#archlinux32 | Logs for 2019-06-18

[00:22:45] -!- MrBIOS has quit [Quit: MrBIOS]
[01:02:29] -!- traace has quit [Quit: Leaving]
[01:02:32] -!- ofara has quit [Ping timeout: 245 seconds]
[01:10:27] -!- guys has quit [Ping timeout: 245 seconds]
[01:42:59] -!- samantaz has quit [Ping timeout: 258 seconds]
[01:44:44] -!- samantaz has joined #archlinux32
[02:14:44] <buildmaster> i686/ripgrep is broken (says rechenknecht).
[02:19:55] -!- ofara has joined #archlinux32
[02:33:46] -!- samantaz has quit [Ping timeout: 246 seconds]
[03:30:15] -!- guys has joined #archlinux32
[03:49:12] -!- guys has quit [Read error: Connection reset by peer]
[03:50:25] -!- guys has joined #archlinux32
[04:15:32] -!- guys has quit [Read error: Connection reset by peer]
[04:16:36] -!- guys has joined #archlinux32
[04:37:45] -!- guys has quit [Read error: Connection reset by peer]
[04:38:48] -!- guys has joined #archlinux32
[05:00:07] -!- guys has quit [Read error: Connection reset by peer]
[05:01:00] -!- guys has joined #archlinux32
[05:05:27] -!- guys has quit [Ping timeout: 245 seconds]
[05:14:14] -!- MrBIOS has joined #archlinux32
[05:21:20] -!- quequotion has quit [Quit: quequotion]
[07:06:37] -!- abaumann has joined #archlinux32
[07:06:38] <buildmaster> Hi abaumann!
[07:06:38] <buildmaster> !rq abaumann
[07:06:38] <phrik> buildmaster: <abaumann> I'm all for an docdepends upstreams and a doc hook, in order to reduce all those silly dependencies just to get documentation nobody reads afterwards, because they don't know about the 'man' command.. (rant of the day)
[07:06:45] -!- abaumann has quit [Client Quit]
[07:10:11] -!- abaumann has joined #archlinux32
[07:10:11] <buildmaster> Hi abaumann!
[07:10:11] <buildmaster> !rq abaumann
[07:10:12] <phrik> buildmaster: <abaumann> I start to get allergic shocks just by having to deal with rust.. :-)
[07:10:33] <abaumann> testing the thermal regime of the new server.. a hot summer day is ideal.. :-)
[07:30:23] <buildmaster> pentium4/java8-openjdk is broken (says nlopc43).
[07:39:01] <buildmaster> pentium4/java10-openjdk is broken (says rechenknecht).
[07:40:36] -!- deep42thought has joined #archlinux32
[07:40:36] <buildmaster> Hi deep42thought!
[07:40:36] <buildmaster> !rq deep42thought
[07:40:37] <phrik> buildmaster: <deep42thought> brtln: you're "the arch linux", we're "the other arch linux" and alarm is "arches linux", then?
[07:45:46] <deep42thought> abaumann: you could harden the conditions even further by putting the server into the other (closed) cabinet ;-)
[08:19:04] <buildmaster> pentium4/java-openjdk is broken (says nlopc43).
[08:24:24] <buildmaster> pentium4/java8-openjdk is broken (says eurobuild3).
[08:31:45] <abaumann> deep42thought: I want to test the machine, not kill it. :-)
[08:31:52] <deep42thought> !grab abaumann
[08:31:52] <phrik> deep42thought: Tada!
[08:31:56] <deep42thought> hmm, good point
[08:32:31] <abaumann> the psu is super silent, the case also, the CPU cooler is big and also good
[08:33:03] <deep42thought> nice :-)
[08:33:04] <abaumann> the case and the motherboard have some disagreement on some front connectors, but I plan to install the power switch only.
[08:39:47] <abaumann> *abaumann considers filming his computer build titled, "tech-professional building a modern computer after having built one 15 years ago"..
[08:40:31] <deep42thought> you wanna sell this as an educational, entertainment or horror movie?
