#archlinux32 | Logs for 2024-04-21
Back
[04:30:31] -!- ssserpent has joined #archlinux32
[07:30:20] -!- abaumann has joined #archlinux32
[07:30:20] <buildmaster> Hi abaumann!
[07:30:20] <buildmaster> !rq abaumann
[07:30:21] <phrik> buildmaster: <abaumann> The next step will be to distribute code to neighbour servers or use the Amazon cluster in the build script to run tests..
[07:31:02] <abaumann> KitsuWhooa: glibc on i486 builds just fine outside of the build system, as soon as I am inside the build system I get all kind of weird issues.
[07:31:25] <abaumann> The weirdest so far is that partitions in /etc/fstab are no longer mounted automatically sind systemd updates.
[07:32:22] <abaumann> bill-auger: new python 3.12, what can possibly go wrong? ;-)
[07:33:24] <abaumann> ./manage-slaves list
[07:33:26] <abaumann> euronuc-i486 enabled sane
[07:33:38] <abaumann> ./manage-slaves enable euronuc-486
[07:33:38] <abaumann> Cannot find build slave "euronuc-486" in the database to enable.
[07:33:43] <abaumann> This also doesn't sound right..
[07:35:10] <abaumann> KitsuWhooa: can you leave it enabled. I disabled the systemd service for the builder and am building manually at the moment to catch the verify errors (something is fundamentally wrong in the build slave)
[07:41:30] <abaumann> glibc ... NOT FOUND
[07:41:38] <abaumann> validity error? huh?
[07:47:32] <abaumann> ==> Validating source files with b2sums... glibc ... NOT FOUND
[07:47:51] <abaumann> source=(git+https://sourceware.org/git/glibc.git#commit=${_commit}
[07:47:52] <phrik> Title: sourceware.org Git - glibc.git/summary (at sourceware.org)
[07:49:01] <abaumann> something is fundamentally broken in the handling of those protocols. I don't understand a) why there is a shasum for this source entry in the first place b) there is no hanlding of that case to siimply ignore checksumming in this case
[07:49:03] -!- drathir_tor has quit [Remote host closed the connection]
[07:49:19] <abaumann> does it come from upstream or are we adding a shasum...?
[07:50:02] -!- drathir_tor has joined #archlinux32
[07:50:42] <abaumann> c) why doesn't this happen in i686/pentium4
[07:50:56] <abaumann> I forgot b) :-)
[07:53:43] <abaumann> maybe a resync of devtools32 from upstream devtools reveils something
[08:26:16] <abaumann> /usr/bin/makepkg is owned by pacman 6.0.2-9.0
[08:26:19] <abaumann> on i486
[08:26:43] <abaumann> aha, so I have to make sure to have pacman/makepkg 6.1 on the 486 vms.. then it should work..
[08:26:59] <abaumann> as a side-effect there is now a new devtools32: https://git.archlinux32.org
[08:27:00] <phrik> Title: devtools32 - Archlinux32 fork of devtools (at git.archlinux32.org)
[08:27:28] <abaumann> it might have interesting things in it for rust..
[08:28:57] <abaumann> so, pacman qualifies as option d)
[11:34:17] <abaumann> yep. done. so now euronuc-i486 should work just fine with makepkg from pacman 6.1
[11:39:55] -!- AtleoS has joined #archlinux32
[12:40:31] -!- abaumann has quit [Quit: leaving]
[13:21:24] <KitsuWhooa> abaumann: I thought it was your IP being banned from upstream due to the constant cloning due to failures
[13:21:30] <KitsuWhooa> and yes, upstream added a way to hash git repositories
[13:22:53] <KitsuWhooa> maybe we should push all the 3.11 stuff to stable before 3.12 hits
[15:14:40] -!- ssserpent has quit [Quit: WeeChat 4.2.2]
[15:38:00] -!- bdju has quit [Ping timeout: 260 seconds]
[15:39:48] -!- bdju has joined #archlinux32
[17:02:15] -!- bdju has quit [Quit: Reconnecting]
[17:02:32] -!- bdju has joined #archlinux32
[19:17:42] -!- drathir_tor has quit [Ping timeout: 260 seconds]
[19:29:44] -!- drathir_tor has joined #archlinux32
[21:38:15] -!- AtleoS has quit [Ping timeout: 245 seconds]
[21:38:48] -!- AtleoS has joined #archlinux32