#archlinux32 | Logs for 2021-02-11

[05:34:16] <buildmaster> !rq buildmaster
[05:34:17] <phrik> buildmaster: I might be insane, but never confused ... ;-)
[05:34:27] <buildmaster> i486/bees are broken (says nlopc46-i486bs1): https://archlinux32.org/buildmaster/build-log.php?a=i486&p=bees
[05:36:07] <trotz> 2021/02/11 05:35 CRIT archlinux32.org Home Certificate HTTP CRITICAL - Invalid HTTP response received from host on port 443: cURL returned 7 - Couldn't connect to server
[05:39:07] <buildmaster> ACTION goes insane.
[05:39:30] <abaumann> of course..
[05:40:48] <buildmaster> ACTION resumes sanity.
[05:41:26] <abaumann> not sure, if it's enough to put a file into build-support, especially for i486-only clients.
[05:41:50] <abaumann> I think, we should consider to build i486 directly on x86_64 hosts too..
[05:44:03] <abaumann> gpg was brilliantly broken lately on i486, now libseccomp/libmagic/file hits again
[05:45:07] <trotz> 2021/02/11 05:44 OK archlinux32.org Home Certificate OK - Certificate 'archlinux32.org' will expire on Mon 29 Mar 2021 12:10:44 AM GMT +0000.
[05:48:57] <buildmaster> i486/doctl is broken (says nlopc46-i486bs1): https://archlinux32.org/buildmaster/build-log.php?a=i486&p=doctl
[06:10:48] <abaumann> hope, file is ok now..
[06:56:07] -!- torv has joined #archlinux32
[06:59:39] <buildmaster> Hi abaumann!
[06:59:39] <buildmaster> !rq abaumann
[06:59:40] <phrik> buildmaster: gogo-gadgeto-strace
[07:02:21] -!- KeiraT has joined #archlinux32
[07:06:40] <abaumann> titus-livius?
[08:44:30] -!- titus_livius has joined #archlinux32
[08:44:35] <girls> ... finally
[08:45:09] <girls> logs, anyone?
[09:17:18] <trotz> 2021/02/11 09:16 WARN buildmaster Total Processes PROCS WARNING: 280 processes with STATE = RSZDT
[09:27:19] <trotz> 2021/02/11 09:26 OK buildmaster Total Processes PROCS OK: 206 processes with STATE = RSZDT
[10:34:17] -!- eschwartz has quit [Ping timeout: 256 seconds]
[10:40:27] -!- eschwartz has joined #archlinux32
[11:22:38] <buildmaster> i486/docker is broken (says nlopc46-i486bs1): https://archlinux32.org
[11:25:35] -!- abaumann has joined #archlinux32
[11:25:35] <buildmaster> Hi abaumann!
[11:25:36] <buildmaster> !rq abaumann
[11:25:36] <phrik> buildmaster: <abaumann> if your software depends on the gcc version, you should really stop developping software because your development process is seriously flawed.
[11:26:04] -!- AndrevS has quit [Read error: Connection reset by peer]
[11:29:21] -!- deep42thought has joined #archlinux32
[11:29:21] <buildmaster> Hi deep42thought!
[11:29:21] <buildmaster> !rq deep42thought
[11:29:22] <phrik> buildmaster: <deep42thought> dear gals: If a guy says, he fixes it, he will fix it - it's not necessary to remind him each year of it.
[11:29:25] <deep42thought> Hi abaumann!
[11:29:39] <deep42thought> so ghc is broken in hscolour, now?
[11:31:27] -!- torv has quit [Ping timeout: 268 seconds]
[11:31:33] <abaumann> hi deep42thought
[11:32:19] <abaumann> yeah, looks like it.
[11:32:46] <abaumann> If have to see, but I'm still recovering from the file trouble on my build slaves.
[11:32:58] <abaumann> I get nice error messages like "/etc/pacman.conf not found"
[11:33:08] <abaumann> no clue yet.
[11:33:14] <abaumann> but after lunch..
[11:33:18] -!- torv has joined #archlinux32
[11:35:46] <abaumann> mmh. the buildmaster still spits out haskell packages, so eventually hscolour will get fixes, thus ghc will build then..