[08:40:46] <abaumann> good question, last, I presume.
[08:41:14] <deep42thought> audience: "oh, I didn't know how to do this - now I does" vs. "haha, look at that fool" vs. "oh my god - nooooo"
[08:41:17] <abaumann> you forgot a class: humour
[08:41:29] <deep42thought> that's what I meant with "entertainment"
[08:41:34] <abaumann> :-)
[08:41:52] <abaumann> well, I used the chassis screws to mount the motherboard (fixed now).
[08:42:08] <abaumann> and I had to attach one fan with some straps, because it doesn't fit the case.
[08:42:15] <abaumann> so - entertainment. :-)
[08:42:19] <deep42thought> :-)
[08:42:21] <abaumann> but the thing works.
[08:42:42] <abaumann> I'll do some stress testing with kernel compilation and such today
[08:45:29] <deep42thought> you can do stress testing with setting up real build slaves, too ;-)
[08:45:46] <deep42thought> well, probably not with setting them up, but with actually running them
[08:49:54] <abaumann> that's true.
[08:50:17] <abaumann> I'm not sure yet, whether I should virtualize some of them, in order to give them "real" i486, i686, pentium4
[08:50:31] <abaumann> i486, yes, those I must.
[08:50:51] <deep42thought> I think, real i686/pentium4 only makes sense for testing purposes
[08:50:54] <deep42thought> chroots are much faster
[08:50:59] <abaumann> yep
[08:51:18] <deep42thought> otoh you need to virtualize anyways in order to run multiple build slaves, so ...
[08:51:24] <abaumann> so, it should be enough to create 4-8 users 'slave1' to 'slaveN', right?
[08:52:04] <abaumann> ah. that's from the chroot-staging-script-point-of-view, I'm talking of
[08:52:16] <abaumann> separater slave directories, one per user?
[08:52:20] <abaumann> this should actually work.
[08:52:24] <abaumann> or - containers.
[08:52:32] <deep42thought> separate users work as long as you don't invoke "-c"
[08:52:38] <abaumann> mmh. chroots in containers..
[08:52:40] <abaumann> maybe not :-)
[08:52:55] <deep42thought> rechenknecht is a container on the master mirror
[08:52:59] <deep42thought> so that works :-)
[08:53:06] <abaumann> ah :-)
[08:53:13] <deep42thought> though, it's only one
[08:53:38] <deep42thought> maybe it's easier to hack some configuration, so it uses different build chroots (also root/) for each slave on the bare metal host
[08:54:51] <abaumann> yeah, I'll try. But first I have to make sure the machine is stable under load and is silent enough under load.
[08:55:01] <deep42thought> ok
[09:13:39] <abaumann> thought my hard drive is too fat to fit the bay, till I realized, I used the wrong screws. :-)
[09:13:53] <abaumann> *fat
[09:14:12] <abaumann> *the fit is fit and the fat is fat
[09:14:34] <abaumann> * abaumann thinks it's getting too hot again to do brain work
[09:14:41] * deep42thought is confused
[09:20:33] -!- quequotion has joined #archlinux32
[09:30:22] <deep42thought> ah, I know what the problem with perl-data-optlist is - I managed to get my IP blacklisted by cpan :-D
[09:34:59] <abaumann> oups.
[09:38:42] -!- bill-auger has quit [Ping timeout: 258 seconds]
[09:42:38] -!- bill-auger has joined #archlinux32
[11:15:28] -!- abaumann has quit [Ping timeout: 245 seconds]
[11:26:08] <buildmaster> pentium4/java8-openjdk is broken (says rechenknecht).
[11:29:18] -!- thePiGrepper has joined #archlinux32
[11:41:49] -!- abaumann has joined #archlinux32
[11:41:49] <buildmaster> Hi abaumann!
[11:41:49] <buildmaster> !rq abaumann
[11:41:49] <phrik> buildmaster: <abaumann> doesn't ## uncomment the #? ;-)
[11:42:11] -!- mrkiko has quit [Quit: Leaving ...]
