#archlinux32 | Logs for 2019-04-17

[00:13:44] <buildmaster> pentium4/deepin-picker is broken (says eurobuild3).
[00:21:35] <buildmaster> pentium4/deepin-clone is broken (says eurobuild3).
[00:48:48] <buildmaster> pentium4/wireshark is broken (says buildknecht).
[01:03:21] <buildmaster> pentium4/libffado is broken (says eurobuild3).
[01:16:59] -!- horus125 has joined #archlinux32
[01:18:37] <horus125> what graphical bootloaders/managers would you recommend for a 32bit legacy bios computer? I' d like to achieve something that looks like refind (horizontal icons)
[01:28:02] <buildmaster> pentium4/libsigrok is broken (says buildknecht).
[01:32:11] <buildmaster> pentium4/qjackctl is broken (says buildknecht).
[01:36:20] <buildmaster> pentium4/qsynth is broken (says buildknecht).
[02:12:22] -!- slacka123 has joined #archlinux32
[02:56:50] -!- thePiGrepper has joined #archlinux32
[04:16:50] <buildmaster> pentium4/deepin-image-viewer is broken (says buildknecht).
[04:21:30] <buildmaster> pentium4/deepin-network-utils are broken (says buildknecht).
[04:25:48] <bill-auger> horus125: you can make GRUB look like anything - you just make a theme for it
[04:30:36] <horus125> bill-auger: after a lot of searching and asking I managed to find one with horizontal icons (the matrix one) https://www.gnome-look.org/p/1115912/
[05:38:58] <buildmaster> pentium4/poppler is broken (says eurobuild3).
[05:45:01] <buildmaster> Hi deep42thought!
[05:45:01] <buildmaster> !rq deep42thought
[05:45:01] <phrik> buildmaster: because you're a pessimist with no hope - or you're a realist afraid of change ... pick your evil
[05:45:19] <deep42thought> elibrokeit: some (all?) of the build slaves seem unable to retrieve gpg keys
[05:45:24] <deep42thought> I have no idea, why
[05:45:41] <deep42thought> usually I download them manually then and transfer them to the slave in that case
[05:46:05] <deep42thought> additionally the logic to dl the key needs one failed build to trigger
[05:46:20] <deep42thought> and there is currently one slave running which only tries to build each package once :-)
[05:48:37] <elibrokeit> welp
[05:48:57] <elibrokeit> I guess you know how to fix that, then.
[05:49:18] <deep42thought> yeah, I just need to look at it ...
[05:51:19] <deep42thought> 2019-04-16 10:42:42: building package "libretro-blastem" (revisions 755fa96c2afd43c2f8df64ac3872d63fa37bdb8a 6de73e1920f4ab16e37602cace587a3d9084ed81, repository community, straw :) for i686 ...Username for 'https://github.com':
[05:51:20] <phrik> Title: The world’s leading software development platform · GitHub (at github.com)
[05:51:21] <deep42thought> wtf?
[05:53:40] <elibrokeit> github will try to ask for passwords if you use https:// sources, and the repo does not exist
[05:53:48] <elibrokeit> the theory is that it may be a private repo
[05:54:03] <elibrokeit> and you aren't allowed to know whether there is one or not
[05:54:32] <elibrokeit> so instead it always asks for a password and usually tells you you aren't authorized to look for private repos there
[05:55:37] <buildmaster> pentium4/deepin-dock is broken (says eurobuild3).
[05:55:54] <deep42thought> but libretro-blastem should not pull from a private repo, right?
[05:56:20] <elibrokeit> my naive suspicion is that upstream deleted that repo and replaced it conceptually with https://github.com/libretro/blastem
[05:56:47] <elibrokeit> either way the PKGBUILD is broken, because the repo it references is gone
[05:57:21] <deep42thought> do you want to open a bug or should I?
