#archlinux32 | Logs for 2019-05-23

Back
[00:03:04] -!- aliemjay has quit [Quit: Konversation terminated!]
[00:19:35] <buildmaster> pentium4/dart is broken (says nlopc43).
[00:22:30] <buildmaster> pentium4/python-yara is broken (says nlopc43).
[00:22:56] <buildmaster> pentium4/ctags are broken (says buildknecht2).
[00:24:44] -!- Dimtree has quit []
[00:58:03] -!- eschwartz has quit [Ping timeout: 244 seconds]
[01:49:47] -!- MrBIOS has quit [Read error: Connection reset by peer]
[01:50:05] -!- MrBIOS has joined #archlinux32
[02:49:52] -!- eschwartz has joined #archlinux32
[03:43:11] <buildmaster> pentium4/firefox is broken (says buildknecht2).
[04:38:40] <buildmaster> pentium4/facter is broken (says buildknecht2).
[04:44:47] -!- eschwartz has quit [Ping timeout: 248 seconds]
[04:51:59] <buildmaster> pentium4/logstash is broken (says nlopc43).
[04:59:16] -!- eschwartz has joined #archlinux32
[05:36:44] <buildmaster> pentium4/libreoffice-fresh is broken (says buildknecht2).
[05:50:18] <buildmaster> pentium4/qt5-webkit is broken (says nlopc43).
[05:53:32] eschwartz is now known as guys
[05:55:42] guys is now known as ztrawhcse
[06:00:31] -!- thePiGrepper has quit [Ping timeout: 248 seconds]
[06:11:04] <buildmaster> pentium4/sysdig is broken (says buildknecht2).
[06:44:20] -!- ztrawhcse has quit [Ping timeout: 272 seconds]
[06:58:46] -!- ztrawhcse has joined #archlinux32
[07:01:33] -!- buildmaster has quit [Remote host closed the connection]
[07:02:09] -!- buildmaster has joined #archlinux32
[07:02:09] <buildmaster> !rq buildmaster
[07:02:10] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[07:05:08] <buildmaster> pentium4/intellij-idea-community-edition is broken (says nlopc43).
[07:07:17] -!- abaumann has joined #archlinux32
[07:07:18] <buildmaster> Hi abaumann!
[07:07:18] <buildmaster> !rq abaumann
[07:07:18] <phrik> buildmaster: <abaumann> it usually automatically fails. :-)
[07:07:24] -!- deep42thought has joined #archlinux32
[07:07:24] <buildmaster> Hi deep42thought!
[07:07:24] <buildmaster> !rq deep42thought
[07:07:25] <phrik> buildmaster: <deep42thought> good night, python - cu tomorrow python2!
[07:07:28] <abaumann> hi deep42thought
[07:07:30] <deep42thought> Hi abaumann!
[07:07:33] <abaumann> faster :-)
[07:07:34] <deep42thought> I rebooted the buildmaster
[07:07:36] <deep42thought> (kernel update)
[07:07:42] <abaumann> yep. Seen that.
[07:08:00] <abaumann> I feel safe now with md_clear ;-)
[07:08:06] <deep42thought> :-D
[07:08:28] <abaumann> the 'causa recursively_umount_and_rm'
[07:08:35] <deep42thought> ah, yes
[07:08:49] <deep42thought> this is a horrible hack ... which does not work (anymore?)
[07:08:54] <abaumann> I can only think of 'sudo chown' the buildroot to give it to the build user before deleting it.
[07:09:06] <deep42thought> yeah, or sudo the find
[07:09:12] <abaumann> yep
[07:09:31] <deep42thought> I think, the *proper* way should be to fix archbuild
[07:09:33] <abaumann> the danger zone gets bigger.. but not as big as when running the build script as root.
[07:09:37] <deep42thought> but I'm not sure, why it's failing
[07:10:15] <abaumann> was there an update? I don't remember having updated devtools32 this week..
[07:10:31] <deep42thought> no
[07:10:34] <deep42thought> there is no update
[07:12:02] <abaumann> now with || true root artifacts just persist when cleaning the chroot, then the next rsync writes over them.
[07:12:22] <deep42thought> does "archbuild -c" work on that? or does it hickup?
[07:12:44] <abaumann> mmh. I have to check on eurobuild1.. there I had the errors.. just a sec
[07:17:32] <abaumann> staging-i486-build -c /var/lib/archbuild/staging-i486/
[07:17:34] <abaumann> worked fine.
[07:17:39] <deep42thought> ok
[07:17:43] <abaumann> but maybe the chroot was empty?
[07:17:52] <deep42thought> it should not be
[07:18:09] <deep42thought> e.g. "usually it is not"
[07:18:19] <abaumann> ah. another test..
[07:19:49] <abaumann> Deleting chroot copy 'build'...
[07:19:51] <abaumann> Deleting chroot copy 'root'...