[11:35:49] <abaumann> ..let's see. :-)
[11:39:53] <buildmaster> i486/skopeo is broken (says nlopc46-i486bs1): https://archlinux32.org
[12:32:30] <buildmaster> i486/grub is broken (says nlopc46-i486bs1): https://archlinux32.org
[12:33:59] -!- abaumann has quit [Quit: leaving]
[13:46:05] -!- AndrevS has joined #archlinux32
[14:39:40] -!- abaumann has joined #archlinux32
[14:39:40] <buildmaster> Hi abaumann!
[14:39:40] <buildmaster> !rq abaumann
[14:39:41] <phrik> buildmaster: <abaumann> should I make a tmpfs for /usr/share? ;-)
[14:43:49] -!- KeiraT has quit [Remote host closed the connection]
[14:45:07] -!- KeiraT has joined #archlinux32
[15:54:42] <deep42thought> Hehe, I like driving on roads with my bike, where *I* do have better traction than the cars :-D
[15:55:41] <abaumann> :-)
[15:55:55] <abaumann> ==> ERROR: pacman configuration file '/etc/pacman.conf' not found.
[15:55:58] <abaumann> ==> ERROR: /etc/makepkg.conf not found.
[15:56:03] <abaumann> chrootbuild
[15:56:03] <deep42thought> did you solve that error?
[15:56:11] <abaumann> I have no clue what's going wrong.
[15:56:42] <deep42thought> works for me(tm)
[15:56:57] <deep42thought> which devtools do you have installed?
[15:57:46] <abaumann> maybe old ones.
[15:57:56] <abaumann> I usually take the master from devtools32 git directly
[15:58:07] <deep42thought> should work
[15:58:11] <abaumann> but I remember, I had pgp problems before..
[15:58:17] <abaumann> ..and file with EBADSYS
[15:58:22] <abaumann> but those are fixed now(tm)
[15:58:24] <deep42thought> this is your normal build slave? or the i486 one?
[15:58:29] <abaumann> no, a i486 one
[15:58:33] <abaumann> only those don't work
[15:58:56] <abaumann> ok, next try, now really with the master
[15:59:07] <abaumann> the master (git) on the slave (machine)
[15:59:33] <deep42thought> !grab abaumann
[15:59:34] <phrik> deep42thought: Tada!
[15:59:49] <abaumann> nope, same.
[15:59:56] <deep42thought> git -C slave checkout origin/master
[16:00:13] -!- KeiraT has quit [Remote host closed the connection]
[16:00:15] <abaumann> I could track it down to systemd-nspawn of chrootbuild
[16:00:28] <deep42thought> so which pacman.conf is actually missing?
[16:00:32] <abaumann> I'm trying now to debug that script to see what it is doing
[16:00:38] -!- KeiraT has joined #archlinux32
[16:00:45] <abaumann> yeah, that's the funny part, all those files are around.
[16:00:54] <deep42thought> errr
[16:00:55] <abaumann> I suspect glibc 2.33, systemd, seccomp mahem again.
[16:01:07] <deep42thought> forks for me(tm)
[16:01:13] <abaumann> the only way a file cannot be found is if some stupid wrapper code is doing something funny.
[16:01:17] <abaumann> you are on glibc 2.33?
[16:01:30] <deep42thought> ah, wait
[16:01:48] <deep42thought> I'm not on glibc 2.33 on the vm, but in the chroot, I believe
[16:01:55] <deep42thought> glibc 2.32-5.0
[16:01:58] <abaumann> aha.
[16:01:59] <deep42thought> try using that :-)
[16:02:02] <abaumann> don't upgrade :-)
[16:02:25] <deep42thought> ok, won't do :-)
[16:04:14] <jonathon> The AUR package glibc-linux4 should help for build containers on outdated hosts; it works for me on an x86_64 host.
[16:05:32] <abaumann> this is for kernel 4.x?
[16:05:38] <abaumann> for older syscalls?
[16:06:02] <abaumann> --host=i686-pc-linux-gnu
[16:06:04] <abaumann> hrum :-)
[16:06:29] <abaumann> let me see if it's the hosts glibc interfereing (i486 host) or the chroot (also i486).
[16:07:38] <abaumann> all this eating our own dogfood is fine.. but maybe exagerated on i486..?
