectl label command allows you to remove labels with the syntax of foo- to remove the label foo. This works with label selectors and --all.

 ~ kubectl label nodes foo=bar --all
node/pikube-0 labeled
node/pikube-2 labeled
node/pikube-1 labeled
 ~ kubectl get nodes -l foo=bar
NAME       STATUS   ROLES    AGE    VERSION
pikube-0   Ready    master   237d   v1.18.6+k3s1
pikube-2   Ready    <none>   237d   v1.18.6+k3s1
pikube-1   Ready    <none>   237d   v1.18.6+k3s1
 ~ kubectl label nodes foo- --all
node/pikube-0 labeled
node/pikube-2 labeled
node/pikube-1 labeled
 ~ kubectl get nodes -l foo=bar
No resources found.

 ~ kubectl label nodes pikube-0 foo=bar
node/pikube-0 labeled
 ~ kubectl get nodes -l foo=bar
NAME       STATUS   ROLES    AGE    VERSION
pikube-0   Ready    master   237d   v1.18.6+k3s1
 ~ kubectl label nodes -l foo=bar foo-
node/pikube-0 labeled
 ~ kubectl get nodes -l foo=bar
No resources found.

It's tucked away as the last example of the help text. We could probably make that more clear.

Does that solve your use case?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

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/test-infra repository.

[tnozicka]

This is still needed. I run into this yesterday when I wanted to clear conditions with a patch for all instances of my CRD.

/reopen
/sig cli