[06:00:48] <elibrokeit> you, please :)
[06:00:59] <deep42thought> ok
[06:02:16] <elibrokeit> I mean I highly doubt libretro is using their paid github plan (which I doubt is paid) to change the repo into a private repo... and if they did then the PKGBUILD still needs a proper source :p
[06:03:09] <deep42thought> :-D
[06:05:25] <deep42thought> !bug 62372
[06:05:26] <phrik> https://bugs.archlinux.org/task/62372
[06:09:08] <elibrokeit> assigned
[06:54:13] <buildmaster> i686/gnuradio is broken (says nlopc43-manual-bs).
[07:15:45] <buildmaster> i686/libosinfo is broken (says nlopc43-manual-bs).
[07:32:37] <buildmaster> i686/at-spi2-atk is broken (says eurobuild3).
[07:47:41] <buildmaster> i686/valabind is broken (says buildknecht).
[08:12:44] <buildmaster> i686/ruby-gio2 is broken (says buildknecht).
[08:12:54] <buildmaster> i686/ruby-gdk_pixbuf2 is broken (says nlopc43-manual-bs).
[08:14:28] <buildmaster> i686/ruby-pango is broken (says nlopc43-manual-bs).
[08:18:14] <buildmaster> Hi deep42thought!
[08:18:14] <buildmaster> !rq deep42thought
[08:18:16] <phrik> buildmaster: my cluster of 1k 4/86's will be happy to iterate Maxwell-Vlasov-equations :-D
[08:27:07] -!- horus125 has quit [horus125 has left #archlinux32]
[08:30:32] -!- titus_livius has joined #archlinux32
[08:49:15] <buildmaster> i686/connman is broken (says eurobuild3).
[08:52:43] <buildmaster> i686/user-manager is broken (says eurobuild3).
[08:54:25] <deep42thought> slacka123: to answer your question: there was no major rebuild, we merely copied over all the packages
[08:54:36] <deep42thought> the "re"build will only happen for new packages
[08:55:49] -!- DCyrax has joined #archlinux32
[09:15:57] <buildmaster> i686/unarchiver is broken (says nlopc43-manual-bs).
[09:16:18] -!- quequotion has joined #archlinux32
[09:35:18] <buildmaster> i686/libsoup is broken (says buildknecht).
[09:46:09] -!- quequotion has quit [Quit: quequotion]
[10:03:21] -!- DCyrax has quit [Remote host closed the connection]
[10:05:01] -!- bill-auger has quit [Ping timeout: 252 seconds]
[10:05:20] -!- bill-auger has joined #archlinux32
[10:07:41] -!- AndrevS has joined #archlinux32
[10:11:27] -!- oaken-source has quit [Quit: Lost terminal]
[10:19:08] -!- DCyrax has joined #archlinux32
[10:22:10] <buildmaster> pentium4/virtualbox is broken (says nlopc43-manual-bs).
[10:24:00] <buildmaster> i686/deepin-dock is broken (says nlopc43-manual-bs).
[10:24:40] <buildmaster> pentium4/virtualbox-modules-arch is broken (says eurobuild3).
[10:29:06] <buildmaster> pentium4/teamspeak3-server is broken (says eurobuild3).
[10:29:33] <deep42thought> oops
[10:29:48] <deep42thought> this one comes with a sha256sums_i686 but not a sha256sums_pentium4 :-D
[10:33:39] <buildmaster> pentium4/libretro-blastem is broken (says eurobuild3).
[11:01:33] -!- oaken-source has joined #archlinux32
[11:01:46] -!- oaken-source has quit [Client Quit]
[11:02:12] -!- oaken-source has joined #archlinux32
[11:23:43] <buildmaster> i686/glibc is broken (says nlopc43-manual-bs).
[11:37:06] <deep42thought> how can this be:
[11:37:08] <deep42thought> rm: skipping '/var/lib/archbuild/staging-i686/root', since it's on a different device
[11:37:10] <deep42thought> but
[11:37:20] <deep42thought> umount: /var/lib/archbuild/staging-i686/root: not mounted.