[16:07:59] <deep42thought> I thought, we did this to trick smart cpu-flag detection scripts?
[16:08:08] <abaumann> yes.
[16:08:15] <abaumann> but we don't do it for i686 and pentium4
[16:08:28] <deep42thought> true
[16:08:30] <abaumann> and there scripts could also "detect" a SSSE3 or so
[16:09:08] <abaumann> if we built i486 on a x86_64 host we would just get artifacts which don't run
[16:09:09] <deep42thought> to be honest, I already force-scheduled some i486 packages onto nlopc43
[16:09:12] <abaumann> more likely..
[16:09:20] <abaumann> ah..
[16:09:21] <abaumann> me too.
[16:09:27] <deep42thought> glibc, for example
[16:09:50] <abaumann> for instance my forced rebuilds of file, openssh all went onto a slave on eurobuild6 directly.
[16:09:58] <abaumann> glibc doesn't to sniffing.
[16:10:09] <abaumann> unless the latest version with wrong ISA levels. :->
[16:10:25] <abaumann> those where there for i686 only, not i486 and pentium4
[16:10:31] <deep42thought> <.<
[16:10:57] <abaumann> so, glibc-2.32-5.0-i486 on the i486 host now
[16:11:04] <abaumann> installing a glibc 2.33 into the chroot
[16:11:19] <abaumann> I'm pretty sure it's a seccomp/glibc/systemd thingy. :-)
[16:11:51] <abaumann> nope. same effect.
[16:12:46] <abaumann> ok, let me try to downgrade glibc in the chroot and the run arch-nspawn /var/lib/archbuild/staging-i486/build --bind=/home/build/builder/work/tmp.lA5l3x:/startdir --bind=/home/build/builder/work/tmp.lA5l3x:/srcdest --bind=/var/cache/archbuild32:/var/cache/pacman/pkg /bin/bash -x /chrootbuild --syncdeps --noconfirm --log --holdver --skipinteg --install
[16:13:53] <abaumann> sudo: error in /etc/sudo.conf, line 0 while loading plugin "sudoers_policy"
[16:13:54] <abaumann> sudo: unable to load /usr/lib/sudo/sudoers.so: /usr/lib/libc.so.6: version `GLIBC_2.33' not found (required by /usr/lib/libcrypto.so.1.1)
[16:13:56] <abaumann> yeah.
[16:14:07] <deep42thought> uff
[16:14:18] <abaumann> downgrading sudo :-)
[16:14:46] <abaumann> pacman
[16:14:46] <abaumann> pacman: /usr/lib/libc.so.6: version `GLIBC_2.33' not found (required by /usr/lib/libcrypto.so.1.1)
[16:14:49] <abaumann> ahem. ok.
[16:14:52] <abaumann> that's not so easy
[16:15:41] <abaumann> ok, stracing..
[16:18:10] <abaumann> faccessat2(AT_FDCWD, "/etc/pacman.conf", R_OK, AT_EACCESS) = -1 EPERM (Operation not permitted)
[16:18:19] <deep42thought> ah, permissions
[16:18:45] <abaumann> -rw-r--r-- 1 root root 2830 Feb 11 16:17 /var/lib/archbuild/staging-i486/build/etc/pacman.conf
[16:18:52] <deep42thought> hmmm
[16:18:55] <deep42thought> strange, then
[16:18:55] <abaumann> faccessat2 sounds like a new syscall.
[16:19:33] <abaumann> did I say that I consider seccomp a flawed idea?
[16:20:07] <abaumann> Ok, I have seen what I wanted to see..
[16:20:12] <deep42thought> I have to little knowledge to judge that
[16:20:57] <abaumann> every time glibc wraps new syscalls, libseccomp falls behind (and a set of other packages like openssh/seccomp jail, file/libmagic, firefox/chromium/*webkit for sandboxing, etc.)
[16:21:19] <deep42thought> ah, yes, right, I remember
[16:21:31] <abaumann> its like with SELinux where all commands fail.. all sysadmins start an audit2xxx script and the add the rules to the exceptions..
[16:21:37] <abaumann> this is not security.. this is b*
[16:22:16] <abaumann> anf for seccomp to work it should be added to _ALL_ of the userland..
