nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 356040

aarch64-darwin

Mentioning #349268. A newer version has been released, some of the dependencies in 17.2.0 contain known vulnerabilities.

cc @zimbatm @globin @krav @talyz @yayayayaka

nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 356040

aarch64-darwin

❌ 1 package failed to build:

Not sure why nixpkgs-review is failing to build when nix-build directly on the package.nix worked as expected. Investigating now

Thank you for your contribution!

In #359329 we update to 17.6.1, then we can also continue here. Can you please update this PR to 17.6.0 and then we should be able to merge this then :)

17.6.0 or 17.6.1?

AFAIK there is no gitlab-runner 17.6.1 release, so 17.6.0 should be fine.

Updated to version 17.6.0

nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 356040

x86_64-linux

nixpkgs-review result

Generated using nixpkgs-review.

Command: nixpkgs-review pr 356040

aarch64-darwin

@bkp5190 : Some strange test fails:

       >              Error Trace:    /build/source/helpers/homedir/homedir_test.go:44
       >                                                       /build/source/helpers/homedir/homedir_test.go:82
       >               Error:          Target error should be in err chain:
       >                                   expected: "homedir variable is not set"
       >                                in chain:
       >              Test:           TestFix/HOME_variable_is_not_set_and_homedir_value_is_empty
       > FAIL
       > FAIL gitlab.com/gitlab-org/gitlab-runner/helpers/homedir     0.005s
       > FAIL
       For full logs, run 'nix-store -l /nix/store/yg45yixbwa94mql6n0jxvvyrgzk9n13d-gitlab-runner-17.6.0.drv'.

on linux x86... do these passthru test work on your machine?

@bkp5190 : Some strange test fails:

       >              Error Trace:    /build/source/helpers/homedir/homedir_test.go:44

       >                                                       /build/source/helpers/homedir/homedir_test.go:82

       >               Error:          Target error should be in err chain:

       >                                   expected: "homedir variable is not set"

       >                                in chain:

       >              Test:           TestFix/HOME_variable_is_not_set_and_homedir_value_is_empty

       > FAIL

       > FAIL gitlab.com/gitlab-org/gitlab-runner/helpers/homedir     0.005s

       > FAIL

       For full logs, run 'nix-store -l /nix/store/yg45yixbwa94mql6n0jxvvyrgzk9n13d-gitlab-runner-17.6.0.drv'.

on linux x86... do these passthru test work on your machine?

Yes they do, the other run is attached from linux x86

@bkp5190 17.9.1 is out now with a few security fixes, so it would be good to get this PR updated to 17.9.1. If you don't have time, I'm happy to provide a diff?

@leona-ya gitlab-runner is significantly out of date (we're on 17.2 from 2024-10-22, 4+ months) - it would be great if you could re-review this, so that this package can be updated, especially given recent runner security updates in 17.9.0/17.9.1.

Last time I checked this was very broken on darwin :c I tried to fix it, but it needed like 5+ fixes. Maybe now it's right to first merge this and then fix on darwin..

I propose a switch to make as upstream does: leona-ya@9bb4e9b

@bkp5190 17.9.1 is out now with a few security fixes, so it would be good to get this PR updated to 17.9.1. If you don't have time, I'm happy to provide a diff?

@leona-ya gitlab-runner is significantly out of date (we're on 17.2 from 2024-10-22, 4+ months) - it would be great if you could re-review this, so that this package can be updated, especially given recent runner security updates in 17.9.0/17.9.1.

Yea a diff would be greatly appreciated! I have not looked at these changes in a while, thanks.

I propose a switch to make as upstream does: leona-ya@9bb4e9b

@leona-ya Thank you! I think this change makes a lot of sense! do you think it would be best to open a new PR with those changes (I can take care of this)?

perhaps if @bkp5190 has the spare time we could push that change to this branch and then update the title?

either way, I definitely think using make and mirroring more closely the upstream build approach would be good for reducing nixpkgs maintainer burden

I opened a PR with my changes + a few patches to fix build on aarch64-linux and darwin

Fixed Darwin builds in #395906.

Bumping to 17.10.1 in #396939.