#archlinux-ports | Logs for 2018-03-09

[00:13:38] eschwartz is now known as anyone
[00:58:37] -!- oaken-source has quit [Ping timeout: 256 seconds]
[01:24:35] -!- guys has quit [Ping timeout: 240 seconds]
[01:49:30] -!- chromicant has joined #archlinux-ports
[02:07:07] -!- guys has joined #archlinux-ports
[02:08:40] anyone is now known as eschwartz
[02:15:31] -!- |ast| has joined #archlinux-ports
[02:42:55] <buildmaster> haskell-http-api-data is broken (says buildknecht3).
[03:12:37] -!- |ast| has quit [Quit: Lost terminal]
[03:13:57] -!- |ast| has joined #archlinux-ports
[05:14:45] <buildmaster> supercollider is broken (says buildknecht).
[05:16:06] -!- yar has quit [Excess Flood]
[05:19:27] -!- yar has joined #archlinux-ports
[05:22:53] -!- wyre has quit [Ping timeout: 240 seconds]
[05:24:24] -!- wyre has joined #archlinux-ports
[05:24:35] -!- hringriin_ has joined #archlinux-ports
[05:27:41] -!- hringriin has quit [Ping timeout: 245 seconds]
[05:27:42] hringriin_ is now known as hringriin
[05:43:42] -!- |ast| has quit [Quit: leaving]
[06:11:00] -!- deep42thought has joined #archlinux-ports
[06:12:26] <deep42thought> tyzoid: I think, it's easiest do to that, when everything works via the database.
[06:12:44] <deep42thought> Because then we must only restore that anywhere and check for sanity
[06:13:20] <deep42thought> plus: we should check, that the restored database also contains the correct stored functions
[06:29:20] <deep42thought> strange: I replaced the hardware of my archlinux32 with a "new" one, but it still freezes ocassionally :-/
[06:54:11] -!- deep42thought has quit [Quit: Leaving.]
[07:49:49] -!- oaken-source has joined #archlinux-ports
[08:19:44] -!- deep42thought has joined #archlinux-ports
[08:54:42] -!- oaken-source has quit [Remote host closed the connection]
[10:04:05] -!- oaken-source has joined #archlinux-ports
[10:14:03] -!- abaumann has joined #archlinux-ports
[11:08:43] <buildmaster> deep42thought: my database is dirty again ...
[11:08:48] <deep42thought> \o/
[11:12:11] <deep42thought> wrong column separator => new_repository=NULL => all moved packages did not arrive in their repository in the mysql database
[11:12:25] <deep42thought> at least, they were not dropped from the database :-)
[11:26:56] <deep42thought> plus: I started enumerating the repositories at 1, so I _know_ which packages need to be moved :-)
[11:30:09] * buildmaster resumes sanity.
[11:49:44] <buildmaster> deep42thought: my database is dirty again ...
[11:54:18] -!- abaumann has quit [Quit: leaving]
[12:07:01] * buildmaster resumes sanity.
[14:34:26] <buildmaster> deep42thought: my database is dirty again ...
[14:42:38] -!- abaumann has joined #archlinux-ports
[14:43:06] <deep42thought> what
[14:43:11] <deep42thought> ah
[14:43:16] <deep42thought> never mind
[14:52:59] -!- abaumann has quit [Quit: leaving]
[14:56:07] * buildmaster resumes sanity.
[14:56:50] <deep42thought> let's try that again ...
[15:02:15] -!- isacdaavid has quit [Quit: Leaving.]
[15:06:54] <buildmaster> deep42thought: my database is dirty again ...
[15:07:21] <deep42thought> at least, it's reproducible :-D
[15:10:38] <deep42thought> hmm, actually not really: a different set of packages is missing from the database :-/
[15:15:00] * buildmaster resumes sanity.
[15:17:17] <tyzoid> deep42thought: I'd argue that it's almost more important to test it now when we've got state thrown everywhere
[15:17:25] <tyzoid> to make sure that we've captured all the state
[15:17:55] <deep42thought> but what/how would you check?
[15:18:17] <tyzoid> Well the current buildmaster is somewhere in a vm, right?
[15:18:25] <tyzoid> in rewby's rack, iirc?