[16:22:43] <abaumann> syscalls differ per architecture and version of the kernel, glibc is supposed to create a stable API for the userland..
[16:23:08] <abaumann> ..with seccomp this is not achievable because it croaks about syscalls a layer too deep.
[16:23:28] <abaumann> seccomp makes sense in container (systemd), browsers (sandboxing), openssh (to isolate children).
[16:23:37] <abaumann> Otherwise it's questionable.
[16:24:00] <abaumann> (contradicting myself above with _ALL_).. mmmh :-)
[16:24:28] <abaumann> ok.. fixes for upstream first.. then we can reenable those broken slaves..
[16:46:45] -!- sunshavi has quit [Ping timeout: 264 seconds]
[17:02:41] -!- abaumann has quit [Quit: leaving]
[17:04:48] -!- abaumann has joined #archlinux32
[17:04:48] <buildmaster> Hi abaumann!
[17:04:48] <buildmaster> !rq abaumann
[17:04:49] <phrik> buildmaster: <abaumann> *abaumann feelds Sunday morning sarcasm entering the room - not that it would be any different from weekday sarcasm.
[17:07:45] <abaumann> patch: **** Can't open patch file ../mpdecimal-2.5.1.patch : No such file or directory
[17:08:25] <abaumann> cd Python-${pkgver}
[17:08:25] <abaumann> patch -p1 -i ../mpdecimal-2.5.1.patch
[17:08:25] <abaumann> patch -p1 -i ../bpo-42794.patch
[17:08:29] <abaumann> where is bpo-42794.patch
[17:08:45] <abaumann> ah, https://github.com
[17:09:16] <abaumann> but source=("https://www.python.org/ftp/python/${pkgver%rc*}/Python-${pkgver}.tgz")
[17:11:08] <abaumann> yeah, a little bit brutal, this patching of mine. :-)
[17:16:57] <abaumann> mmh.
[17:17:00] <abaumann> https://archlinux32.org
[17:17:01] <phrik> Title: Buildmaster for Archlinux32 packages (i486, i686, pentium4, any) (at archlinux32.org)
[17:17:05] <abaumann> now it's running nicely :-)
[17:17:08] -!- abaumann has quit [Quit: leaving]
[17:25:53] -!- deep42thought has parted #archlinux32
[18:00:41] -!- MrBIOS has joined #archlinux32
[18:14:07] -!- sunshavi has joined #archlinux32
[18:21:51] -!- Cthulu202 has joined #archlinux32
[18:23:45] -!- Cthulu201 has quit [Ping timeout: 240 seconds]
[18:32:01] <buildmaster> i486/python is broken (says nlopc46-i486bs1): https://archlinux32.org
[18:47:09] <buildmaster> i486/libsndfile is broken (says nlopc46-i486bs1): https://archlinux32.org
[18:47:39] -!- Cthulu202 has quit [Quit: Nowhere special. I always wanted to go there.]
[18:55:06] -!- Cthulu201 has joined #archlinux32
[20:13:40] <buildmaster> pentium4/haskell-gnuidn is broken (says eurobuild6-6): https://archlinux32.org
[20:14:41] <buildmaster> i686/haskell-gnuidn is broken (says rechenknecht): https://archlinux32.org
[20:15:31] <buildmaster> pentium4/haskell-alsa-mixer is broken (says eurobuild6-1): https://archlinux32.org
[20:18:33] <buildmaster> i686/haskell-alsa-mixer is broken (says eurobuild3): https://archlinux32.org
[20:41:08] <buildmaster> pentium4/libvirt is broken (says eurobuild6-6): https://archlinux32.org
[21:15:00] <buildmaster> i686/vault is broken (says eurobuild6-6): https://archlinux32.org
[21:16:08] <buildmaster> i686/brltty is broken (says nlopc43): https://archlinux32.org
[21:17:23] <buildmaster> pentium4/vault is broken (says eurobuild6-5): https://archlinux32.org
[21:20:44] <buildmaster> pentium4/brltty is broken (says eurobuild6-6): https://archlinux32.org
[21:39:02] <buildmaster> i686/haskell-syb is broken (says nlopc43): https://archlinux32.org
[21:43:22] <buildmaster> pentium4/haskell-generic-deriving is broken (says nlopc43): https://archlinux32.org
