#archlinux32 | Logs for 2023-05-22

Back
[01:29:23] -!- drathir_tor has quit [Remote host closed the connection]
[04:09:42] -!- drathir_tor has joined #archlinux32
[06:36:02] -!- T`aZ has joined #archlinux32
[08:15:42] -!- bgyorgy[m] has joined #archlinux32
[09:17:51] <girls> theoretically, the buildmaster should work again - modulus the ever-possible bugs, that I for sure have not thought of
[09:18:34] <girls> ah, a
[09:18:57] <girls> nd there are some features missing, of course
[09:21:52] <girls> abaumann: are you still using local-build-package? this looks broken ...
[10:22:53] -!- abaumann has joined #archlinux32
[10:22:54] <buildmaster> Hi abaumann!
[10:22:54] <buildmaster> !rq abaumann
[10:22:54] <phrik> buildmaster: <abaumann> the Linux kernel goes the Firefox way a) create a foundation, b) let everybody code on it, c) build times exceed the 20 minute margin on modern hardware d) reprogram everything in rust
[10:23:19] <abaumann> deep42thought: I don't really need the local_package thingy anymore.
[10:25:54] <abaumann> (sorry, my webmail frontend "lost" it's PGP keys, I can read your mail only in the evening at home) :-)
[10:26:09] <girls> no hurry :)
[11:41:15] -!- abaumann has parted #archlinux32
[11:41:27] -!- abaumann has joined #archlinux32
[11:41:27] <buildmaster> Hi abaumann!
[11:41:27] <buildmaster> !rq abaumann
[11:41:27] <phrik> buildmaster: <abaumann> Do you have a ii-hook to check for TODO descriptions if I asked for silly features? ;-)
[11:41:47] -!- abaumann has parted #archlinux32
[13:37:05] -!- bill-auger has quit [Ping timeout: 240 seconds]
[13:37:25] -!- bill-auger_ has quit [Ping timeout: 240 seconds]
[13:37:56] -!- bill-auger has joined #archlinux32
[13:38:01] -!- bill-auger_ has joined #archlinux32
[14:45:33] <girls> oh, boy - there is no direct link from the pkgbase of a package to the repository containing the PKGBUILD - mostly, it's identical, but sometimes it just is not: "+" becomes "plus" or "-" (that's, what I have seen so far ...)
[15:31:44] <Menzador> >reprogram everything in Rust
[15:31:48] * Menzador sighs in Ninja
[15:42:36] <girls> yeah, I was pondering to rewrite everything in something else than bash, but then I thought: What language comes with a nice git, curl integration anyways
[15:43:17] <Foxboron> girls: fwiw, there was a test setup available since late January
[15:43:34] <Foxboron> it was announced on arch-dev-public and you could have been given access to stuff for testing purposes
[15:44:15] <girls> yeah, I guessed so - as I said: no heart feelings
[15:46:39] <girls> otoh: I have my foot broken *now* - from where would I have taken the time in January :-D
[15:49:26] <KitsuWhooa> Injuring maintainers so that they take time off to work on their projects 🤔
[15:49:52] <KitsuWhooa> (/j if it wasn't obvious)
[16:21:39] -!- abaumann has joined #archlinux32
[16:21:39] <buildmaster> Hi abaumann!
[16:21:39] <buildmaster> !rq abaumann
[16:21:40] <phrik> buildmaster: <abaumann> canary no. 2 is back on testing. :-)
[16:22:14] <abaumann> yeah, and I didn't have time and I know little about the build infrastructure anyway.. I would have made a terrible mess. :-)
[16:22:32] * abaumann wonders if breaking two feet increases the speed of development ;-)
[16:23:23] -!- abaumann has quit [Client Quit]
[16:39:20] -!- abaumann has joined #archlinux32
[16:39:21] <buildmaster> Hi abaumann!
[16:39:21] <buildmaster> !rq abaumann
[16:39:21] <phrik> buildmaster: <abaumann> being currently under a DoS myself since almost two weeks, so I'm a little bit angry at the internet specifically and the world in general.