[15:18:25] <deep42thought> yes
[15:18:28] <deep42thought> right
[15:18:44] <tyzoid> So the best way to test would be to provision another vm somewhere else and see if we can restore a working buildmaster
[15:19:41] <deep42thought> yes, but that is easier if there's only the database to restore
[15:19:43] <tyzoid> Because if the system is rendered offline by any number of causes, we can be assured we can recover
[15:19:47] <tyzoid> That's true too
[15:20:00] <tyzoid> but depending on how long we're going to remain in the mixed state we're in now
[15:20:11] <deep42thought> the problem I see is, that you'd need the file system and database of the same time
[15:20:25] <tyzoid> And the backup should have both, right?
[15:20:30] <deep42thought> I replaced db-update by a mysql only variant
[15:20:42] <deep42thought> yes, but not at the same time, necessarily
[15:20:52] <tyzoid> Is the database physically on the buildmaster? Or is it on another machine?
[15:21:01] <deep42thought> on the bm
[15:21:09] <deep42thought> so we might have that one in files, too
[15:21:10] <tyzoid> The snapshot is taken at a fixed point in time
[15:21:15] <deep42thought> - if it's not in the ram
[15:21:35] <deep42thought> I'm not sure if you can restore a running mariadbd
[15:21:37] <deep42thought> :-/
[15:21:40] <tyzoid> but after each transaction, mysql/maria will journal those transactions
[15:21:51] <tyzoid> And yes, you can. They're designed to be resilliant to crashing
[15:22:00] <deep42thought> ok
[15:22:40] <deep42thought> well, then, the borg backup should have everything we need
[15:23:30] <tyzoid> You think you'll have any time this weekend to test out a recovery?
[15:23:52] <deep42thought> yes, maybe
[15:24:17] <deep42thought> my wife won't be too euphoric about that, but I think: "yes" :-)
[15:24:53] <deep42thought> oh
[15:24:57] <deep42thought> I see the problem
[15:25:21] <deep42thought> (the other problem, I'm pursuing)
[15:25:25] <tyzoid> Ok, It's more a nice to have at this point, but something that should be done. I wouldn't want it to get in the way of personal time
[15:25:35] <tyzoid> I'll throw up an encrypted tarball some time today that you can play around with when you have some time
[15:25:36] <deep42thought> nah, it's ok
[15:26:23] <buildmaster> deep42thought: my database is dirty again ...
[15:31:43] * buildmaster resumes sanity.
[15:33:03] <deep42thought> tyzoid: archlinux32 in general is "gettin in the way" of personal time ;-) (isn't that the definition of "spare time"?)
[15:37:15] * deep42thought found the bug
[15:39:14] -!- abaumann has joined #archlinux-ports
[15:39:20] <deep42thought> db-update uses different locks than get-package-updates (that's ok) and it calls mysql_cleanup at the beginning to reduce the volume of some joins - but mysql_cleanup removes all "preliminary packages" from the database
[15:39:43] <abaumann> mmh. python-setuptools needs python-pip, python-pip needs python-setuptools in makedepends. How can this possibly be bootstrapped?!
[15:39:44] <deep42thought> so get-package-updates loses all preliminary packages at that point ...
[15:40:04] <deep42thought> abaumann: can you somehow compile both simultanously?
[15:40:18] <abaumann> in parallel, on two virtual machines? :-)
[15:40:26] <deep42thought> no, in one single chroot ;-P
[15:40:39] <abaumann> I don't know how.
[15:40:51] * deep42thought was not 100% serious
[15:40:53] <abaumann> anyway. systemd needs meson, meson needs ninja, ninja needs python2
[15:41:00] <abaumann> and this for a low-level PID=1 system.
[15:41:05] <abaumann> I'm sorry, but this is madness.
[15:41:11] <deep42thought> these are only makedepends, rightß
[15:41:12] <deep42thought> ?
[15:41:17] <abaumann> only.
[15:41:21] <deep42thought> :_D
[15:41:23] <deep42thought> :-D
[15:41:28] <abaumann> yes. if you want to build systemd that is.
[15:41:43] <abaumann> somehow I suspect the last bootstrapping of archlinux happened before systemd came along..
[15:42:03] <deep42thought> maybe, you should bootstrap without it, then?
