[release-1.4] Bump golangci-lint to v2

[ivanvc]

Manual backport of #954 / 675a3be.

Silence linters that fail with the current state of the code, as we don't want to change the stable release code.

Trying to unblock #1157 / The Go version bump (etcd-io/etcd#21458).

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: ivanvc
Once this PR has been reviewed and has the lgtm label, please assign spzala for approval. For more information see the Code Review Process.

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

Details

Needs approval from an approver in each of these files:

• OWNERS

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

[ivanvc]

Should we keep linting Windows files? I'd need to disable the formatter gofmt to make them pass. An alternative would be to not lint in Windows. WDYT, @ahrtr?

[ahrtr]

I'd need to disable the formatter gofmt to make them pass

Can we fix the linter error?

[ivanvc]

I don't think so, it's an obscure error that we found before, which Windows complains of every file's first line:

https://github.com/etcd-io/bbolt/actions/runs/23867689295/job/69591112978?pr=1178#step:7:185

An alternative approach would be to use a separate configuration file for Windows that disables gofmt.

I also don't see much value in linting every file on Linux or Windows. In my opinion, we should have a single Lint CI workflow rather than appending the step to every workflow.

[Elbehery]

I don't think so, it's an obscure error that we found before, which Windows complains of every file's first line:

+1, I hit the same issue last year iirc

I recall I have tried using V2 to resolve it, but then hit the go-version used

we should have a single Lint CI workflow rather than appending the step to every workflow.

+1, good refactoring 👍🏽

use a separate configuration file for Windows that disables gofmt

I would keep this as the last resort, and we could skip it without a separate file imho

[ahrtr]

I don't think so, it's an obscure error that we found before, which Windows complains of every file's first line

Two questions:

• what's the blocker to fix the linter error?
• why there isn't such issue on main branch?

we should have a single Lint CI workflow rather than appending the step to every workflow.

Sounds good to me.

[ivanvc]

Two questions:

* what's the blocker to fix the linter error?

* why there isn't such issue on main branch?

These are good questions.

We have been trying to address this issue several times (i.e., #1073, #994). We don't know why it's not happening on the main branch. We thought it needed a newer Go version, but after the bump, it was still failing.

The problem now is that we need to bump golangci-lint to bump Go. So, that's why I suggest disabling linting on Windows. In the end, the only Windows file that the Linux linter is not linting is bolt_windows.go.

I would like to get this going, to not keep blocking the Go version bump.

we should have a single Lint CI workflow rather than appending the step to every workflow.

Sounds good to me.

I'll open a PR targeting the main branch to open the discussion.

[ahrtr]

We don't know why it's not happening on the main branch. We thought it needed a newer Go version, but after the bump, it was still failing.

OK, we don't have to spend too much time on this for stable releases. Pls feel free to workaround it.