[16:39:28] <abaumann> https://bbs.archlinux32.org
[16:39:29] <phrik> Title: Arch's Git Migration is a Mess: What Can we Learn for ArchLinux32? / Pacman / Pacman Upgrades / Arch Linux 32 Forum (at bbs.archlinux32.org)
[16:39:33] <abaumann> I hope, I put that right.
[16:39:54] <abaumann> Foxboron: I anybody is to blame, then us not adapting the tools. :-)
[16:40:32] <abaumann> Congrats to the migration, I know first hand how it is to move huge software repos (went all the way from RCS, PRCS, Subversion to Git)
[16:40:43] <abaumann> without loosing any vital information, that is.
[16:41:54] <abaumann> pacman -Syu "pacman>=6.0.2-7"
[16:42:00] <abaumann> pacman -Syu "devtools>=1:1.0.0-1"
[16:42:03] <abaumann> pacman -Syu "devtools>=1:1.0.0-1" "pacman>=6.0.2-7"
[16:42:09] <abaumann> error: target not found: devtools>=1:1.0.0-1
[16:42:09] <abaumann> error: target not found: pacman>=6.0.2-7
[16:42:22] <abaumann> mmh. I might be missing something here..
[16:42:35] <abaumann> ah. wrong mirror. :-)
[16:47:19] <abaumann> deep42thought: /scratch is ext4 and upstream-packages is a of the complexity (nof_modules x nof_versions), but as long as we don't iterate through that directory and just access packages by name, it should work.
[16:47:30] <abaumann> otherwise I can change the underlying filesystem
[16:47:39] <abaumann> ls /scratch/srv/http/upstream-packages/ | wc -l
[16:47:41] <abaumann> 3265
[16:47:46] <abaumann> ok, somebody is caching :-)
[16:51:56] <abaumann> 'pkgctl repo clone' works only for authenticated people?
[17:01:05] * abaumann actually starts to read the man page of pkgctl
[17:12:17] <abaumann> ok, I'm not developer to TU, so I failed at cloning a repo..
[17:12:56] <abaumann> This is all a little bit Arch developer centric for my taste :-)
[17:14:14] * abaumann leaves it for another day
[17:14:15] -!- abaumann has quit [Quit: leaving]
[17:24:18] -!- abaumann has joined #archlinux32
[17:24:18] <buildmaster> Hi abaumann!
[17:24:18] <buildmaster> !rq abaumann
[17:24:19] <phrik> buildmaster: <abaumann> I made the experience that when the config is simple and looking nicely it usually also works. :-)
[17:25:11] <abaumann> actually, I can generate the access token and log in and clone with pkgctl just fine.. but I can imagine users of pkgctl which just want to build a package and might not want to go through the same hoops :-)
[17:25:14] -!- abaumann has quit [Client Quit]
[18:10:14] <girls> abaumann, FYI: I hacked a "head -n ..." into get-package-updates, so one can now run it "in slices" - but this is really a hack, one should still run it without this option in the final run.
[18:19:21] <girls> ah, and another warning: because the tracking of the package version was changed, we probably have to rebuild all packages :-/
[18:19:35] <girls> so prepare for some fun, fun, fun
[19:50:06] <girls> fatal: unable to access 'https://gitlab.archlinux.org/archlinux/packaging/state.git/': Failed to connect to gitlab.archlinux.org port 443 after 7 ms: Couldn't connect to server
[19:50:16] <girls> wow, I guess, we hit some rate limiting, now :)
[19:55:56] <girls> Foxboron: is it possible, to get us unblocked again?
[19:56:19] <girls> we're caching the sources anyways - but we have to download each one once :)
[19:56:52] <girls> 85.10.198.216 and 2a01:4f8:a0:5264::2
[20:20:29] <girls> hmm, either you fixed it, or it fixed itself - in the first case: thank you! in any case: issue solved :-)