[15:42:13] <guys> abaumann: hrm, why does setuptools need pip anyway, surely that is why the bootstrap.py exists
[15:42:15] <abaumann> sooner or later it has to be built.
[15:42:37] <abaumann> where is bootstrap.py?
[15:42:43] <abaumann> in python itself?
[15:43:00] <guys> in setuptools source
[15:43:04] <abaumann> ah: python bootstrap.py
[15:43:23] <guys> it writes a stub setuptools.egg-info
[15:43:24] <abaumann> but then I can just drop the makedepends..
[15:43:31] <abaumann> oh. nice. :-)
[15:43:36] <guys> abaumann: if it works, this is a definite bug
[15:43:43] <abaumann> :-)
[15:43:57] <abaumann> let's see..
[15:44:09] <guys> testing...
[15:45:52] <guys> --nocheck works fine without it, now for regular build...
[15:47:03] <guys> it is a checkdepends
[15:47:06] <guys> I will open a bug
[15:47:30] <abaumann> wow. I don't have to open a bug. :->
[15:47:33] <buildmaster> haskell-yesod-auth is broken (says rechenknecht).
[15:47:38] <abaumann> just kidding.. :-0
[15:48:00] <deep42thought> :-D
[15:48:04] <abaumann> ok. I'll patch accordingly..
[15:48:12] <abaumann> thanks guys.
[15:50:13] <abaumann> mmh.. python --without-ensurepip.. in python.. wouldn't that enable a built-in pip to avoid the hen-and-egg-problem?
[15:52:18] <guys> https://bugs.archlinux.org
[15:52:19] <phrik> Title: FS#57770 : [python-setuptools] pip is listed as a makedepends but it is only a checkdepends (at bugs.archlinux.org)
[15:52:41] <guys> but python bundling ensurepip is bloated and we want pip as a package...
[15:54:25] <buildmaster> haskell-yesod-static is broken (says rechenknecht).
[15:54:28] <abaumann> yeah. It would be double then. But this means for bootstrapping python I have to build with the embedded pip, then first build python-setuptools with the embedded pip, rebuild python without embedded pip and then build python-pip as package.
[15:54:28] -!- p71 has quit [Remote host closed the connection]
[15:54:32] <abaumann> sounds ok. :-)
[15:54:41] <abaumann> *doable
[15:54:51] <guys> well, you could just bootstrap with --nocheck
[15:55:34] <abaumann> yeah. I wonder anyway, when I should really start testing the packages.. so far testing draws in too many dependencies anyway. So stage4 I suppose.. :-)
[15:57:08] <abaumann> python setup.py build in python-pip: Traceback (most recent call last):
[15:57:08] <abaumann> File "setup.py", line 6, in <module>
[15:57:08] <abaumann> from setuptools import setup, find_packages
[15:57:08] <abaumann> ModuleNotFoundError: No module named 'setuptools'
[15:57:12] <abaumann> so..
[15:57:26] <abaumann> it's also used for building not only for testing?
[15:57:53] <buildmaster> deep42thought: my database is dirty again ...
[15:57:58] -!- p71 has joined #archlinux-ports
[15:59:33] <guys> abaumann: hence why it is listed as a makedepends, not a checkdepends?
[15:59:53] <guys> It's okay, pip can be bootstrapped after setuptools as long as you use --nocheck
[16:00:06] * buildmaster resumes sanity.
[16:00:23] <abaumann> ah. ok.
[16:00:28] <buildmaster> haskell-yesod-test is broken (says rechenknecht).
[16:00:41] <guys> I'm not positive about some of the things setuptools depends on though
[16:01:12] <abaumann> exactly.
[16:01:24] <guys> But this is at least not *also* a pip problem :D
[16:01:26] <abaumann> most likely they depend on setuptools themselves. :-)
[16:01:50] <guys> Yes, IIRC setuptools was bootstrapped by initially uploading a setuptools package that just vendored everything
[16:02:18] <guys> So build-support should have a python-setuptools-bootstrap package that vendors everything. ;)
[16:03:16] <abaumann> I really thought I could bootstrap without python, because I wanted to avoid this trouble. But with systemd now using meson/ninja/python2 I have no option, I'm afraid..
