#archlinux32 | Logs for 2019-08-29

Back
[00:00:13] <GNUtoo> hi, is there some documentation on the tradeoffs done in community-staging and community?
[00:00:42] <girls> GNUtoo: what kind of tradeoffs do you have in mind?
[00:00:49] <GNUtoo> for instance community-staging looks closer to arch's community, but is the -staging stable enough?
[00:01:18] <GNUtoo> (I wonder which one I should use as repos like community are often out of sync with arch)
[00:01:42] <GNUtoo> by stable I mean that ldd <binary> will not have "not found" in it
[00:01:53] <GNUtoo> and that the thing is consistent
[00:02:50] <girls> testing is a good choice for that
[00:03:07] <GNUtoo> ok thanks
[00:03:16] <girls> upstream's stable packages are built and put into staging
[00:03:37] <girls> and then those get moved to testing - intended with all their libraries (we do make some mistakes sometimes)
[00:03:53] <girls> and when they get tested and approved, they are similarily moved to stable
[00:04:21] <GNUtoo> ok thanks a lot
[00:04:32] <girls> if you choose to run testing, you can also volunteer for reporting tested packages
[00:04:37] <girls> it can be automated :-)
[00:08:29] <GNUtoo> That would be nice, however it can't be automated in my case as I've to make sure that everything really comes from arch32 before reporting
[00:10:58] <girls> that's no problem, we send a hash of the package along with the report
[00:11:04] <girls> so we don't have false positives :-)
[00:11:14] <GNUtoo> does that take into account the dependencies?
[00:11:29] <girls> no, you're right
[00:11:39] <girls> that won't correctly handle foreign dependencies
[00:12:10] <buildmaster> i686/vamp-aubio-plugins are broken (says eurobuild6-6): https://archlinux32.org
[00:12:30] <GNUtoo> ok, it might still be nice to do it from the command line once I've asserted that the cause is 100% from arch32 and not other dependencies
[00:12:44] <GNUtoo> is there some documentation on doing that?
[00:12:48] <girls> yes
[00:13:06] <girls> https://bbs.archlinux32.org
[00:13:07] <phrik> Title: automatic testing via manual installation(s) / Testing / Arch Linux 32 Forums (at bbs.archlinux32.org)
[00:13:11] <girls> hopefully, it's stil up-to-date
[00:13:19] <GNUtoo> thanks
[00:20:28] <GNUtoo> for some reason https://git.archlinux32.org doesn't seem to respond
[00:20:30] <phrik> Title: archlinux32 - Archlinux32 Gitea (at git.archlinux32.org)
[00:20:57] <GNUtoo> ah now it does
[00:21:31] <GNUtoo> (bad exit node)
[00:22:48] <girls> if you only want to report a couple of packages (in contrast to "all packages") you will rip apart report-installed-packages and do it's job manually (by specifying the packages or so)
[00:23:03] <girls> s/will rip/will want to rip/
[00:24:15] <GNUtoo> I don't understand how it works
[00:24:47] <GNUtoo> When reading various source code I only saw it sending the list of packages
[00:25:08] <girls> yes, that's the part, which the tester does
[00:25:18] <GNUtoo> let's say a binary from a package from arch32 doesn't work due to a library not found
[00:25:22] <girls> the build master then sets some flags and moves packages appropriately
[00:25:26] <GNUtoo> or that stuff are out of sync somehow
[00:25:40] <GNUtoo> in both cases I'd like to report a bug
[00:25:47] <girls> then you open a bug and the buildmaster will consider the package "broken"
[00:25:57] <girls> that part is not automatized
[00:25:58] <GNUtoo> oh ok thanks
[02:14:16] -!- MrBIOS has quit [Quit: MrBIOS]
[02:17:58] -!- KitsuWhooa has quit [Quit: Unable to handle kernel NULL pointer dereference at null]
[02:22:33] -!- KitsuWhooa has joined #archlinux32
[02:49:44] eschwartz is now known as guys
[02:52:01] guys is now known as eschwartz
[06:32:59] -!- davor has quit [Read error: Connection reset by peer]
[06:36:41] -!- davor has joined #archlinux32
[07:20:45] <buildmaster> i686/chromium is broken (says eurobuild6-1): https://archlinux32.org
[07:30:37] <buildmaster> pentium4/chromium is broken (says eurobuild6-5): https://archlinux32.org
[07:37:25] -!- abaumann has joined #archlinux32
[07:37:26] <buildmaster> Hi abaumann!
[07:37:26] <buildmaster> !rq abaumann
[07:37:27] <phrik> buildmaster: <abaumann> The future looks bright, no window managers working, Xorg not working, sound and video stuff "optimized", so not working. Gives us a nice text-only distribution. :-)
[07:38:56] <abaumann> deep42thought: I start to see a plateau again in the builds, I have to see, what are the biggest blockers now.