[07:19:55] <abaumann> Creating install root at /var/lib/archbuild/staging-i486/root
[07:19:59] <abaumann> yep, looks fine
[07:20:08] <deep42thought> ok
[07:20:38] <deep42thought> but I guess, when my unexplainable error of left-over stuff happens, archbuild will fail and the hack will not change anything about it :-/
[07:20:54] <abaumann> most likely, yes.
[07:21:03] <abaumann> otoh, this is a detector for stupid packages.
[07:21:12] <deep42thought> hmm? how so?
[07:21:15] <abaumann> no package should ever create files/directories owned by root.
[07:21:23] <abaumann> only fakeroot during install
[07:21:23] <deep42thought> nah
[07:21:31] <deep42thought> it's the _build_ chroot
[07:21:37] <deep42thought> which has root-owned files
[07:21:49] <abaumann> the pkg
[07:21:51] <abaumann> not the src
[07:21:55] <abaumann> ah.
[07:21:57] <abaumann> now I get it.
[07:22:01] <abaumann> yeah. it's the chroot.
[07:22:02] <abaumann> :-)
[07:22:57] <abaumann> at least it will not fail on the same package over and over again.
[07:24:05] <abaumann> buildmaster: I 'git gc' the repos there. The root partition got up to 97% full.
[07:24:14] <deep42thought> O.o
[07:24:35] <abaumann> well. I can always make it bigger..
[07:24:41] <deep42thought> maybe we should add a reclone cronjob :-D
[07:24:52] <deep42thought> for git, not for /
[07:24:57] <abaumann> ah. :-)
[07:25:44] <abaumann> Ah. I changed the pacman CacheDir once, now it was back on /var/cache/pacman/pkg
[07:26:04] <deep42thought> maybe I missed that when merging a new pacman.conf :-/
[07:26:17] <abaumann> /dev/sdb2 32765712 20186676 10884932 65% /
[07:26:20] <abaumann> nice. :-)
[07:26:24] <abaumann> much better.
[07:26:33] <abaumann> aeh.. cause 'webkit2gtk'
[07:26:53] <abaumann> the problem is not so much the SSE2 patching.. the problem is Gnome/upstream sends down broken cmake/make files which fail to build or install..
[07:27:12] <abaumann> I'm not able to build it on 64-bit..
[07:27:14] <deep42thought> so this is an issue on pentium4, too?
[07:27:18] <abaumann> yes.
[07:27:18] <deep42thought> oh
[07:27:27] <deep42thought> well, do they know about it?
[07:27:43] <abaumann> Not yet. I have to do it once more, for reproducability purposes.
[07:27:49] <abaumann> but it takes several hours!
[07:28:18] <abaumann> it's modern software in the end.. it _has_ to compile that long in order to get good.. ;-)
[07:28:27] <deep42thought> !grab abaumann
[07:28:28] <phrik> deep42thought: 🎉
[07:28:56] <abaumann> I also have to discuss with my hoster about the definition of "two cores"..
[07:29:12] <deep42thought> good software is like a good meal: you cannot expect to get something really good if it's done in 5 minutes ...
[07:29:20] <abaumann> ..if they say fair use, I understand I can use 2 cores on the virtual machine for 24x7
[07:29:30] <abaumann> !grab deep42thought
[07:29:31] <phrik> abaumann: 🎉
[07:29:32] <deep42thought> you got a complaint from them?
[07:29:33] <abaumann> exactly.
[07:29:35] <abaumann> yep.
[07:29:48] <abaumann> but it was a mysql/borg mixup somehow.
[07:29:59] <abaumann> it's not happening with the buildmaster/mysqld alone.
[07:30:18] <abaumann> but sometimes I'm using the machine also to build a package, for testing purposes..
[07:30:24] <abaumann> ..like webkit2gtk :->
[07:31:17] <deep42thought> hmm, looks like we need more hardware :-/
[07:32:22] <abaumann> yeah. or clouds giving me what they promise.
[07:33:03] <abaumann> it's not the "hosting a webpage cheapo hoster", so I'm expecting more from them
[07:34:37] <deep42thought> ah, btw: I changed the layout of releng/archiso32 and releng/archiso-dual
[07:34:49] <deep42thought> they both depend on archiso now and can/should be installed in parallel
[07:34:56] <deep42thought> and only contain the /differences/
[07:35:22] <abaumann> did you put pentium4 or i686 on it?
[07:35:57] <deep42thought> I think, it has architecture=auto
[07:44:42] -!- titus_livius has joined #archlinux32
[07:45:02] <deep42thought> I know how to trigger the archbuild bug
[07:46:03] <deep42thought> on non-btrfs, run 'mount -o bind /run /var/lib/archbuild/staging-x86_64/root/run' to simulate a busy /run and then "staging-x86_64 -c" happily fails
[07:46:55] <abaumann> shouldn't archbuild try hard to unmount things in the chroot before trying to delete /run or /proc?
[07:47:18] <deep42thought> rm: skipping '/var/lib/archbuild/staging-x86_64/root/run', since it's on a different device
[07:47:19] <abaumann> but the build continues now..