[21:44:39] <buildmaster> pentium4/haskell-hspec-discover is broken (says nlopc43): https://archlinux32.org
[21:46:39] -!- buildmaster has quit [Remote host closed the connection]
[21:46:50] -!- buildmaster has joined #archlinux32
[21:46:51] <buildmaster> !rq buildmaster
[21:46:52] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[21:49:23] <buildmaster> i686/haskell-base-compat-batteries are broken (says nlopc43): https://archlinux32.org
[21:54:42] <buildmaster> pentium4/haskell-nanospec is broken (says nlopc43): https://archlinux32.org
[22:00:20] <buildmaster> i686/haskell-splitmix is broken (says nlopc43): https://archlinux32.org
[22:05:05] <buildmaster> i686/haskell-colour is broken (says nlopc43): https://archlinux32.org
[22:06:48] <buildmaster> i686/haskell-clock is broken (says nlopc43): https://archlinux32.org
[22:19:20] <buildmaster> pentium4/haskell-random is broken (says nlopc43): https://archlinux32.org
[22:22:37] <buildmaster> i686/haskell-stringbuilder is broken (says nlopc43): https://archlinux32.org
[22:28:50] <buildmaster> pentium4/haskell-stringbuilder is broken (says nlopc43): https://archlinux32.org
[22:35:23] <buildmaster> i686/haskell-vector is broken (says nlopc43): https://archlinux32.org
[22:39:04] <buildmaster> pentium4/haskell-fold-debounce is broken (says eurobuild6-6): https://archlinux32.org
[22:39:06] <buildmaster> i686/haskell-fold-debounce is broken (says eurobuild6-5): https://archlinux32.org
[22:39:15] <buildmaster> pentium4/haskell-loop is broken (says nlopc43): https://archlinux32.org
[22:41:14] <buildmaster> i686/haskell-loop is broken (says eurobuild6-4): https://archlinux32.org
[22:44:07] <buildmaster> i686/haskell-reinterpret-cast is broken (says nlopc43): https://archlinux32.org
[22:44:59] <buildmaster> i686/haskell-validity is broken (says eurobuild6-4): https://archlinux32.org
[22:46:05] <buildmaster> pentium4/haskell-reinterpret-cast is broken (says eurobuild6-6): https://archlinux32.org
[22:48:38] <buildmaster> i686/haskell-with-location is broken (says eurobuild6-4): https://archlinux32.org
[22:48:39] <buildmaster> pentium4/haskell-validity is broken (says nlopc43): https://archlinux32.org
[22:49:33] <buildmaster> pentium4/haskell-with-location is broken (says eurobuild6-6): https://archlinux32.org
[22:52:54] <buildmaster> i686/haskell-newtype-generics are broken (says eurobuild6-5): https://archlinux32.org
[23:00:21] <buildmaster> pentium4/haskell-newtype-generics are broken (says eurobuild6-4): https://archlinux32.org
[23:03:34] <buildmaster> pentium4/haskell-mysql is broken (says eurobuild6-5): https://archlinux32.org
[23:03:45] <buildmaster> i686/haskell-mysql is broken (says rechenknecht): https://archlinux32.org
[23:05:40] <buildmaster> i686/haskell-network-byte-order is broken (says eurobuild6-1): https://archlinux32.org
[23:07:04] <buildmaster> pentium4/haskell-word-wrap is broken (says eurobuild6-4): https://archlinux32.org
[23:14:12] <buildmaster> i686/haskell-word-wrap is broken (says eurobuild6-6): https://archlinux32.org
[23:17:24] <buildmaster> i686/haskell-resourcet is broken (says rechenknecht): https://archlinux32.org
[23:25:07] <buildmaster> i686/haskell-pem is broken (says eurobuild6-1): https://archlinux32.org
[23:28:13] <buildmaster> pentium4/haskell-pem is broken (says eurobuild6-4): https://archlinux32.org
[23:29:34] <buildmaster> pentium4/haskell-resourcet is broken (says nlopc43): https://archlinux32.org
[23:45:12] <buildmaster> pentium4/haskell-mwc-random is broken (says nlopc43): https://archlinux32.org