[07:39:05] <abaumann> Also I saw errors like: https://sources.archlinux32.org
[07:39:11] <abaumann> 404 not found
[07:40:27] <abaumann> what could be the reason nothing gets from staging to testing? Can I selectively db-push -p only from staging to testing, I think not..
[07:40:44] <abaumann> ..there must be a reason, why this doesn't happen automatically..
[07:59:11] <girls> the 404 is uncritical
[07:59:23] <girls> it's just a last resort trial which is failing
[08:00:35] <girls> the -p only for staging -> testing is a good idea
[08:00:44] <girls> I'll look into writing that
[08:01:20] <girls> otoh removing the blockers might be more benefitial - otherwise we face similar problems for testing -> stable
[08:03:19] <girls> it might also be old (=deleted upstream) packages in [testing] which block moving of staging->testing
[08:05:27] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[08:06:08] <abaumann> girls: thanks. I'll have a look at the blocking packages..
[08:10:41] -!- abaumann has quit [Quit: leaving]
[08:27:20] -!- abaumann has joined #archlinux32
[08:27:21] <buildmaster> Hi abaumann!
[08:27:21] <buildmaster> !rq abaumann
[08:27:22] <phrik> buildmaster: <abaumann> a babe is a checksum error
[08:49:48] -!- deep42thought has joined #archlinux32
[08:49:48] <buildmaster> Hi deep42thought!
[08:49:48] <buildmaster> !rq deep42thought
[08:49:50] <phrik> buildmaster: <deep42thought> but let me help put the little terrorists to bed, first ...
[09:06:26] <buildmaster> i486/virtualbox-modules-arch is broken (says eurobuild6-7-i486): https://archlinux32.org
[09:09:51] <buildmaster> i686/virtualbox-modules-arch is broken (says eurobuild6-6): https://archlinux32.org
[09:10:28] <buildmaster> pentium4/virtualbox-modules-arch is broken (says eurobuild6-2): https://archlinux32.org
[09:27:21] <buildmaster> i686/linux-zen is broken (says nlopc46): https://archlinux32.org
[09:38:33] <deep42thought> abaumann: have a look at db-update -o|--only if you want to move packages selectively
[09:41:27] <buildmaster> i686/linux is broken (says eurobuild6-1): https://archlinux32.org
[09:41:34] <abaumann> ah. ok
[09:42:05] <abaumann> ah. it's not in help :-)
[09:42:11] <abaumann> that's why I didn't see it
[09:42:15] <deep42thought> it's somewhat elaborate - maybe someone can compress my code :-)
[09:42:21] <abaumann> :-)
[09:42:53] <deep42thought> it /is/ in help, it's just, that I committed it just now
[09:43:01] <deep42thought> (maybe the buildmaster needs a git pull first)
[09:43:01] <buildmaster> pentium4/linux is broken (says eurobuild6-4): https://archlinux32.org
[09:44:20] <buildmaster> pentium4/linux-zen is broken (says eurobuild6-5): https://archlinux32.org
[09:44:48] <deep42thought> err, these linux errors look like a problem of the archive-tarball server :-/
[09:45:01] <abaumann> oups
[09:47:56] <deep42thought> might be some signing issue, too
[09:54:09] <deep42thought> apparently, the receiving of missing keys takes too long and the reverse proxy times out :-/
[09:54:29] <abaumann> oh.
[09:54:55] <abaumann> archlinux doens't have it's own keyserver (yet)?
[09:55:06] <abaumann> Because then a background process could try hard to fetch the keys..
[09:55:17] <abaumann> ..and build slaves can be sure the keys are always around
[09:55:33] <deep42thought> that's actually something I had in mind for archlinux32
[09:55:44] <abaumann> :-)
[09:56:08] <deep42thought> just throw all known key ids into it and let some background job try to receive them
[09:56:41] <deep42thought> and you can receive/update keys by 'curl keys.archlinux32.org/0xdeadbeef | gpg --import'
[09:57:40] <abaumann> something for the TODO list ;-)
[09:59:49] <deep42thought> the reason why I have not done this yet (not even the TODO entry) is, that I'm not sure on which machine this should run
[10:00:06] <abaumann> true. it's quite a security thingy..
[10:00:33] <abaumann> ..personal opinion: a server quite closed down and doing just that.
[10:00:53] <abaumann> and it is a single-point-of-failure, both in terms of building and security
[10:01:52] <deep42thought> what are the security risks?
[10:02:04] <abaumann> somebody hacking it and adding weird keys?
[10:02:06] <deep42thought> we could set up multiple of those servers
[10:02:16] <abaumann> yeah. but just for Archlinux32?
[10:02:28] <deep42thought> how is getting weird keys from our server different from getting weird keys from a foreign, unknown server?
