#archlinux32 | Logs for 2023-06-17

Back
[01:07:56] -!- lithium_pt has quit [Ping timeout: 240 seconds]
[06:22:03] <girls> abaumann (sry for the slow reply - my desktop irc client is broken): AFAIR, seed-build-list was never able to add packages to the build list, you're looking for copy-to-buildsupport or just wanting to add something in our packages32-repository under build-support/$pkgname. Regarding the "I want to build immediately": theres the forced-package-buils.$slavename files, or (proposal): we could make the "priority column on the build_assignments
[06:22:03] <girls> table the highest rank, i.e. no package with non-highest priority will ever be considered for building, even if it was a dependency of a highest-priority package. What do you think of that?
[06:23:45] <girls> this would also considerably speed up get-assignment, as long as not all build-assignments have the same priority
[07:25:22] <KitsuWhooa> > ssh: connect to host buildmaster.archlinux32.org port 22: Connection refused
[07:25:24] <KitsuWhooa> uh oh
[12:55:13] -!- lithiumpt has joined #archlinux32
[13:10:33] -!- abouvier has quit [Quit: kthxbye]
[13:10:56] -!- abouvier has joined #archlinux32
[13:36:27] <KitsuWhooa> 2023-06-17 14:25:43: building package "firefox" (revisions 38c69a77548b41c624966b8fb1b823ba992b5631 532431f1fc2e6e61f8cfae7a02d8261db7699fed, repository extra, straw :mirrored_source_by_hash:) for pentium4 ...
[13:36:29] <KitsuWhooa> ha ha ha
[14:22:41] <KitsuWhooa> memcached seems to hang whatever builder is trying to build it
[15:18:23] <KitsuWhooa> I've been manually killing the main test process (perl) for memcached so that it'd get marked as broken and unblock the builder
[15:18:54] <KitsuWhooa> seems to have worked: 8:18:03 <buildmaster> i686/memcached is broken (says build-tasossah-3):
[15:35:05] -!- abaumann has joined #archlinux32
[15:35:05] <buildmaster> Hi abaumann!
[15:35:05] <buildmaster> !rq abaumann
[15:35:06] <phrik> buildmaster: <abaumann> I have a theory: Linux decided they want to take over the Windows market, but this is of course only possible by getting the same quality and bugs as Windows has.. ;-)
[15:35:50] <abaumann> deep42thought: no worries, slow replies are not an issue. Good idea to prioritize packages this way. Of course one has to know what one is doing if giving a package that high priority. :-)
[15:36:56] <abaumann> KitsuWhooa: yeah, sadly this sometimes things are hanging, the usual culprit are tests which can be ommitted with 'unset check' in a PKGBUILD diff file.
[15:37:31] <KitsuWhooa> abaumann: if you can push it for memcached, that'd be great
[15:37:52] <abaumann> ok. lemme see what the trouble is there..
[15:38:14] <KitsuWhooa> It was on i686. Idk about pentium4
[15:48:03] <abaumann> Seccomp policy failure. Caught syscall 403. This is either an exploit attempt, or your system is not supported yet.
[15:48:19] <abaumann> this thing is plastered with seccomp wrapping.. it won't run, even if it builds..
[15:49:06] <KitsuWhooa> oh huh
[15:59:44] <abaumann> I checked in, tests seem to work fine without seccomp.
[16:00:18] <abaumann> I never understood the idea behing libseccomp anyway. It just creawtes problems with C library shims to the kernel and I doubt it really improves security..
[16:00:42] <KitsuWhooa> is seccomp at fault then?
[16:01:02] <abaumann> hard to say. maybe libseccomp is stale (because of Python bindings not building)
[16:01:21] <KitsuWhooa> ah
[16:27:43] <abaumann> pentium4 0000000000000000000000000000000000000000 90479ec80eead2de31d3aca3f4c61dfb1925f81d blacklist fatal: pathspec 'blacklist/pentium4/PKGBUILD' did not match any files
[16:27:46] <abaumann> tar: This does not look like a tar archive
[16:27:48] <abaumann> tar: blacklist/pentium4/PKGBUILD: Not found in archive
[16:27:51] <abaumann> tar: Exiting with failure status due to previous errors
[16:27:53] <abaumann> ==> ERROR: pkgrel is not allowed to be empty.
[16:27:56] <abaumann> ==> ERROR: pkgver is not allowed to be empty.
[16:27:58] <abaumann> ==> ERROR: pkgname is not allowed to be empty.
[16:28:01] <abaumann> "make_source_info" had empty output - eh, what?
[16:28:03] <abaumann> and this after running for 30 minutes
[16:30:20] <abaumann> aha: 90479ec80eead2de31d3aca3f4c61dfb1925f81d belongs to extra/python-prefile: bootstrapping
[16:30:50] <abaumann> instead of python-pefile :-
[16:30:51] <abaumann> )
[16:31:31] <abaumann> mmh. strange. looks ok to me.
[16:35:17] <abaumann> I'm removing it again. the buildmaster doesn't like it, it seems.
[17:26:05] -!- abaumann has quit [Quit: leaving]
[17:28:54] -!- abaumann has joined #archlinux32
[17:28:58] -!- abaumann has parted #archlinux32
[17:29:01] -!- abaumann has joined #archlinux32
[17:29:08] -!- abaumann has parted #archlinux32
[17:29:14] -!- abaumann has joined #archlinux32
[17:29:19] <abaumann> pentium4 0000000000000000000000000000000000000000 13e9ad47db413bfc1e2b832725697889c6448b6b blacklist fatal: pathspec 'blacklist/pentium4/PKGBUILD' did not match any files
[17:29:38] <abaumann> still, also after removing python-pefile I get this error with get-package-updates
[17:29:39] -!- abaumann has quit [Client Quit]
[17:31:26] <bill-auger> seem to have some validation holes there
[17:32:23] <bill-auger> ==> ERROR: PKGBUILD is not allowed to not exist
[19:47:47] <girls> fixed the blacklist bug, also made the priority as proposed
[19:48:02] <girls> ... let's see ...
[21:57:55] <KitsuWhooa> What did you do, lol
[21:58:03] <KitsuWhooa> the queued packages went back up to 20k
[22:11:12] <bill-auger> no problem - if they average 4 seconds per package, that should only take a week
[23:43:11] -!- T`aZ has quit [Ping timeout: 246 seconds]
[23:43:26] -!- T`aZ has joined #archlinux32