[16:04:53] <deep42thought> abaumann: it does not matter if systemd depends on python
[16:05:03] <deep42thought> if you want to have python at some point, you need to bootstrap it
[16:05:13] <deep42thought> s/python/python-setuptools/
[16:05:56] <abaumann> systemd needs arch-meson to build.
[16:06:02] <abaumann> meson bases on ninja.
[16:06:15] <abaumann> ninja is written in python, as far as I recall.
[16:06:27] <abaumann> Another option is to reactivate the non-meson buildsystem of systemd.
[16:06:38] <deep42thought> no
[16:06:59] <abaumann> no?
[16:07:17] <deep42thought> do you want to have python in the end? (yes!) => you need to bootstrap it at some point => you can do that before systemd without any additional pain
[16:07:33] <abaumann> ah. that's true.
[16:07:47] <guys> abaumann: meson is written in python, ninja is not
[16:07:48] <abaumann> that's something I ask myself at every cross-road.
[16:07:53] <deep42thought> that's what I meant with "bootstrap without systemd" in the beginning
[16:08:16] <abaumann> yeah. but sooner or later you want to build systemd :->
[16:08:39] <abaumann> bootstrapping with build-support is no option if you want to cross-compile.
[16:09:07] <deep42thought> I'm just saying, that you need python anyway, so you do not need to bother to compile systemd without it
[16:09:07] <abaumann> but you can build a python setuptool without the packages and use that one.
[16:09:57] <abaumann> BTW: the testing in python build is for profile based optimization, can be disabled with --disable-optimizations.
[16:10:06] <abaumann> Python itself is no problem to bootstrap.
[16:10:16] <deep42thought> ah, ok :-)
[16:11:16] <abaumann> I have to see what oaken-source did in his RISCV scripts..
[16:12:10] <abaumann> ah. He uses meson with cross-compile, meson from the host.
[16:12:41] <abaumann> so. no need to bootstrap a meson..
[16:12:50] <abaumann> ..when doing it this way.
[16:12:52] -!- guys has quit [Ping timeout: 240 seconds]
[16:13:07] <abaumann> ok. python it is.. *sigh*
[16:13:31] -!- p71 has quit [Remote host closed the connection]
[16:16:02] * buildmaster goes insane.
[16:16:09] <deep42thought> *grrrrr*
[16:16:17] <abaumann> oups. trouble?
[16:16:33] <deep42thought> no, I just thought, I removed the very last bug :-)
[16:16:59] * buildmaster resumes sanity.
[16:17:00] <deep42thought> it looks like that was just temporary
[16:17:03] <deep42thought> (I did nothing)
[16:17:40] <abaumann> temporary mental illness ;-)
[16:18:13] <deep42thought> the mysql queries are now cached in a file (so I can redo them if they fail) and the sanity-check also checks for leftover files
[16:18:40] <abaumann> ah.. sweet.
[16:18:43] <deep42thought> and it appeared, that is saw such a file, which was removed (after the successful query) before I could have a look
[16:18:57] <deep42thought> s/is/it/
[16:19:20] <abaumann> you could move them to another directory, so you can have a glance at what the buildmaster corrected automatically..
[16:19:52] <deep42thought> I would end up with _a_lot_ of such files
[16:20:18] <abaumann> yeah. but you can still put a cron or make this a debug option..
[16:20:32] <deep42thought> I don't see the advantage
[16:20:32] <abaumann> ..better than to have unsolvable puzzles..
[16:20:47] <deep42thought> I already have a cumulative file of stdin,stdout and stderr to/from mysql
[16:20:54] <abaumann> ah. ok.
[16:21:03] <deep42thought> that's truncated to 10k lines
[16:21:17] <deep42thought> but it won't get filled further if the build master goes insane
[16:21:45] <abaumann> sounds good.
[16:21:52] <deep42thought> btw, tyzoid: did you have any progress with the database interface so far?
[16:22:09] <tyzoid> Nope, unfortunately
[16:22:22] <deep42thought> ok, np - I'm just curious :-)
[16:23:29] <tyzoid> It's on the list, but above it are 1) Fix/reduce Forum Spam issue, 2) Resolve periodic failures of iso build scripts
[16:24:27] <tyzoid> And at some point, I'd like to migrate my stuff from docker to kubernetes
[16:24:44] <deep42thought> ok, no hurry
[16:24:53] <deep42thought> forum spam is much more important :-D
[16:34:38] <abaumann> tyzoid: remote shell execution, check out shmux..