[10:02:40] <deep42thought> abaumann: why not - I would use it personally, too :-D
[10:02:50] <abaumann> aha. :-)
[10:03:05] <abaumann> normally build slaves takes keys from a pool of keyservers (I hope).
[10:03:13] <deep42thought> "multiple" = 2 would suffice circumventing the single-point-of-failure problem
[10:03:15] <abaumann> So it's not so easy to fool the system
[10:03:19] <abaumann> true
[10:03:35] <deep42thought> fooling the system is easy - see recent attacks on sks
[10:03:42] <abaumann> mmh db-update -o is very handy :-)
[10:03:47] <deep42thought> \o/
[10:22:52] <abaumann> https://buildmaster-status.archlinux32.org
[10:23:03] <abaumann> why is percona-server there? isn't the default mysql server mariadb?
[10:23:14] <abaumann> same vor java-environment, why jdk7
[10:23:53] <deep42thought> this is probably just an inaccuracy in "blocks" column
[10:23:53] <abaumann> is this just a presentation thingy, becasue I think makeing the package will use the "usual" providing package for java-environment or mysql
[10:24:06] <deep42thought> it does not see, that the "mysqld" dependency is also provided by mariadb
[10:24:15] <abaumann> ah.
[10:24:23] <abaumann> ok. so I can safely ignore them..
[10:24:33] <deep42thought> I could implement this piece of logic, but then this query would become slow, too ;-)
[10:24:42] <abaumann> yeah. true :-)
[10:25:07] <deep42thought> well, not the web query, but the background query refreshing the `blocks` column there
[10:25:38] <abaumann> because. I tried to fix jdk7, and that's a beast (also for 64-bit)
[10:29:43] <abaumann> ah jack2 and jack. similar.
[10:36:29] -!- bill-auger has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
[10:38:26] <nit-picker> commit 010738591cacfc6deb2e61c2c4699e788a5ab9b3 is not an ancestor of HEAD 9bf3a2110fba5e329123b4afaa4006b5a803f524 in repository archlinux32
[10:39:11] -!- bill-auger has joined #archlinux32
[10:39:24] <deep42thought> why the hell is the current head at 9bf...?
[10:39:42] <deep42thought> ah, because you pushed it forward :-D
[10:40:56] -!- bill-auger has quit [Client Quit]
[10:41:09] <abaumann> oups..
[10:41:11] <abaumann> sorry :-)
[10:41:14] <deep42thought> np
[10:41:18] <deep42thought> it's a bug in the nit-picker
[10:41:40] <abaumann> that's another thing I'm usually doing. I use get-db-packages to get new package descriptions if I fix a package with a lot of dependencies
[10:41:43] -!- nit-picker has quit [Remote host closed the connection]
[10:41:55] <abaumann> This time it was gspell
[10:41:58] -!- nit-picker has joined #archlinux32
[10:41:58] <buildmaster> Hi nit-picker!
[10:41:59] <buildmaster> !rq nit-picker
[10:42:00] <phrik> buildmaster: <nit-picker> abaumann: but there are only 18 "intermediate" versions
[10:42:00] <abaumann> it blocks half of Gnome
[10:42:15] <abaumann> ./seed-build-list: line 196: /tmp/tmp.seed-build-list.c9epvH4yMx/blacklist: No such file or directory
[10:42:18] <abaumann> huh
[10:42:27] <deep42thought> üöäh
[10:42:35] <abaumann> seed-build-list cannot find the temporary file
[10:42:41] <abaumann> ./seed-build-list -w -p '^gspell$'
[10:43:16] <deep42thought> dammit I'm stupid
[10:43:27] * abaumann doesn't think so :-)
[10:43:41] <deep42thought> missing "tee" :-D
[10:43:46] <abaumann> :-)
[10:43:54] <abaumann> better than missing 'tea'
[10:44:11] <deep42thought> I have enough Tee over here
[10:44:46] <deep42thought> but as you can confirm, I also have quite some throughput (regarding tea)
[10:45:19] <abaumann> :-)
[10:46:08] <deep42thought> seed-build-list should be fixed, now
[10:46:16] <abaumann> thanks
[10:46:36] <abaumann> https://github.com
[10:46:37] <phrik> Title: LLVM ERROR: out of memory building rustc on powerpc ppc64 · Issue #60294 · rust-lang/rust · GitHub (at github.com)
[10:46:46] <abaumann> this is a nice hint how to not run out of memory when building rust
[10:47:45] <deep42thought> but it says, the default debug level is already quite low
[10:48:00] <deep42thought> does someone overwrite the default setting?
[10:48:26] <abaumann> possibly
[10:48:53] <abaumann> it's the most helpful hint I got so far :-)
[10:49:13] <abaumann> ./seed-build-list: line 199: /tmp/tmp.seed-build-list.8kZ9KxZJnO/ignore-package: No such file or directory
[10:49:16] <abaumann> ahem :-)
[10:49:22] <abaumann> there might be another one?