[11:43:07] <buildmaster> pentium4/java10-openjdk is broken (says rechenknecht).
[12:08:00] -!- djmoch has quit [Quit: ZNC 1.7.3 - https://znc.in]
[12:11:20] -!- djmoch has joined #archlinux32
[12:25:18] <buildmaster> pentium4/glibc is broken (says nlopc43-manual-bs).
[12:41:58] -!- djmoch has quit [Read error: Connection reset by peer]
[12:43:55] -!- djmoch has joined #archlinux32
[13:26:04] <abaumann> ah. a broken glibc and a broken java on pentium4. now things get interesting. :-)
[13:26:12] <deep42thought> glibc was my fault
[13:26:27] <deep42thought> I added the wrong proxy to my (blacklisted) build slave
[13:27:05] <abaumann> ah. then CPAN will block your proxy..
[13:27:13] <deep42thought> it's tor
[13:35:09] <abaumann> *abaumann thinks tcr doesn't stand for "The Colbert Show"..
[13:35:20] <abaumann> ah tor, not tcr
[13:35:29] <deep42thought> :-D
[13:35:37] <abaumann> I had some dirt on my screen
[13:35:41] <deep42thought> lol
[13:35:54] <deep42thought> you encrypted your screen
[13:36:11] <abaumann> question of the day: can I reboot an installed system remotely and it will actually work on the first try..
[13:36:32] <deep42thought> you enabled the network and sshd services?
[13:36:41] <deep42thought> (in contrast to only "start"ing them)
[13:36:54] <deep42thought> ... that's usually why it does not work when I try
[13:37:24] <abaumann> yeh, I did, but I don't remember if I set the USB stick to priority boot.
[13:37:41] <abaumann> and that's the only problem with a gamer motherboard, there is no serial on a remote console..
[13:37:47] <abaumann> ..at least I didn't find any.
[13:38:58] <abaumann> grub-mkconfig seems to have a bug: if you add 'root=/dev/md/root' it also adds a 'root=/dev/md127' additionally.
[13:40:48] -!- MrBIOS has quit [Ping timeout: 248 seconds]
[13:41:38] <deep42thought> abaumann: https://archlinux32.org now comes with a "Blocks" column which counts how many packages are being blocked by each build order
[13:42:02] <deep42thought> it's inaccurate in several ways:
[13:42:18] <deep42thought> 1st, it's updated asynchronously - because it's really slow :-(
[13:43:03] <deep42thought> 2nd, I have not thought about what means "block" too much - I currently just look at all depends=() stuff and ignore makedepends=() totally (in contrast to how get-assignment works)
[13:43:25] <abaumann> very handy.
[13:43:29] <deep42thought> 3rd, I did not think about architectures too much
[13:43:49] <deep42thought> but anyways, it should give a rough idea, what is at the bottom of everything :-)
[13:43:51] <abaumann> on the performace. I should maybe once have a look at the Mysql statements, maybe I see something which can be improved.
[13:44:13] <deep42thought> if you look at them, you will _for_sure_ see something :-D
[13:50:55] <deep42thought> oh, and I have the dependencies reversed O.o
[13:51:03] * deep42thought fixes this stupid bug right away
[14:18:23] <abaumann> the list looks nice. :-)
[14:18:50] <abaumann> and things like the toolchain, mesa, librsvg, then gtk3, qt5-base are on top if ordered by 'Blocks' descending
[14:18:56] <abaumann> so, this sounds right
[14:25:58] <abaumann> ln -s /dev/null /etc/udev/rules.d/80-net-setup-link.rules
[14:26:11] <abaumann> this makes booting a device with network configuration so much more predicable :-)
[14:26:31] <abaumann> if I have on NIC, then eth0 is really a good name for it.
[14:26:47] <abaumann> There is of course '*' in Match..
[14:27:00] <abaumann> *abaumann is very happy the machine actually booted
[14:28:22] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[14:40:00] <abaumann> b
[14:40:06] <abaumann> oups.