[16:34:56] <abaumann> .. stumbled upon today in the news of VoidLinux.
[16:36:18] <tyzoid> "The most likely way for the world to be destroyed, most experts agree, is by accident. That's where we come in. We're computer professionals. We cause accidents."
[16:36:21] <tyzoid> Love their tagline
[16:37:01] <abaumann> lol
[16:37:50] <tyzoid> looks pretty neat, though
[16:38:38] <abaumann> especially because it's scriptable and you can put results in separate textfiles per host..
[16:38:45] <abaumann> ..error handling looks quite neat.
[16:39:10] <deep42thought> goota go, cu later!
[16:39:14] <tyzoid> ok, see ya
[16:39:14] -!- deep42thought has quit [Quit: Leaving.]
[16:39:27] <buildmaster> python-xcffib is broken (says buildknecht2).
[16:39:30] <abaumann> cu
[16:40:56] <abaumann> ok. have to rush..cu.
[16:41:00] -!- abaumann has quit [Quit: leaving]
[16:58:10] -!- yokel has quit [Remote host closed the connection]
[16:59:21] -!- p71 has joined #archlinux-ports
[17:00:36] -!- yokel has joined #archlinux-ports
[17:04:11] <tyzoid> deep42thought: PM me when you get back, I've got the archive ready for you.
[17:04:52] <tyzoid> also, let me know if you want an account on my ZNC bouncer, it'd allow you to receive offline PMs
[17:18:58] <buildmaster> embree is broken (says rechenknecht).
[17:24:57] -!- |ast| has joined #archlinux-ports
[17:26:44] -!- isacdaavid has joined #archlinux-ports
[17:31:31] <buildmaster> haskell-hopenpgp is broken (says rechenknecht).
[17:34:24] -!- yokel has quit [Ping timeout: 260 seconds]
[17:36:55] -!- yokel has joined #archlinux-ports
[17:54:54] <buildmaster> haskell-hspec-wai-json is broken (says rechenknecht).
[17:55:11] -!- CalimeroTeknik has quit [Quit: バイバイ]
[18:05:24] <buildmaster> haskell-hjsonschema is broken (says buildknecht3).
[18:11:34] -!- deep42thought has joined #archlinux-ports
[18:12:31] <deep42thought> tyzoid: re
[18:22:43] <buildmaster> xmobar is broken (says buildknecht3).
[18:24:06] <buildmaster> tamarin-prover is broken (says buildknecht2).
[18:26:36] <buildmaster> idris is broken (says eurobuild3).
[18:45:02] <buildmaster> vlc is broken (says rechenknecht).
[19:01:58] <buildmaster> haskell-criterion is broken (says rechenknecht).
[19:07:33] <buildmaster> haskell-http-api-data is broken (says buildknecht3).
[19:36:43] -!- bill-auger has quit [Remote host closed the connection]
[19:40:35] -!- bill-auger has joined #archlinux-ports
[19:52:16] -!- shentey has joined #archlinux-ports
[20:10:00] -!- tyler has joined #archlinux-ports
[20:10:23] tyler is now known as Guest82280
[20:10:51] -!- Guest82280 has quit [Client Quit]
[20:11:18] -!- tyler1 has joined #archlinux-ports
[20:11:29] -!- tyler1 has quit [Client Quit]
[20:13:51] <buildmaster> hopenpgp-tools is broken (says rechenknecht).
[20:17:51] -!- tyler1 has joined #archlinux-ports
[20:19:51] -!- tyler1 has quit [Client Quit]
[20:21:26] -!- tyler1 has joined #archlinux-ports
[20:21:27] -!- tyler1 has quit [Client Quit]
[20:21:37] -!- oaken-source has quit [Ping timeout: 248 seconds]
[20:23:36] <buildmaster> postgrest is broken (says rechenknecht).
[20:30:26] -!- guys has joined #archlinux-ports
[20:31:08] <deep42thought> I'm afraid, all these haskell failures are due to the new db-update :-/
[20:32:20] -!- CalimeroTeknik has joined #archlinux-ports
[20:34:58] <buildmaster> firefox-developer-edition is broken (says buildknecht2).