[10:49:24] <deep42thought> next missing tee?
[10:50:20] <deep42thought> no, just another random typo :'-(
[10:50:29] <abaumann> :-)
[10:50:41] <deep42thought> hot fixed for now
[10:50:50] <deep42thought> (e.g. "should work")
[10:50:52] <abaumann> thanks :-)
[10:51:09] <abaumann> aha, hotfixing on the buildmaster
[10:51:14] <deep42thought> yes
[10:51:17] <abaumann> got ya
[10:51:19] <abaumann> ;-)
[10:51:36] <deep42thought> when it works, I'll commit all the typo-fixes
[10:51:48] * deep42thought is ashamed, he apparently did not test his previous commits
[10:52:00] * abaumann sees no more error messages now :-)
[10:52:06] <deep42thought> good
[10:52:29] -!- bill-auger has joined #archlinux32
[10:52:51] * abaumann considers to rename himself into nick-picker ;-)
[10:53:02] <deep42thought> that nick is taken, sry
[10:53:05] <abaumann> :-)
[11:12:04] -!- infides has joined #archlinux32
[11:14:01] -!- bill-auger has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
[11:19:17] -!- bill-auger has joined #archlinux32
[11:20:24] <buildmaster> pentium4/haskell-hspec-meta is broken (says nlopc46) - I rescheduled: haskell-hspec-expectations, haskell-hunit: https://archlinux32.org
[11:20:49] -!- buildmaster has quit [Remote host closed the connection]
[11:22:14] -!- buildmaster has joined #archlinux32
[11:22:14] <buildmaster> !rq buildmaster
[11:22:15] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[11:36:16] <buildmaster> i686/opencv is broken (says eurobuild6-1): https://archlinux32.org
[11:36:33] <buildmaster> pentium4/opencv is broken (says eurobuild6-5): https://archlinux32.org
[11:47:57] -!- DCyrax has quit [Remote host closed the connection]
[11:52:41] * buildmaster goes insane.
[11:57:00] * buildmaster resumes sanity.
[11:57:13] <deep42thought> abaumann: you repaired that?
[11:57:56] -!- DCyrax has joined #archlinux32
[11:58:07] <abaumann> yep.
[11:58:19] <abaumann> I have the deletion-build-list-sql in the history now ;-)
[12:03:10] <deep42thought> some packages seem to wrongly enter the deletion-list
[12:04:57] <abaumann> yeah. I think, it's the once I force (seed and priorize) by hand
[12:05:01] <abaumann> *ones
[12:41:28] * buildmaster goes insane.
[12:43:58] * buildmaster resumes sanity.
[12:47:50] -!- infides has quit [Ping timeout: 272 seconds]
[13:11:10] <buildmaster> pentium4/haskell-configurator-pg is broken (says eurobuild6-3) - I rescheduled: haskell-test-framework-hunit: https://archlinux32.org
[13:20:04] -!- abaumann has quit [Quit: leaving]
[13:20:28] -!- abaumann has joined #archlinux32
[13:20:29] <buildmaster> Hi abaumann!
[13:20:29] <buildmaster> !rq abaumann
[13:20:30] <phrik> buildmaster: <abaumann> if your software depends on the gcc version, you should really stop developping software because your development process is seriously flawed.
[13:39:10] <buildmaster> i686/performous are broken (says eurobuild6-1): https://archlinux32.org
[13:42:34] <buildmaster> any/kotlin is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:54:39] <abaumann> upstream: Query {SELECT pref_name, pref_value FROM `flyspray_prefs`} with params {} Failed! (MySQL server has gone away)"
[13:54:58] <deep42thought> O.o
[13:55:16] <abaumann> back again.
[13:55:23] <abaumann> seems to have been maintainance :-)
[13:56:03] <abaumann> it seems to be a bad idea to trigger a rebuild..
[13:56:11] <abaumann> ..almost everything in Gnome is broken.
[13:56:18] <abaumann> and then there is gettext problems everywhere..
[13:56:21] <deep42thought> due to the rebuild?
[13:56:26] <abaumann> yes.
[13:56:30] <deep42thought> :-/
[13:56:48] <abaumann> the problem is. you can not safely assume that upstream repo is sort of a tagged version and everything is rebuilding just fine.
[13:56:54] -!- thePiGrepper has joined #archlinux32
[13:57:00] <deep42thought> I always thought, a rebuild cannot break more than it fixes, thus generating some semi-invariant which guarantees the algorithm to converge
[13:57:27] <abaumann> yeah. the problem is: gettext got upgraded, not everything using it (wrongly) was rebuilt
[13:57:50] <abaumann> in gnome mutter is broken, which is quite central for Gnome to work
[13:59:40] <deep42thought> we could/should trigger rebuilds for fixed git revisions
[13:59:55] <deep42thought> though, that doesn't sound practical ...