[07:57:47] -!- oaken-source has quit [Quit: Lost terminal]
[08:23:29] -!- deep42thought has quit [Quit: Leaving.]
[08:24:24] -!- ztrawhcse has quit [Ping timeout: 272 seconds]
[08:45:43] -!- abaumann has quit [Quit: leaving]
[08:50:44] <buildmaster> pentium4/haskell-aeson-qq is broken (says eurobuild1).
[08:57:06] -!- ztrawhcse has joined #archlinux32
[09:39:00] <buildmaster> pentium4/opencv is broken (says buildknecht2).
[10:53:14] -!- ztrawhcse has quit [Ping timeout: 272 seconds]
[11:08:02] -!- ztrawhcse has joined #archlinux32
[12:09:39] -!- deep42thought has joined #archlinux32
[12:09:39] <buildmaster> Hi deep42thought!
[12:09:39] <buildmaster> !rq deep42thought
[12:09:40] <phrik> buildmaster: <deep42thought> *mimimi* tyzoid *mimimimi*
[12:39:34] <buildmaster> pentium4/nfs-utils are broken (says eurobuild1).
[12:48:54] <buildmaster> pentium4/js60 is broken (says buildknecht).
[12:52:54] -!- ztrawhcse has quit [Ping timeout: 258 seconds]
[13:06:56] -!- ztrawhcse has joined #archlinux32
[13:28:41] <buildmaster> pentium4/haskell-cassava-megaparsec is broken (says eurobuild1).
[13:32:28] <buildmaster> pentium4/idris are broken (says buildknecht).
[13:42:00] -!- thePiGrepper has joined #archlinux32
[13:46:11] -!- ofara_ has quit [Ping timeout: 258 seconds]
[13:48:19] -!- ofara_ has joined #archlinux32
[14:09:04] <buildmaster> pentium4/shellcheck is broken (says eurobuild1).
[14:11:21] <buildmaster> pentium4/virtualbox-modules-arch is broken (says buildknecht).
[14:21:38] <buildmaster> pentium4/broadcom-wl is broken (says buildknecht).
[14:36:39] <buildmaster> pentium4/wireguard-arch is broken (says eurobuild3).
[14:38:15] <buildmaster> pentium4/nvidia-390xx is broken (says buildknecht).
[14:53:16] -!- ztrawhcse has quit [Ping timeout: 272 seconds]
[15:07:31] -!- ztrawhcse has joined #archlinux32
[15:42:40] -!- ztrawhcse has quit [Ping timeout: 272 seconds]
[16:05:18] <buildmaster> pentium4/nfs-utils are broken (says rechenknecht).
[16:07:23] -!- deep42thought has quit [Quit: Leaving.]
[16:28:50] <buildmaster> i686/nfs-utils are broken (says eurobuild3).
[16:48:28] <buildmaster> pentium4/shellcheck is broken (says rechenknecht).
[17:24:00] -!- thePiGrepper has quit [Ping timeout: 272 seconds]
[17:30:38] <buildmaster> pentium4/startdde is broken (says rechenknecht).
[17:31:56] <buildmaster> pentium4/deepin-dock is broken (says eurobuild3).
[17:39:47] <buildmaster> i686/qt5-declarative is broken (says eurobuild3).
[17:53:22] <buildmaster> i686/webkit2gtk is broken (says rechenknecht).
[18:06:08] -!- MrBIOS has quit [Quit: MrBIOS]
[18:39:06] <buildmaster> i686/bat is broken (says eurobuild3).
[18:46:01] <buildmaster> i686/intel-graphics-compiler is broken (says rechenknecht).
[19:08:27] <buildmaster> pentium4/golang-golang-x-image is broken (says buildknecht).
[19:21:54] -!- MrBIOS has joined #archlinux32
[19:31:06] -!- MrBIOS has quit [Quit: MrBIOS]
[19:31:33] -!- MrBIOS has joined #archlinux32
[19:47:40] -!- oaken-source has joined #archlinux32
[20:02:52] -!- davor has quit [Read error: Connection reset by peer]
[20:06:06] -!- davor has joined #archlinux32
[20:13:34] -!- AndrevS has joined #archlinux32
[20:15:37] -!- DepositePirate has quit [Ping timeout: 256 seconds]
[20:17:30] -!- DepositePirate has joined #archlinux32
[20:58:55] <elibrokeit> Interesting, GCC 9 indicates that it reduces memory use for LTO by a significant margin.
[21:00:23] -!- MrBIOS_ has joined #archlinux32
[21:01:18] -!- MrBIOS has quit [Ping timeout: 245 seconds]
[21:01:18] MrBIOS_ is now known as MrBIOS
[21:08:49] <buildmaster> i686/thunderbird is broken (says eurobuild3).
[21:21:58] <buildmaster> i686/firefox is broken (says rechenknecht).
[22:34:24] -!- eschwartz has joined #archlinux32
[23:43:35] <buildmaster> i686/js60 is broken (says rechenknecht).