[12:08:10] <deep42thought> ah, apparently it's a btrfs subvolume
[12:08:39] * deep42thought wasn't aware, that this build slave has btrfs
[13:37:51] -!- thePiGrepper has joined #archlinux32
[13:54:20] -!- thePiGrepper has quit [Read error: Connection timed out]
[13:54:38] -!- thePiGrepper has joined #archlinux32
[15:56:26] -!- deep42thought has quit [Quit: Leaving.]
[18:09:22] -!- rcf has quit [Quit: WeeChat 2.4]
[18:10:31] -!- rcf has joined #archlinux32
[18:12:03] -!- rcf has quit [Client Quit]
[18:12:34] -!- rcf has joined #archlinux32
[18:56:44] -!- isacdaavid has joined #archlinux32
[19:04:42] -!- samantaz has joined #archlinux32
[19:36:38] -!- abaumann has joined #archlinux32
[19:36:38] <buildmaster> Hi abaumann!
[19:36:38] <buildmaster> !rq abaumann
[19:36:40] <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..
[19:38:51] -!- MrBIOS has joined #archlinux32
[19:55:33] <abaumann> restart my slaves, they insisted there are no more packages to build.. :-)
[19:55:38] -!- abaumann has quit [Quit: leaving]
[20:11:26] -!- deep42thought has joined #archlinux32
[20:11:26] <buildmaster> Hi deep42thought!
[20:11:26] <buildmaster> !rq deep42thought
[20:11:27] <phrik> buildmaster: <deep42thought> it was too slow - they don't offer that speed anymore, so they rounded down to 0
[20:11:37] <deep42thought> abaumann: glibc was blocking some stuff
[20:11:42] <deep42thought> but that should be done, now ...
[20:14:19] <deep42thought> now it's gcc :-)
[20:15:54] <deep42thought> explanation: I updated a bunch of PKGBUILDs to do the same (or similar stuff) on pentium4 as they do on i686
[20:16:03] <deep42thought> so a lot of packages were scheduled for rebuild
[20:18:32] -!- samantaz_ has joined #archlinux32
[20:20:49] -!- samantaz has quit [Ping timeout: 246 seconds]
[20:23:27] -!- abaumann has joined #archlinux32
[20:23:27] <buildmaster> Hi abaumann!
[20:23:27] <buildmaster> !rq abaumann
[20:23:28] <phrik> buildmaster: <abaumann> singularity applied to software development - releases and changes get faster and faster.. till..
[20:23:34] <abaumann> deep42thought: ah. yes. glibc
[20:23:35] <deep42thought> Hi abaumann!
[20:23:42] <abaumann> hi deep42thought
[20:23:56] <abaumann> gcc is on my i486 slave. :-)
[20:23:59] <deep42thought> though, something seems wrong, as gcc on pentium4 currently blocks glibc on i686 or something like that :-D
[20:24:04] <deep42thought> that, too
[20:24:09] <abaumann> mmh
[20:24:11] <deep42thought> but that should be unrelated to i686 and pentium4
[20:24:32] <deep42thought> the problem with i686/pentium4 /might/ be, that both are built on the same clients
[20:25:01] <abaumann> thus blocking.. could be..
[20:25:08] <abaumann> you are the (build)master. ;-)
[20:25:15] <deep42thought> !grab abaumann
[20:25:16] <phrik> deep42thought: Tada!
[20:25:47] <deep42thought> that's too much honor - I just /try/ to be the build master and kick the real build master's ass
[20:25:56] <deep42thought> but besides that I'm not mastering anything
[20:26:25] <abaumann> buildmaster: kick ass
[20:26:36] <abaumann> mmh.. :-)
[20:26:40] <buildmaster> ouch
[20:26:47] <abaumann> there you go. :-)
[20:26:56] <deep42thought> ah, just another thing: we now have https://git3.archlinux32.org
[20:27:07] <deep42thought> which is a reverse proxy to git.archlinux32.org
[20:27:13] <deep42thought> to circumvent the broken cert
[20:27:17] <abaumann> oh my..