[14:00:01] <abaumann> .. yeah.
[14:00:03] <abaumann> not really.
[14:03:01] <buildmaster> pentium4/performous are broken (says eurobuild6-2): https://archlinux32.org
[14:07:17] <abaumann> "libotf: OpenType Font library", errors in checking for freetype-config... no Freetype library wan't found in your system!
[14:07:23] <abaumann> this is almost comical. :-)
[14:08:00] <abaumann> https://bugs.archlinux.org
[14:08:01] <phrik> Title: FS#60427 : [libotf] build fails in configure - freetype not found (at bugs.archlinux.org)
[14:08:31] <deep42thought> sounds familiar
[14:08:39] <abaumann> yes. you reported it :-)
[14:08:49] <deep42thought> and the last comment is from me :-)
[14:08:58] <abaumann> that's a problem in the error build list. sometimes bug comments (blocks) run away
[14:09:08] <abaumann> when packages get retriggered most likely.
[14:09:13] <deep42thought> definitely
[14:09:16] <deep42thought> that's a feature
[14:09:28] <deep42thought> the buildmaster does not matter about the content of the block reason
[14:09:35] <deep42thought> so it assumes the issue has been fixed
[14:09:43] <deep42thought> when it gets rescheduled (for whatever reason)
[14:09:46] <abaumann> which is a good assumption (usually) :-)
[14:10:06] <deep42thought> there is also a tool to find upstream bug reports to failed builds
[14:10:08] <deep42thought> let me find it ...
[14:10:22] <abaumann> ah? cool!
[14:10:41] <deep42thought> check-upstream-bug-tracker
[14:10:45] <deep42thought> it's interactive
[14:11:19] <abaumann> ah. sweet.
[14:11:21] <deep42thought> it autogenerates those "wait for FS#12345" block reasons
[14:11:36] <abaumann> and I'm copy pasting them since years... ;-)
[14:11:41] <deep42thought> lol
[14:12:07] <deep42thought> Is it also "lazy" to not search for a tool which automates a tedious task?
[14:12:39] * abaumann mumbles something which is not quite understandable..
[14:14:01] <abaumann> you have a similar script checking for flagged-out-of-date upstream?
[14:14:09] <deep42thought> no
[14:14:22] <deep42thought> I think, it's on a todo to automate that
[14:14:27] <abaumann> to be fair: this is not a frequent case.
[14:14:48] <deep42thought> here you go: https://archlinux32.org
[14:14:49] <phrik> Title: Todos in the build scripts (at archlinux32.org)
[14:15:16] <abaumann> Do you have a ii-hook to check for TODO descriptions if I asked for silly features? ;-)
[14:15:18] <deep42thought> the plan is to check them right on return
[14:15:25] <deep42thought> !grab abaumann
[14:15:25] <phrik> deep42thought: Tada!
[14:15:42] <deep42thought> no, but I can add it to the todo list ;-P
[14:15:48] <abaumann> lol
[14:17:06] <abaumann> Query {SELECT pref_name, pref_value FROM `flyspray_prefs`} with params {} Failed! (Too many connections)
[14:17:16] <abaumann> mmh. Am I reporting too many bugs..? ;-)
[14:17:21] <deep42thought> lol
[14:18:49] <deep42thought> I don't want to sound spiteful, but it's nice to see, others have problems with their database, too.
[14:19:05] <abaumann> :-)
[14:26:30] <abaumann> https://buildmaster-status.archlinux32.org
[14:26:32] <phrik> Title: Buildmaster for Archlinux32 packages (i686, pentium4, any) (at buildmaster-status.archlinux32.org)
[14:26:36] <abaumann> that's a slope I like
[14:26:42] <abaumann> 1618 blue line
[14:27:09] <deep42thought> should we make a graph of the derivatives of those lines, too ;-)
[14:27:49] <abaumann> the mathemtician/physisist in me rejoices. :-)
[14:27:59] * abaumann thinks it's only a small mathemtician/physisist
[14:28:15] <deep42thought> I can do all kind of analysis on those graphs
[14:28:56] * deep42thought sights his mathematica scripts intended for spectum analysis and simulation data evaluation
[14:39:30] * buildmaster goes insane.
[14:42:23] -!- deep42thought has quit [Remote host closed the connection]
[14:42:24] -!- nit-picker has quit [Remote host closed the connection]
[14:44:37] * buildmaster resumes sanity.
[14:45:14] -!- nit-picker has joined #archlinux32
[14:45:14] <buildmaster> Hi nit-picker!
