[Piuparts-devel] Bug#1036705: override: adduser:admin/required

[Cyril Brulebois]

Hi,

With only my random DD and d-i hats, leaving the release one aside…

Helmut Grohne <helmut at subdivi.de> (2023-05-24):
> I am requesting to override the priority of adduser to become
> required.

Watching from the sideline, this seems to come in horribly late.

> apt used to depend on adduser and apt is required, so adduser is
> transitively required in bullseye. Johannes and myself worked towards
> making apt not depend on adduser and that work succeeded.

FSVO “success” then, given the rest of the mail…

> We've now fixed such postrm scripts to no longer do that, but we agreed
> with the release team that it should be difficult to remove for bookworm
> in order to make purging packages left over from bullseye just work
> after and upgrade to bookworm. Originally, the idea was to add back the
> dependency from apt.

Out of curiosity, why wasn't that easy fix implemented?

> Instead, we made apt "Protected: yes".

Via olasd/#debian-release: adduser got that field, not apt.

> This still doesn't install it by default, but makes removal difficult
> which is what saves postrm purge scripts, so all should be good.
> Except that this makes piuparts unhappy as it tries to remove adduser
> and apt being unhappy about it. This is presently breaking testing
> migration for a number of packages. So now we thought about it again
> and got to the conclusion that adduser should also be Priority:
> required for bookworm (and unstable until bookworm is released).

Same question as before, why not just add the dependency back?

> Doing so is a late change, I know. However, it gets us back to the
> bullseye state and in being required, debootstrap --variant=minbase
> will install adduser again, which will fix piuparts. So an we do that?

Aren't we risking a redux of “we turned another knob, and now we're
discovering yet another issue”?


I'm not particularly worried about people using d-i to install minbase
specifically, and even with the open questions above, I wouldn't
normally object to the proposed change from a d-i perspective.

But I'm very much worried about possible side effects at this critical
stage of the freeze.


Cheers,
-- 
Cyril Brulebois (kibi at debian.org)            <https://debamax.com/>
D-I release manager -- Release team member -- Freelance Consultant
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/piuparts-devel/attachments/20230524/5ad9657b/attachment.sig>