[20:27:25] <abaumann> nice one. :-)
[20:27:33] <deep42thought> well, but it's triple the trafic :-D
[20:27:49] <deep42thought> tyzoid's server <-> my server <-> client
[20:27:58] <deep42thought> (all https, of course)
[20:27:58] <abaumann> no chances to reach tyzoid?
[20:28:07] <deep42thought> haven't reached him so far
[20:28:19] <deep42thought> my hope was for after easter
[20:28:22] <abaumann> he might be on a safari on the North Pole..
[20:28:38] <abaumann> ..no internet there.. ;-)
[20:28:42] <deep42thought> maybe he is fasting computers/internet/archlinux32?
[20:28:53] <abaumann> :-)
[20:29:20] <abaumann> super. i can log in to gitea :-)
[20:30:19] <deep42thought> that was the idea
[20:31:36] <abaumann> "run opcode sniffer for i686 (and i486) on regular basis", yeah. don't like my script approach.. a little (aka very) too slow
[20:31:56] <deep42thought> let it be run by the build slaves
[20:31:58] <deep42thought> and we're fine
[20:32:16] <deep42thought> they need to have the extracted package anyways
[20:32:24] <deep42thought> so just hack the package() function and we're good
[20:32:37] <abaumann> could work..
[20:33:03] <deep42thought> this could end up as a new namcap test, too
[20:33:34] <deep42thought> "namcap ERROR: binary with sse2 in i686 package"
[20:33:44] <deep42thought> similar to the "efi in any-package" error
[20:33:50] <deep42thought> *elf
[20:33:53] <deep42thought> ... not efi :-D
[20:34:18] <abaumann> yep. also an option.. :-)
[20:34:27] <deep42thought> would be more flexible
[20:34:35] <abaumann> we should make a .. not bios ;-)
[20:34:37] <deep42thought> e.g. we could also test manually
[20:35:04] <abaumann> I plan to set up X and stuff on one of my real P3 without SSE3.
[20:35:26] <abaumann> The only trouble is, that the Matrox graphic card driver seems pretty broken in X
[20:35:38] <deep42thought> hmm
[20:35:45] <abaumann> mmh. maybe easier on the Alix with Geode
[20:40:41] <deep42thought> my alix is currently sometimes running out of memory when I run apt-get upgrade
[20:40:54] <deep42thought> ... not a really nice experience
[20:43:01] <abaumann> apt-get running out of memory.. well, there is some room to "improve" memory usage of pacman :-)
[20:43:21] <abaumann> pacman is really efficient.. even when installing on 486 with 64 MB RAM.
[20:43:38] <deep42thought> yeah, I expect pacman to be more efficient (in terms of memory and speed)
[20:43:51] <abaumann> one of the best. I think.
[20:43:55] <deep42thought> otoh apt-get offers really nice, elaborate solutions for conflicts
[20:44:00] <abaumann> the voidlinux one might be comparable.
[20:48:20] <KitsuWhooa> <abaumann> The only trouble is, that the Matrox graphic card driver seems pretty broken in X <-- I have a matrox card and last I checked things were okay. Maybe try turning off acceleration :p
[20:50:33] <deep42thought> abaumann: the build master is running fine? It did not buckle under the additional load of another architecture so far?
[20:51:43] <abaumann> KitsuWhooa: oh! that's a good point. I'll try..
[20:51:54] <KitsuWhooa> I think it's NoAccel
[20:51:57] <KitsuWhooa> I can't remember
[20:52:12] <abaumann> deep42thought: yeah. the buildmaster is tucking along fine. :-)
[20:52:30] <abaumann> the usual 2x100% CPU, but otherwise.. :-)
[20:52:35] <deep42thought> umm
[20:52:48] <abaumann> as in old mainframe days: we spent a million on that CPU, it's better up to 100% usage..