[14:45:14] <buildmaster> !rq nit-picker
[14:45:15] <phrik> buildmaster: <nit-picker> abaumann: but there are only 18 "intermediate" versions
[14:49:53] <buildmaster> i486/nodejs-lts-dubnium is broken (says eurobuild6-7-i486): https://archlinux32.org
[15:02:20] -!- abaumann has quit [Quit: leaving]
[15:14:13] <buildmaster> i486/clasp is broken (says tyzoid-srv0-vm486): https://archlinux32.org
[15:19:55] <buildmaster> pentium4/kresus are broken (says eurobuild6-3): https://archlinux32.org
[15:39:16] -!- thePiGrepper has quit [Ping timeout: 248 seconds]
[15:41:22] -!- thePiGrepper has joined #archlinux32
[15:53:24] -!- thePiGrepper has quit [Ping timeout: 272 seconds]
[15:54:56] -!- thePiGrepper has joined #archlinux32
[16:18:29] <buildmaster> i686/haskell-sbv is broken (says eurobuild6-2): https://archlinux32.org
[16:29:44] <buildmaster> i686/shellcheck is broken (says eurobuild6-1): https://archlinux32.org
[17:04:54] * buildmaster goes insane.
[17:07:14] -!- MrBIOS has joined #archlinux32
[17:42:30] * buildmaster resumes sanity.
[17:47:36] -!- thePiGrepper has quit [Quit: leaving]
[18:06:51] <buildmaster> i686/hedgewars are broken (says eurobuild6-4): https://archlinux32.org
[18:08:39] -!- thePiGrepper has joined #archlinux32
[18:10:35] -!- infides has joined #archlinux32
[18:17:08] -!- infides has quit [Ping timeout: 258 seconds]
[18:36:48] -!- thePiGrepper has quit [Ping timeout: 272 seconds]
[19:03:15] -!- eschwartz has quit [Ping timeout: 250 seconds]
[19:06:20] -!- eschwartz has joined #archlinux32
[19:16:31] <buildmaster> pentium4/nvidia-390xx is broken (says eurobuild6-3): https://archlinux32.org
[19:25:15] -!- thePiGrepper has joined #archlinux32
[19:34:30] <thePiGrepper> for some reason, pulseaudio 12.2-2.6 broke audio on my system, downgrading to 2.4 fixed it, so Im asking myself how could it be arch32 fault if pulseaudio has no patch for pulseaudio in the first place...
[19:58:26] -!- abaumann has joined #archlinux32
[19:58:27] <buildmaster> Hi abaumann!
[19:58:27] <buildmaster> !rq abaumann
[19:58:28] <phrik> buildmaster: <abaumann> Working in IT makes working in an asylum for mental people look like holidays.. ;-)
[19:59:09] <abaumann> thePiGrepper: just tried, alsa mplayer words, pulseaudio says "audio device got stuck"
[19:59:31] <abaumann> there is also this one: https://bbs.archlinux32.org
[19:59:33] <phrik> Title: Audio hardware not detected / Kernel & Hardware / Arch Linux 32 Forums (at bbs.archlinux32.org)
[19:59:56] * abaumann hates pulseaudio with a passion..
[20:00:26] <abaumann> basically for the same reasons as systemd.. overenginered, badly tested stuff.
[20:00:33] <abaumann> personal opinion, sorry
[20:00:57] <abaumann> yep. pulseadio doesn't recognize my sound card.. cool
[20:01:07] * abaumann stickes to alsa
[20:01:59] <abaumann> this doesn't mean I don't want to fix it :-)
[20:03:17] -!- abaumann has quit [Quit: leaving]
[20:11:01] <thePiGrepper> haha, dont worry about it. it's quite a popular opinion ;-)
[20:12:36] <thePiGrepper> however, as I said, 2.4 works just fine, so what gives? whats the difference between 2.4 and 2.6 on arch32's part? there are no patches, so it's basically just a recompiled package with some newer dependencies 'maybe'?
[20:15:33] -!- abaumann has joined #archlinux32
[20:15:33] <buildmaster> Hi abaumann!
[20:15:33] <buildmaster> !rq abaumann
[20:15:35] <phrik> buildmaster: <abaumann> if fail -> try again (hard)
[20:15:38] * abaumann shrugs
[20:15:54] <abaumann> just noticed, I never tested sound on my virtual machines (staging, testing)..
[20:17:08] <abaumann> mmh. ICH6 ICH9 or AC97..
[20:17:36] <abaumann> ah. and it's a server without sound output. adding some speakers will probably help. :-)
[20:18:27] <abaumann> ICH6, I have the same on the eeepc, so..
[20:18:49] <buildmaster> i486/celluloid is broken (says eurobuild6-7-i486): https://archlinux32.org
[20:21:44] <thePiGrepper> mine is ICH7
[20:21:56] <thePiGrepper> what kind of eeepc do you have?
[20:29:54] <buildmaster> any/gnucash-docs are broken (says eurobuild6-7-i486): https://archlinux32.org
[20:32:43] <abaumann> eepc 701 4G
[20:32:59] * abaumann is sorry, had to hunt some power cables for 20 year old sound boxes
[20:50:19] <abaumann> mmh. so on my server alsa and pulse are both working on 64-bit with pulseaudio 12.2
[20:50:22] <abaumann> so..
