Ignore not found in kubectl wait --for=delete

[zhouya0]

What type of PR is this?
/kind feature

What this PR does / why we need it:
If the condition of kubectl wait if delete， then we should ignore 404 not found errors.

Before:

[root@master ~]# kubectl wait jobs job1 --for=delete
Error from server (NotFound): jobs.batch "job1" not found
[root@master ~]# echo $?
1

With this PR:

[root@master ~]# ./kubectl wait jobs job1 --for=delete
[root@master ~]# echo $?
0

Which issue(s) this PR fixes:

Fixes ##87747

Special notes for your reviewer:
/cc @soltysh

Does this PR introduce a user-facing change?:

Add kubectl wait  --ignore-not-found flag

[zhouya0]

/assign @seans3

[zhouya0]

/retest

[zhouya0]

/retest

[zhouya0]

/retest

[zhouya0]

/cc @brianpursley Can you help me to review this? Thanks!

[brianpursley]

Are we sure we want to add a new flag for this, instead of just changing the behavior of waiting for delete?

It seems like it is already trying to handle not found here, but I wonder if it needs more. I don't know off-hand if List() returns a 404 if the list is empty, so this might not actually be doing what was intended.

What about in func IsDeleted, just adding another if apierrors.IsNotFound(err) condition like this? Would this work?

objWatch, err := o.DynamicClient.Resource(info.Mapping.Resource).Namespace(info.Namespace).Watch(context.TODO(), watchOptions)
if apierrors.IsNotFound(err) {   # <-- Added this
	return info.Object, true, nil
}
if err != nil {
	return gottenObj, false, err
}

I'm not entirely opposed to having a command-line option, but I'm trying to imagine when I would not want to set this when I'm waiting for a deletion, or a case where I would want to set this when I'm waiting for a condition.

[brianpursley]

/cc @soltysh
/cc @seans3

Thoughts on adding an --ignore-not-found flag vs changing the behavior of waiting for a delete to always ignore not found errors? (See my previous comment)

[zhouya0]

@brianpursley I tried your suggestion and apparently it doesn't work because the Notfound error happens before. In visitor := o.ResourceFinder.Do() in RunWait and it will call

kubernetes/staging/src/k8s.io/cli-runtime/pkg/resource/builder.go

Line 792 in a8e4303

func (b *Builder) visitorResult() *Result {

It's will firstly send the request to get the info and that's where error happens :)

[soltysh]

I'm also leaning towards what Brian said, since we already ignore not found, I'd make it explicit there as well.

[zhouya0]

/retest

[zhouya0]

/retest

[soltysh]

Oh, and please squash your changes.

[soltysh]

/lgtm
/approve
/milestone v1.20
/priority important-longterm

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: soltysh, zhouya0

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

The pull request process is described here

Needs approval from an approver in each of these files:

• staging/src/k8s.io/kubectl/OWNERS [soltysh]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[soltysh]

Changing kind per linked issue.