[20:48:44] <tyzoid> deep42thought: I got some errors on a recent iso build, too
[20:48:55] <deep42thought> oh noez
[20:49:32] <tyzoid> One sec, grabbing log
[20:50:59] <deep42thought> what's "recent"? <24h?
[20:52:07] <tyzoid> Ran it about 2 hrs ago, so yeah
[20:52:15] <tyzoid> https://ptpb.pw
[20:52:35] <tyzoid> warning: cannot resolve "libvpx.so=5-32", a dependency of "ffmpeg"
[20:53:10] <tyzoid> It's possible it's a mirror state issue
[20:54:41] <tyzoid> Forced a mirror update, and I'm going to re-run the iso build to see
[20:56:03] <deep42thought> it's more probable, that the new db-update moved some packages it should not have moved
[20:56:29] <tyzoid> Generally, I like to make sure the issue isn't on my end, since 99% of the time it is
[20:56:38] <deep42thought> yes
[20:56:38] <tyzoid> But thought I'd give a heads up in case
[20:56:55] <deep42thought> but I can already see, where the issue is on my end :-(
[20:57:01] <tyzoid> ah
[20:57:06] <deep42thought> (mentally "see")
[20:57:17] <deep42thought> the database does not (yet) have versioned dependencies
[20:57:25] <tyzoid> Ah, makes sense
[20:57:33] <deep42thought> but here, something depends on "=5-32"
[20:57:36] <deep42thought> :-/
[20:57:43] -!- isacdaavid has quit [Quit: Leaving.]
[20:58:13] <tyzoid> In that case, I'll do the iso build out of the archive
[20:58:19] <tyzoid> *yay for the archive(
[20:58:31] <tyzoid> s/(/*/
[21:03:21] <deep42thought> on the other hand, we have enumerated dependencies of libvpx.so in the database
[21:08:03] -!- shentey has quit [Quit: Konversation terminated!]
[21:12:15] <deep42thought> hmm, extra/ffmpeg depends on libvpx.so.5 which is only provided from staging/libvpx
[21:12:19] <deep42thought> that seems wrong
[21:12:35] <deep42thought> and the buildmaster _knows_ it ...
[21:36:03] -!- |ast| has quit [Remote host closed the connection]
[21:36:21] <deep42thought> ah, no it's testing/libvpx ... sill wrong
[21:36:25] <deep42thought> s/sill/still/
[21:55:42] <tyzoid> deep42thought: And it's a new issue, using yesterday's mirror archive works fine
[21:56:06] <deep42thought> yes, it's something, db-update did wrong
[21:56:18] <deep42thought> but I'm too tired to usefully think about that, now :-/
[22:09:04] <tyzoid> No problem.
[22:14:24] -!- yans has joined #archlinux-ports
[22:25:57] -!- guys has quit [Ping timeout: 240 seconds]
[22:27:45] <tyzoid> deep42thought: Maybe some some sleep is in order?
[22:28:04] <deep42thought> yes, but I wanted to give my archlinux32 box another stress test
[22:28:06] <tyzoid> At some point, it might just make more sense to tackle in the morning
[22:28:10] <deep42thought> I suspect a failing hdd
[22:28:12] <tyzoid> ah
[22:28:21] -!- p71 has quit [Remote host closed the connection]
[22:28:39] <deep42thought> yeah, but in the morning, there will be two little distractions running around and pulling my legs ;-)
[22:28:54] <tyzoid> Yeah, that might be a small problem
[22:29:30] <tyzoid> deep42thought: How you manage to do this much work on arch32, your own work, and raise a family is beyond me
[22:30:05] <deep42thought> just don't think too hard about it - that's what I do
[22:30:15] <tyzoid> I can hardly keep up, and I (a) live at home and (b) have no kids
[22:30:20] <tyzoid> lol
[22:35:33] <deep42thought> AAAHHH
[22:35:50] * deep42thought is not too tired to find his own mistakes
[22:37:41] <tyzoid> :)
[22:38:02] <deep42thought> well, it is _some_ issue, but nothing which may have caused the current problem
[22:38:48] <deep42thought> we may move a package too early if it does not require removing a package - a package may be moved into a repository, where not all dependencies of that package are available
[22:39:28] <tyzoid> I see, but yeah, that shouldn't be the cause of the ffmpeg / vpx issue
[22:39:51] <deep42thought> the current logic is (should be): if a package is replaced by a newer version, check if anything still depends on the old one
[22:46:14] <deep42thought> well, it _can_ be this issue: if ffmpeg was moved prematurely
[22:47:25] <deep42thought> tyzoid: can you easily check if extra/ffmpeg was updated?