[20:51:42] <abaumann> E: [pulseaudio] ltdl-bind-now.c: Failed to open module /usr/lib/pulse-12.2/modules/module-alsa-card.so: /usr/lib/pulse-12.2/modules/libalsa-util.so: undefined symbol: pa_alsa_ucm_device_update_available
[20:51:49] <abaumann> ok. this one has to be solved first..
[20:56:16] <thePiGrepper> indeed, pulseaudio 12.2 is just fine. in my x64 systems, as well as in my arch32 systems, up to 2.4. have you tried rebuilding 2.6? or maybe checking which dependencies changed version between 2.4 and 2.6?
[20:57:31] <thePiGrepper> abaumann: how did you get that 'ltdl-bind-now.c: Failed to open module' error?
[20:57:42] <abaumann> journalctl -fl
[20:57:50] <abaumann> while starting to play sound
[20:58:08] <abaumann> or.. while starting pulseaudio by hand..
[20:58:19] <abaumann> as root
[20:58:23] <thePiGrepper> which version of pulseaudio?
[20:58:25] <abaumann> otherwise I get silly dbus errors
[20:58:34] <abaumann> 12.2
[20:58:39] <abaumann> pulseaudio 12.2-2.6
[20:58:39] <thePiGrepper> on x64?
[20:58:43] <abaumann> on 32-bit
[20:58:45] <thePiGrepper> oh, ok
[20:58:48] <abaumann> i686 and pentium4
[20:58:57] <thePiGrepper> then that's the issue for sure
[20:58:59] <abaumann> this one is the one missing symbol mentioned in the forum.
[20:59:06] <thePiGrepper> I'm on pentium4 btw
[20:59:27] <abaumann> https://bbs.archlinux32.org
[20:59:28] <phrik> Title: Audio hardware not detected / Kernel & Hardware / Arch Linux 32 Forums (at bbs.archlinux32.org)
[20:59:31] <abaumann> yep. that's the one.
[20:59:38] <thePiGrepper> thx, I was about to ask you for the link.
[20:59:47] <abaumann> :-)
[21:02:02] <abaumann> mmh. some too agressive linking removing unused symbols, I wonder?
[21:03:39] <abaumann> ah. 12.2 is flagged out of date, 12.99 is in testing (upstream)
[21:06:24] <abaumann> 10 hours on a pentium4 64-bit for building the kernel.
[21:06:26] <abaumann> well..
[21:07:12] <thePiGrepper> yeah ,Ive just read that here https://www.linuxquestions.org
[21:07:14] <phrik> Title: [SOLVED] Pulseaudio Mon May 13 22:53:13 UTC 2019 (at www.linuxquestions.org)
[21:08:01] <thePiGrepper> do you think that 12.99 has that patch? I assume so right?
[21:08:09] <abaumann> probably
[21:08:42] <abaumann> https://gitlab.freedesktop.org
[21:08:44] <phrik> Title: Failed to load module "module-alsa-card" (#718) · Issues · PulseAudio / pulseaudio · GitLab (at gitlab.freedesktop.org)
[21:08:50] <abaumann> this is actually quite a sensless bugreport.
[21:08:56] <abaumann> "please try on Ubuntu"
[21:09:04] <abaumann> so, they admit basically they are testing Ubuntu only
[21:09:19] <abaumann> the report came from Antergos
[21:09:27] <abaumann> and Antergos has no 32-bit version..
[21:10:44] <thePiGrepper> I built it an hour ago, from the AUR, however I dont really feel like testing that while working on my machine hehe. maybe this afternoon. how would you do to install the new package and restart pulse?
[21:11:07] <abaumann> pulseaudio -k
[21:11:10] <thePiGrepper> abaumann: indeed , I just read that report, is really stupid that they admit that even..
[21:11:11] <abaumann> and makepkg -si
[21:11:31] <abaumann> well. they have to develop on something. so picking a mainstream distribution is ok.
[21:11:47] <abaumann> there are open build system (OBS), or travis or whatever to test the builds.
[21:12:09] <abaumann> I'm sure redhat/opensuse/etc are testing when they integrate the sources
[21:12:12] <abaumann> so is archlinux.
[21:12:41] <abaumann> why it's breaking in archlinux32.. I had trouble before that the linker is not working exactly as on 64-bit
[21:12:49] <abaumann> so, sometimes symbols can run away :-)
[21:13:02] <abaumann> ../pulseaudio/src/pulsecore/svolume_mmx.c:239:1: warning: unsupported size for integer register
[21:13:05] <abaumann> ahem..
[21:13:33] <abaumann> well. that's a nice experiment for me to test sound..
