[4.7] fix: check wsl npipe when executing podman compose

[vrothberg]

Does this PR introduce a user-facing change?

fix podman compose command not working with wsl

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: vrothberg

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:

• OWNERS [vrothberg]

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

[openshift-merge-robot]

PR needs rebase.

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.

[vrothberg]

@cevich CI on this branch is sick

[Luap99]

@cevich CI on this branch is sick

Looks like #20341 was merged without passing tests.

But the better question is why do we even run validate on more than one disto/arch? Shouldn't one time be enough? Validate checks based on the source code so checking more than once seems wasteful.

[vrothberg]

@baude shall we merge and get CI back later?

[TomSweeneyRedHat]

Just adding the upstream PR for reference: #20478

[TomSweeneyRedHat]

LGTM
I'd be fine with pushing the red merge button at this point while @cevich digs at the CI issue. I'll let @bbaude or @mheon make the final decision.

[cevich]

The only reason I can think of to run it across multiple platforms is to ensure it also functions if a developer runs it locally (on their random platform). Maybe we don't care. I'm fine if someone wants to strip out the aarch64 validate.

[mheon]

I think that the test definitely has value so long as we build for Macs, and we aren't planning on dropping that any time soon. Any idea what's going on - is it just this branch?

[mheon]

@Luap99 I think the multiple validates is mostly about build tags - we have a lot of things (esp. in machine) build-tagged for a single architecture or OS

[vrothberg]

Can we get this in?

[cevich]

I think the multiple validates is mostly about build tags

Aaaahhh, that sounds like a good reason. Otherwise those files would all just be ignored.

In any case, the python failure here I'm clueless about. The pre-commit (python) check has no use across multiple arches (that I'm aware of). Unfortunately this is an item that does runtime installs as well, so...well...there you go.

It's been a long while since I looked at these CI codepaths, but IIRC they're all lumped in together under a single make target :S

[mheon]

Alright, I'm just going to press the button. I would say we should investigate further, but we only have to maintain this branch for another ~3 weeks, so it's probably not worth it.

[cevich]

maintain this branch for another ~3 weeks

Oh forget it, definitely not worth it. And we don't see this happening on main so 🤷‍♂️