Re: towards a reproducible forky

Jochen Sprickerhof <[email protected]> Sun, 19 Jul 2026 17:51:32 +0200
Newsgroups	gmane.linux.debian.devel.release
Organization	The Debian Project
Message-ID	<alzygQ3s9ShZ0yzN@mpd>
Hi Sebastian,

* Sebastian Ramacher <[email protected]> [2026-07-19 15:54]:
>On 2025-11-06 19:44:38 +0100, Jochen Sprickerhof wrote:
>> Hi,
>>
>> * Paul Gevers <[email protected]> [2025-09-27 17:48]:
>> > On IRC Sebastian also brought up that we'd need to agree on what we'd
>> > expect the environment of r.d.n to look like. Obviously CPU, RAM,
>> > hostname and exact build times will be different from the buldds, but
>> > most other things we'd expect the same, like umasks, timezones and
>> > locale settings. Are those the same across all buildds? A point of
>> > attention is the packages installed on the buildds which aren't
>> > specified in the build dependencies and thus not logged in the buildinfo
>> > files: apt, openssl, ca-certificates and fakeroot (the latter depending
>> > on Rules-Requires-Root: binary-targets if I get it right).
>>
>> We assembled a list of differences during the reproducible builds summit in
>> Vienna. Fell free to adopt it for the release team documentation.
>>
>> General system properties:
>>
>>   - system time
>>   - system uptime
>>   - CPU model, load, number of cores
>>   - RAM amount, speed
>>   - kernel version, process scheduling, various kernel settings
>>   - disk space, speed
>>   - file order, mtime, ctime, atime, inodes
>>   - /sys/* (due to different system)
>>
>> Specific to debrebuild:
>>
>>   - apt, openssl, ca-certificates not installed by default (not in buildinfo file)
>>   - fakeroot not used for Rules-Requires-Root (not in buildinfo file)
>>   - dpkg install order is not guaranteed - can have effects on maintainer scripts, UIDs added by packages..
>>   - /etc/apt/sources.list* content
>>   - /run/* (due to mmdebstrap cleaning it up)
>>   - /var/log/*
>>   - LD_LIBRARY_PATH does not contain fakeroot when running with RRR: binary-targets
>>
>> Specific to reproduce.debian.net:
>>
>>   - tmpfs vs ext4 (will probably be fixed)
>>   - nice level
>
>Why is the nice level different from the buildds? It is reported on IRC
>that this causes packages to be blocked. What is the benefit from this
>test?
>
>FTBFS under a different nice level than on buildds is an important bug
>at most. But as FTBFS on r.d.n is now causing packages to be blocked
>from migrating to testing, please remove this variation.

I think that can be argued both ways and I am also fine with dropping 
it. Added Holger as he is in charge of this.

Cheers Jochen
signature.asc (application/pgp-signature, 833 B)