[14:45:02] -!- abaumann has quit [Quit: leaving]
[14:58:50] -!- traace has joined #archlinux32
[15:00:47] -!- abaumann has joined #archlinux32
[15:00:47] <buildmaster> Hi abaumann!
[15:00:47] <buildmaster> !rq abaumann
[15:00:47] <phrik> buildmaster: <abaumann> wait... go-go-gadgeto-i686-test-machine
[15:06:58] <deep42thought> ah, damn, the bug that gcc, binutils and glibc block each other cross-architecture popped up again ...
[15:07:41] <deep42thought> e.g. https://www.archlinux32.org
[15:07:42] <phrik> Title: Todos in the build scripts (at www.archlinux32.org)
[15:48:16] -!- MrBIOS has joined #archlinux32
[15:51:17] -!- guys has joined #archlinux32
[16:13:56] -!- deep42thought has quit [Quit: Leaving.]
[16:15:23] -!- abaumann has quit [Remote host closed the connection]
[17:44:50] -!- slacka123 has quit [Ping timeout: 252 seconds]
[17:48:46] -!- slacka123 has joined #archlinux32
[17:54:20] -!- traace has quit [Quit: Leaving]
[18:17:49] -!- samantaz has joined #archlinux32
[19:16:47] -!- slacka123 has quit [Quit: Leaving]
[19:18:27] -!- slacka123 has joined #archlinux32
[19:43:31] <buildmaster> i686/gcc is broken (says nlopc43-manual-bs).
[20:03:08] -!- buildmaster has quit [Remote host closed the connection]
[20:03:49] -!- buildmaster has joined #archlinux32
[20:03:49] <buildmaster> !rq buildmaster
[20:03:49] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[20:19:29] <buildmaster> pentium4/java11-openjdk is broken (says nlopc43).
[20:29:49] <slacka123> could we change the build flags of pentium4/java11-openjdk to match i686? That would fix FS#75
[20:49:33] -!- thePiGrepper has joined #archlinux32
[21:23:37] <girls> slacka123: if I understand the issue correctly, we would need to change the build flags of glibc - effectively making i686 and pentium4 identical for almost all use cases ...
[21:38:39] -!- phrik has quit [Read error: Connection reset by peer]
[21:40:39] -!- phrik has joined #archlinux32
[21:40:39] <buildmaster> Hi phrik!
[21:40:39] <buildmaster> !rq phrik
[21:43:35] <buildmaster> pentium4/python-wrapt is broken (says nlopc43).
[21:44:15] <phrik> buildmaster: <phrik> i am bot
[21:56:28] -!- thePiGrepper has quit [Ping timeout: 272 seconds]
[21:57:18] <buildmaster> pentium4/nfs-utils are broken (says rechenknecht).
[22:10:08] <buildmaster> pentium4/python-sqlalchemy is broken (says nlopc43).
[22:19:13] -!- Cthulu201 has quit [Ping timeout: 246 seconds]
[22:19:49] -!- Cthulu201 has joined #archlinux32
[22:23:43] <buildmaster> pentium4/virtualbox-modules-arch is broken (says nlopc43).
[22:31:43] -!- ofara_ has quit [Ping timeout: 245 seconds]
[22:36:07] <buildmaster> pentium4/python-awkward is broken (says nlopc43).
[22:49:35] -!- traace has joined #archlinux32
[22:55:13] -!- ofara_ has joined #archlinux32
[23:08:45] -!- MrBIOS has quit [Quit: MrBIOS]
[23:10:50] <buildmaster> pentium4/grafana is broken (says nlopc43).
[23:41:49] <slacka123> girls: then maybe we should use "-march=i686 -msse2 -mtune=generic -mfpmath=sse -mstackrealign"
[23:42:17] <slacka123> I just checked Java on 32-bit Fedora 30 and it works along with ant
[23:42:51] <slacka123> https://fedoraproject.org
[23:42:52] <phrik> Title: Changes/Update i686 architectural baseline to include SSE2 - Fedora Project Wiki (at fedoraproject.org)