[20:53:06] <deep42thought> similar with the laser in the lab
[20:53:30] <deep42thought> "one day running that thing costs 10k, one day not running it costs 10k, too - so just shoot on anything"
[20:54:00] <abaumann> wow. that's an expensive toy. :-)
[20:54:22] <deep42thought> well, maybe the number is slightly to high, but the order of magnitude should be around there
[20:55:43] <deep42thought> my supervisor when I made my first shot with the laser: He handed me the remote and said: "Hey Erich, want to destroy a 20 Mio€ device?"
[20:56:07] <abaumann> !grab deep42thought
[20:56:08] <phrik> abaumann: Tada!
[20:56:15] <abaumann> LOL
[20:58:47] <abaumann> 01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Rage 4 [Rage Fury/Xpert 128/Xpert 2000 AGP]
[20:58:56] <abaumann> I should try not to start a Matrox Xorg driver here.
[20:59:08] <abaumann> Somehow I forgot I changed the video card.. :-)
[21:00:06] <deep42thought> maybe you were frustrated with the broken driver on the old one?
[21:00:37] <abaumann> yeah. I think so.
[21:14:29] <abaumann> "Input not supported".. I really love old VGA-only LCD displays and Xorg..
[21:14:50] <abaumann> They are so old, you cannot even find the frequency parameters for your Xorg.conf
[21:15:16] <abaumann> The 'r128' driver seems - well - to be xorg 1.19 and older only..
[21:19:53] -!- abaumann has quit [Quit: leaving]
[21:22:43] <KitsuWhooa> <abaumann> "Input not supported".. I really love old VGA-only LCD displays and Xorg.. <-- Most VGA monitors should support DDC
[21:22:53] <KitsuWhooa> so that the computer can read the EDID
[21:23:22] <KitsuWhooa> some manufacturers though simply didn't bother putting pins for those signals in the connectors and wiring them, so the computer has no way to talk to the monitor to get the data
[21:23:27] <KitsuWhooa> in cables, I mean
[21:24:03] <KitsuWhooa> and then there are those DVI-I to VGA passive adapters that have the pins there but don't even bother wiring the signals and make me pull my hair out
[21:39:28] -!- abaumann has joined #archlinux32
[21:39:29] <buildmaster> Hi abaumann!
[21:39:29] <buildmaster> !rq abaumann
[21:39:29] <phrik> buildmaster: <abaumann> We should really not build a package more than 99 times. ;-)
[21:39:48] <abaumann> KitsuWhooa: yeah. now it works. I used some very basic VESA timing settings in xorg.conf
[21:39:56] <abaumann> and i used the xf86-video-vesa driver
[21:40:29] <abaumann> all I want to do on that machine is to test whether i686 packages are working without SSE2
[21:40:39] <KitsuWhooa> ah
[21:41:10] <abaumann> Pentium III (Coppermine), just mmx and sse :-)
[21:43:00] <KitsuWhooa> yup
[21:43:30] <KitsuWhooa> This reminds me, I found out why I couldn't use speedstep-smi. The BIOS didn't support it
[21:43:42] <KitsuWhooa> Turns out toshiba did their own thing
[21:44:06] <KitsuWhooa> on the bright side, with my adventure I got TV out working using the same thing :p
[21:44:51] <abaumann> cool :-)
[21:46:13] <abaumann> I really like tinkering on those old machines.. especially, when they run the newest software again.. :-)
[21:50:23] <abaumann> mmh. on the positive side firefox is not crashing with SIGILL.. on the negative side it's a SEGSEGV in mozalloc_abort, gtk_css_provider_load_from_file
[21:50:26] <abaumann> nice.
[21:50:44] <abaumann> I'll use dillo meanwhile.. :-)
[21:56:33] <abaumann> ah. seamonkey is working perfectly. :-)
[21:56:39] <abaumann> so. enough for today..
[21:56:40] -!- abaumann has quit [Quit: leaving]
[22:19:49] <elibrokeit> deep42thought: you haven't seen conflict resolution until you've seen Gentoo conflict resolution spanning multiple years of updates as a flow chart :p
[23:07:57] -!- AndrevS has quit [Remote host closed the connection]