Re: Heads Up: cirrus-ci is shutting down June 1st

Hi,

Thank you for looking into this!

On Thu, 28 May 2026 at 23:50, Jacob Champion
<jacob(dot)champion(at)enterprisedb(dot)com> wrote:
>
> On Thu, May 28, 2026 at 10:06 AM Nazir Bilal Yavuz <byavuz81(at)gmail(dot)com> wrote:
> > > > + $prefix = (python -c "import sys; print(sys.prefix)").Trim()
> > > > + Add-Content $env:GITHUB_ENV "PYTHONHOME=$prefix"
> > > > + Add-Content $env:GITHUB_PATH $prefix
> > > > + Write-Host "PYTHONHOME=$prefix"
> > > > + Write-Host "Prepended $prefix to PATH"
> > >
> > > GRJGJKLJKJDFJKDF.
> >
> > I re-checked this since Jelte wasn't completely sure about this [2]
> > but this is unfortunately correct :(
>
> What are the chances we can strip Mercurial out of the PATH instead of
> messing with PYTHONHOME? I foresee pain in the future if we override
> that globally.

I think your suggestion is better, done.

> > v3 is attached.
>
> > + uses: msys2/setup-msys2(at)v2
>
> Should we pin this? It's the only third-party action we reference, and
> Scorecard [1] complains. (I'm not convinced its other complaints in
> this category are something we want to worry about, but this caught my
> eye.)
>
> We'd need to figure out how to keep it up to date, if we pinned it.
> But we probably need to figure out how to keep it up to date anyway.

Instead of using 'msys2/setup-msys2(at)v2', I directly installed the
packages by using msys2's package installer (pacman). There was one
problem, it seems that default installation of msys2 is on C: drive
and C: drive is somehow lots slower compared to D: drive [1]. I am
talking about total runtime being ~15 minutes slower (22m -> 35m). So,
I moved msys2 to D: drive and used it from there as a solution.

> Scorecard doesn't report any `Dangerous-Workflow` violations, so that's good.

Nice!

v4 is attached, GA run:
https://github.com/nbyavuz/postgres/actions/runs/26628396798

[1] https://github.com/actions/runner-images/issues/8755

--
Regards,
Nazir Bilal Yavuz
Microsoft