14:58:40 <h01ger> #startmeeting reproducible-builds.org in January
14:58:40 <MeetBot> Meeting started Tue Jan 31 14:58:40 2023 UTC.  The chair is h01ger. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:58:40 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:59:03 <h01ger> #topic introduction round, please indicate your presence, agenda at https://pad.riseup.net/p/rb-irc-meetings-keep
14:59:09 <h01ger> also feel free to amend the agenda
14:59:15 * h01ger = Holger Levsen
14:59:23 <Foxboron> <- Morten Linderud, Arch Linux dude
14:59:24 * lamby - Chris Lamb, working on all things reproduble, especially diffoscope and Debian
14:59:31 * h01ger will gives 3min or so now for this round...
14:59:35 * rclobus - Roland Clobus, working on Debian Live images
15:02:23 <h01ger> hello everybody, great that you're!
15:02:33 <lamby> \o/
15:02:47 <h01ger> we'll move on to the next topic now but please do feel free to say hi later/now
15:03:19 <h01ger> the last meeting in November 2022 didnt have any action iteams \o/ ;-p
15:03:28 <h01ger> so that was easy
15:04:12 * h01ger will be going to FOSDEM 2023 and is looking forward to meet many people there
15:04:43 <h01ger> afaik there are no dedicated talks about r-b, though i guess i will be corrected now, or later
15:05:01 <lamby> That's a good point - I assume the schedule is released.
15:05:04 <lamby> I'm not available to go, alas.
15:05:13 <Foxboron> I'll be going to FOSDEM as well :)
15:05:19 <lamby> Might watch some live streams of the interesting talks though, timezone permitting. ^_^
15:05:30 <h01ger> there will be an SFC booth somewhere which i intend to hang around a bit on sunday, else you can find me around the debian booth, probably outside around the entrances though
15:05:51 <rclobus> There appear to be some talks about reproducible builds: https://fosdem.org/2023/search/?q=reproducible
15:06:17 <h01ger> #info There appear to be some talks about reproducible builds: https://fosdem.org/2023/search/?q=reproducible
15:06:17 <Foxboron> The only one actually about reprobuilds is the container one
15:06:24 <Foxboron> Rest of them are about declarative environemtns
15:06:27 <Foxboron> environments*
15:07:11 <h01ger> at least something :)
15:07:14 <Foxboron> yess
15:07:17 <lamby> https://fosdem.org/2023/search/?q=supply+chain  <- bunch of relevant things are being labelled as 'supply chain' these days too
15:07:31 <h01ger> #info https://fosdem.org/2023/search/?q=supply+chain  <- bunch of relevant things are being labelled as 'supply chain' these days too
15:07:38 <lamby> … #info  you beat me :)
15:07:56 * h01ger has been a meetbot user for more than a decade :-)
15:08:38 * raboof[m] late introduction: Arnout Engelen, mostly working from the Nix perspective these days but upstreaming as much as possible
15:08:55 <lamby> hi raboof[m]
15:09:31 <h01ger> many interesting talks again
15:09:46 <h01ger> i guess we can move on to the next topic?!?
15:10:11 <Foxboron> yes
15:10:21 <h01ger> #topic short time slots for checkins from various projects
15:11:02 <h01ger> Ariadne: jelle: fepitre[m]: _hc[m]: aparcar:  are you there?
15:11:20 <h01ger> next would be Alpine with Ariadne
15:11:22 <Foxboron> I can take the Arch one if jelle doesnt appear :)
15:11:33 <h01ger> Foxboron: \o/ (& somehow i figured :)
15:11:47 <h01ger> #topic no update on Alpine Linux
15:12:05 <h01ger> #topic short time slot: Arch Linux (jelle / Foxboron)
15:12:23 <h01ger> also the question about the status of the live/installer .iso being reproducible..
15:12:44 <Foxboron> Not a lot has happened on repro builds. Some bugfixing in repro. Mostly focusing on migrating package repos from svn to git :p
15:13:10 <Foxboron> .iso reprodacbility is on my (long) todo, but sadly been focusing on other distro projects :/
15:13:17 <h01ger> :)
15:13:42 * h01ger is relieved that other projects also still have some svn2git migrations to do ;)
15:13:57 <h01ger> anything else about Arch?
15:14:13 <Foxboron> Nope :)
15:15:12 <h01ger> #topic short time slot: Debian ( rclobus / fepitre[m] / h01ger )
15:15:14 <rclobus> #info https://lists.reproducible-builds.org/pipermail/rb-general/2023-January/002821.html <- Regular status update
15:15:19 <h01ger> \o/
15:15:32 * h01ger echoes lamby's words from the ML: awesome work
15:15:45 <rclobus> Reproducible status: still all live images are reproducible without additional hacks
15:16:13 <rclobus> However, occasionally some images cannot be generated, due to ongoing transitions (Qt)
15:16:38 <h01ger> i think the two next important steps are: a.) non-free-firmware integration b.) making those builds official and ship with Bookworm
15:17:03 <rclobus> And: for the official regular builds, the local Debian mirror will be used, so I'll have to work on verifying that type of image (I'll need 2 timestamps)
15:17:23 <h01ger> rclobus: you said, that official builds would not be build against snapshot.d.o and i wondered why this is problematic? we just need to use snapshot for the rebuilds then?!
