Samuel Thibault writes:
Hi,
> jann...@gnu.org, le mar. 05 nov. 2024 23:50:44 +0100, a ecrit:
>> > For GNU Guix I'm trying to upgrade our packages gnumach to
>> > v1.8+git20240406 and hurd to v0.9.git20240714, the most recent tags (as
>> > a step towards possible 64bit support), but the resulting VM does not
>> > boot.
>>
>> Two more data points: downgrading the hurd package only to our current
>> version: v0.9.git20231217 works, so with new v1.8+git20240406 gnumach.
>>
>> Upgrading the hurd package to the very latest commit
>> 00b1784c3a3d90e1abac40b6c7de3ebbbb05b6a2, also doesn't boot.
>>
>> Do you think it would make sense to bisect between v0.9.git20240714 and
>> latest for a hurd that boots?
>
> Yes, that would be welcome.
Found it, and as so often in retrospect, prretty obvious:
--8<---------------cut here---------------start------------->8---
cb6b4024b5e0b79f6416dc9d2553208b981f4550 is the first bad commit
commit cb6b4024b5e0b79f6416dc9d2553208b981f4550
Author: Damien Zammit <dam...@zamaudio.com>
Date: Tue Mar 5 01:30:02 2024 +0100
ext2fs: New default: use xattrs to store translator records
Replaces experimental option --x-xattr-translator-records
with --no-xattr-translator-records to allow rolling back to
previous behaviour.
--8<---------------cut here---------------end--------------->8---
Of course, it would (have) be(en) nice if the boot would not silently
hang but say something. Oh well, guess that was what the "shameless
plug" was aiming at. :)
Removing "--x-xattr-translator-records" from our grub config generation
fixes the boot.
Greetings,
Janneke
--
Janneke Nieuwenhuizen <jann...@gnu.org> | GNU LilyPond https://LilyPond.org
Freelance IT https://www.JoyOfSource.com | Avatar® https://AvatarAcademy.com