[22:47:41] <tyzoid> Between days?
[22:47:47] <deep42thought> yesterday and now
[22:48:25] * deep42thought should seriously set up a log of moved packages
[22:48:35] <tyzoid> No, ffmpeg did not change
[22:48:46] <tyzoid> both are showing 1:3.4.1-3.1
[22:49:05] <deep42thought> yeah, but now we have 3.4.2
[22:49:28] <tyzoid> in testing, not extra
[22:49:36] <deep42thought> +-------+------------------+------------+-------+--------+--------+------------+------------+-----------+---------+--------------+------------------+
[22:49:36] <deep42thought> | id | build_assignment | repository | epoch | pkgver | pkgrel | sub_pkgrel | has_issues | is_tested | pkgname | architecture | is_to_be_deleted |
[22:49:36] <deep42thought> +-------+------------------+------------+-------+--------+--------+------------+------------+-----------+---------+--------------+------------------+
[22:49:36] <deep42thought> | 33712 | 19249 | 2 | 1 | 3.4.2 | 1 | 0 | | | ffmpeg | 2 | |
[22:49:36] <deep42thought> | 43710 | 34603 | 5 | 1 | 3.4.2 | 2 | 0 | | | ffmpeg | 2 | |
[22:49:37] <deep42thought> +-------+------------------+------------+-------+--------+--------+------------+------------+-----------+---------+--------------+------------------+
[22:49:46] <deep42thought> 1:3.4.2-1.0 in extra
[22:49:54] <deep42thought> and 1:3.4.2-2.0 in testing
[22:50:17] <tyzoid> huh
[22:50:28] <deep42thought> extra/ffmpeg-1:3.4.2-1.0-i686.pkg.tar.xz
[22:50:39] <tyzoid> could it be the database is out of date?
[22:50:48] <tyzoid> I was getting the current info from the package api, which grabs from the database
[22:50:49] <deep42thought> the last one was from the master mirror
[22:50:56] <tyzoid> the filename itself is 3.4.2
[22:51:11] <deep42thought> maybe I fried the update sync-file script
[22:51:12] <deep42thought> lemme check
[22:51:40] <deep42thought> nope, lastupdate is current
[22:52:11] <deep42thought> is your mirror up to date?
[22:52:20] <tyzoid> Yeah
[22:52:26] <tyzoid> Again, the filename itself is showing 3.4.2
[22:52:34] <deep42thought> ummm
[22:52:35] <tyzoid> and the pkgapi goes and check against the master mirror
[22:52:46] <deep42thought> I *think* that the consistency of the database is checked, too
[22:52:47] <tyzoid> Try a `pacman -Syyi ffmpeg`
[22:53:27] <deep42thought> umm, I don't have a not-running-on-testing machine handy
[22:53:45] <deep42thought> Repositorium : extra
[22:53:45] <deep42thought> Name : ffmpeg
[22:53:45] <deep42thought> Version : 1:3.4.2-1.0
[22:53:49] <deep42thought> looks ok to me
[22:53:49] <tyzoid> one sec, I can patch you through to my already-running vm
[22:57:15] <deep42thought> https://github.com
[22:57:16] <phrik> Title: builder/bootstrap-mysql at master · archlinux32/builder · GitHub (at github.com)
[22:57:25] <deep42thought> at least, it's documented, now :-)
[23:03:42] -!- yans has quit [Quit: chaos is the only true answer]
[23:05:22] -!- yans has joined #archlinux-ports
[23:07:13] <deep42thought> good night, cu all later
[23:07:17] -!- deep42thought has quit [Quit: Leaving.]
[23:16:06] -!- guys has joined #archlinux-ports