[21:13:37] <abaumann> ..another day.
[21:13:44] * abaumann thinks he is sleepy now
[21:13:52] <thePiGrepper> a min..
[21:13:56] <abaumann> huh?
[21:14:05] <thePiGrepper> do you know if it makes sense for me to still be hearing sound
[21:14:10] <thePiGrepper> after pulseaudio -k
[21:14:10] <thePiGrepper> ?
[21:14:19] <abaumann> oeh.
[21:14:27] <abaumann> yes. if sound is going via alsa. :-)
[21:14:33] <thePiGrepper> oh, ok..
[21:14:41] <abaumann> if pulseaudo is killed, then some software falls back to alsa
[21:14:47] <thePiGrepper> ohhh
[21:14:49] <thePiGrepper> I see
[21:14:52] <abaumann> :-)
[21:14:53] <thePiGrepper> that kinda explains it
[21:15:03] <thePiGrepper> then how cmus didnt picked alsa
[21:15:09] <thePiGrepper> with pulseaudio 2.6?
[21:15:14] <thePiGrepper> when it broke
[21:15:33] <abaumann> cmus? a, didn't know that one..
[21:15:37] <abaumann> cmus is command line.
[21:15:47] <thePiGrepper> well, I'll try to install the new build then
[21:15:53] <abaumann> pulseaudio is a little bit more guish usually
[21:15:57] <thePiGrepper> btw, how do you restart pulse after killing it with -k
[21:16:19] <abaumann> mmh. should happen automatically I though, as soon as a software needs pulse
[21:16:33] <abaumann> I might be wrong, some time I tinkered with this stuff..
[21:16:37] <thePiGrepper> abaumann: really? because I tend to use pactl to set pulseaudio
[21:17:36] <abaumann> if you use 'pactl list' then notice the little bit slower output after you killed pulseaudio (the daemon)?
[21:17:49] <abaumann> that's when the pulseaudio daemon is started in the background
[21:18:26] <thePiGrepper> let me see
[21:18:39] <abaumann> ../pulseaudio/src/tests/thread-test.c:108:E:thread:thread_test:0: (after this point) Test timeout expired
[21:18:47] <thePiGrepper> no,not really slower
[21:18:48] <abaumann> sweet. the developer uses a faster machine, it seems..
[21:19:09] <abaumann> how did pulseaudio get into staging.. *puzzle8
[21:19:10] <thePiGrepper> hehe. typical
[21:19:52] <thePiGrepper> well, apparently I built it on a tmpfs *before* I rebooted my machine out of frustration
[21:19:55] <thePiGrepper> so, yeah..
[21:20:11] <thePiGrepper> Ill restart 2.4 and start rebuilding
[21:20:13] <thePiGrepper> lol
[21:20:16] <abaumann> you need persistent tmpfs
[21:20:27] * abaumann is kidding
[21:20:33] <thePiGrepper> lol
[21:21:02] <thePiGrepper> well I'll try to install optane in my netbook
[21:21:43] <thePiGrepper> arent they trying to make dimms compatible optine now lol
[21:21:55] <buildmaster> i686/hedgewars are broken (says eurobuild6-4): https://archlinux32.org
[21:22:59] <abaumann> yeah. 12.99.2 works fine.
[21:23:12] <abaumann> so let's wait. get's published soon upstream..
[21:23:18] <thePiGrepper> oh, nice to hear!
[21:23:30] <thePiGrepper> I'll keep 2.4 then,for now
[21:23:36] <abaumann> yes.
[21:23:44] <abaumann> ok. must sleep now.
[21:23:46] <abaumann> bye :-)
[21:23:50] <thePiGrepper> ok, bye
[21:23:52] -!- abaumann has quit [Quit: leaving]
[21:29:59] <buildmaster> i686/nvidia-390xx is broken (says eurobuild6-3): https://archlinux32.org
[21:40:22] <buildmaster> girls, please have a look at my dirty database
[21:40:22] * buildmaster goes insane.
[22:48:19] <girls> fyi: the insanity was due to a version downgrade of lua-sdl2 upstream
[22:48:48] * buildmaster resumes sanity.
[22:51:18] -!- ofara_ has quit [Ping timeout: 245 seconds]
[22:52:40] -!- ofara_ has joined #archlinux32
[23:05:36] <buildmaster> i486/cups are broken (says eurobuild6-7-i486): https://archlinux32.org
[23:19:52] <buildmaster> any/gandi.cli is broken (says eurobuild6-7-i486): https://archlinux32.org
[23:35:16] <buildmaster> i686/grafana is broken (says eurobuild6-2): https://archlinux32.org
[23:45:39] <buildmaster> pentium4/grafana is broken (says eurobuild6-2): https://archlinux32.org
[23:46:09] <buildmaster> pentium4/bup is broken (says eurobuild6-6): https://archlinux32.org