15:17:28 <h01ger> ah
15:17:36 <h01ger> you basically just explained
15:17:43 <rclobus> h01ger: Indeed. Item b will be the most important step, because then there could be more testers than just a few.
15:18:07 <h01ger> yeah
15:18:24 <h01ger> those user will want non-free-firmware too though :)
15:18:57 <rclobus> For testing reproducibility, it will be slightly more complicated, but that could be done while Bookworm has been released.
15:19:27 <rclobus> Before the release of Bookworm, at least b must have been done, but definately a would be good...
15:19:48 <rclobus> So far, Jenkins and openQA have been tremendously helpful in finding issues.
15:20:04 <h01ger> yay. i guess thats it for debian live builds for now?
15:20:14 <rclobus> Yes. Thanks for your time :-)
15:20:31 <h01ger> awesome, thanks for your summary!
15:20:35 <h01ger> about snapshot.r-b.o: SDDs have arrived at OSUOSL and there's hope they will be put into the machine this week (via replacing the HDDs with SDDs and hoping the HW raid will do ;) - if this works out, we dont need to copy the data again...
15:21:45 <rclobus> Nice.
15:22:18 <h01ger> about general Debian status: we've done quite some NMU sprints in December and January but since January 12th Debian has started to become frozen and starting February 12th definitly is not a good time anymore to do NMUs fixing reproducibility issues, so this aspect will stall a bit in the coming weeks
15:23:23 <h01ger> somewhat related, i've been nudging Lukas to prepare the new into-toto and securesytemslib versions for bookworm, but so far they are still busy with the  upstream releases..
15:23:33 <h01ger> .
15:25:14 <h01ger> any other commends etc about Debian?
15:26:50 <h01ger> #topic no update on F-Droid but Felix Stegermann has contributed some nice patches for diffoscope to better deal with APK files
15:27:08 <h01ger> #topic no update on OpenSUSE
15:27:20 <lamby> ^ Yes, I hope to be able to find some time later in the week to do a good clear-out of all outstanding diffoscope stuff, including APK updates
15:27:29 <h01ger> nice
15:27:43 <h01ger> #topic no updates on OpenWrt
15:27:58 <h01ger> #topic short time slots: NiX (raboof)
15:28:04 <h01ger> raboof[m]: the stage is yours!
15:28:26 <raboof[m]> for our minimal installer .iso reproducibility, we kinda put that on hold for now until after the release of 23.05, since that's when we bump to python 3.11 which may solve some of the remain
15:28:26 <raboof[m]> ing problems (https://github.com/NixOS/nixpkgs/issues/188978)
15:28:57 <raboof[m]> we're trying to reduce our bus factor for reproducibility reporting, moving to more shared infrastructure. Next step: getting https://nix-rb.bzzt.net/ published at https://reproducible.nixos.org
15:29:17 <h01ger> sounds great!
15:29:36 <h01ger> We can’t connect to the server at reproducible.nixos.org.
15:29:43 <h01ger> or is that a planned server? :)
15:30:03 <raboof[m]> we're now reporting on our gnome iso build closure as well (https://arnout.engelen.eu/nixos-iso-gnome-r13y/). Note that 'build closure' means it's not just everything that is in the ISO, but also (almost) everything that is used to build the ISO.
15:30:11 <raboof[m]> yeah that's the planned url :)
15:30:48 <h01ger> hehe, ok
15:31:06 * h01ger <3 r13y.com as a url too
15:31:19 <h01ger> anything else on NiX?
15:31:33 <raboof[m]> the list of problems in the gnome iso build closure is pretty long, but quite many of those are potentially-duplicate Haskell problems. That's it for now I think - in summary, steadily improving ;)
15:31:47 <h01ger> :thumbsup:
15:32:07 <h01ger> (i've not #info anything because it felt too much in the flow. that said, anyone can use #info :)
15:32:24 <h01ger> so, final topic i guess?
15:32:40 <h01ger> #topic Any Other Business
15:33:17 <lamby> Nothing here. :)
15:33:23 <rclobus> Even though attendance is relatively low, I would still like to keep the montly meetings
15:33:33 * h01ger has none, i'm just happy we've picked up these meetings again and that we've managed to have a short and (IMO) good one
15:33:53 <h01ger> rclobus: same here. i do expect some people do sometimes read the logs too
15:34:00 * mapreri shyly say hello at the very end
15:34:13 <h01ger> mapreri: ! o/
15:34:32 <mapreri> I'll make sure to read the backlog :)
15:34:50 <lamby> hey mapreri
15:36:43 <h01ger> any other business?
15:37:24 <rclobus> Nope
15:38:13 <h01ger> so lets's wrap this up!
15:38:23 <lamby> Thanks for running this h01ger
15:38:29 <h01ger> hey vagrantc, we're just closing the meeting :)
15:38:31 * vagrantc waves
15:38:34 <h01ger> thank you all for attending!
15:38:38 * vagrantc fails at timezones
15:39:41 <h01ger> dont we all, once in a while?
15:39:44 <h01ger> #endmeeting