| Monday, June 20th, 2011, 08:13 UTC | ||
| [08:13:35] | stuarta: | well i looks like my buildslave is behaving. bit slow at times but it's working. must not like doing more than 1 build at a time (which it's now limited to) |
| [08:29:50] | stuarta: | on the bright side, its generally taking less time to compile than the PPC builder, which is what i would expect |
| [08:54:45] | Beirdo: | heh |
| [08:55:05] | Beirdo: | what was the motivation for another? just better distro coverage? |
| [08:57:36] | kth (kth!~kth@unaffiliated/kth) has joined #mythtv | |
| [08:57:38] | stuarta: | that, and getting familiarity with what it does on the slave side |
| [08:57:49] | Beirdo: | cool |
| [08:57:51] | stuarta: | so that i can man handle the OSX build to fit that model |
| [08:58:14] | stuarta: | since i need to convert it from build everything including qt |
| [08:58:26] | stuarta: | to boostrap everything including qt, then buildslave mythtv |
| [08:59:14] | Beirdo: | well, a makefile to build the bootstrap part would work |
| [08:59:40] | stuarta: | it'll be 85% osx-packager |
| [08:59:47] | Beirdo: | then it can run the makefile each time, and not have to rebuild as there's no change |
| [09:00:10] | stuarta: | i'm going to make it another script in the packaging repo |
| [09:00:20] | Beirdo: | ummm |
| [09:00:22] | stuarta: | so it'll need to pull that each time, which is simple |
| [09:00:25] | Beirdo: | there's a small issue |
| [09:00:39] | Beirdo: | buildbot will be only watching one repo |
| [09:00:48] | stuarta: | yeah that's fine |
| [09:00:59] | stuarta: | not bothered by that |
| [09:01:27] | stuarta: | if i do update dependencies then the build can always be forced to get the rebuild underway |
| [09:01:28] | Beirdo: | so you'll make it manually do a pull on the packaging repo? |
| [09:01:34] | stuarta: | aye |
| [09:01:43] | Beirdo: | should work, I guess ;) |
| [09:01:50] | ** stuarta knows it'll work ** | |
| [09:02:14] | Beirdo: | hehe, that's the spirit |
| [09:02:34] | stuarta: | bending build systems to do my will is one of my many talents |
| [09:02:41] | stuarta: | :) |
| [09:02:56] | Beirdo: | hehe. Yeah, it usually works fairly well for me too |
| [09:03:14] | Beirdo: | if you run out of ideas or anything, feel free to bounce it off me |
| [09:03:44] | Beirdo: | I've been chasing the lame mysql exit messages |
| [09:03:48] | stuarta: | yeah, no worries. nothing like actually doing it |
| [09:03:54] | stuarta: | hence the extra buildd |
| [09:04:08] | Beirdo: | I think it's the Qt MySQL drivers not using the C API quite right |
| [09:04:19] | stuarta: | joy |
| [09:04:42] | Beirdo: | they have a big fat warning that you can't share connections across threads, blah blah blah |
| [09:05:22] | Beirdo: | the message is a nuisance, but we can live. it's the 5s extra delay that offends me most :) |
| [09:05:45] | kth (kth!~kth@unaffiliated/kth) has quit (Quit: Leaving.) | |
| [09:05:46] | kth1 (kth1!~kth@dyndsl-085-016-232-243.ewe-ip-backbone.de) has joined #mythtv | |
| [09:07:43] | martin__ (martin__!~quassel@static-88.131.29.2.addr.tdcsong.se) has joined #mythtv | |
| [09:07:59] | Beirdo: | well, I'll fight it again tomorrow |
| [09:08:08] | Beirdo: | tilting at windmills is just so fun |
| [09:09:27] | stuarta: | that's an image i'd never previously considered |
| [09:14:08] | kth1 (kth1!~kth@dyndsl-085-016-232-243.ewe-ip-backbone.de) has quit (Ping timeout: 244 seconds) | |
| [09:14:19] | Beirdo: | hehe |
| [09:14:43] | Beirdo: | it's actually from classic literature (Don Quixote, IIRC) |
| [09:15:39] | Beirdo: | http://bugreports.qt.nokia.com/browse/QTBUG-2365 |
| [09:15:40] | stuartm: | yeah, it's Don Quijote de la Mancha |
| [09:15:46] | Beirdo: | my ass it's closed |
| [09:16:12] | Beirdo: | fixed in 4.5.2, eh? |
| [09:16:27] | Beirdo: | I guess they unfixed it later, or there's yet another thing causing it |
| [09:17:48] | stuarta: | defn coming from mysql? |
| [09:19:12] | Beirdo: | yes |
| [09:19:12] | kth (kth!~kth@unaffiliated/kth) has joined #mythtv | |
| [09:19:21] | Beirdo: | Error in my_thread_global_end(): 5 threads didn't exit |
| [09:19:30] | Beirdo: | or 1 threads |
| [09:19:40] | Beirdo: | the error is from the mysql C API |
| [09:20:02] | Beirdo: | and it's when the threads aren't shutdown in an orderly fashion somehow. |
| [09:20:37] | Beirdo: | sigh |
| [09:20:58] | Beirdo: | but I'll take another run at it tomorrow, including reading over some Qt code |
| [09:22:33] | Beirdo: | which is always fun |
| [09:22:44] | Beirdo: | but I need sleep so I can go to work tomorrow |
| [09:23:09] | Beirdo: | night, and good luck with your OSX fun |
| [09:23:11] | ** stuarta waves nnn ** | |
| [09:27:42] | stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has quit (Ping timeout: 240 seconds) | |
| [09:27:57] | stuartm (stuartm!~stuartm@cpc4-derb9-0-0-cust534.8-3.cable.virginmedia.com) has joined #mythtv | |
| [09:27:57] | stuartm (stuartm!~stuartm@cpc4-derb9-0-0-cust534.8-3.cable.virginmedia.com) has quit (Changing host) | |
| [09:27:58] | stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has joined #mythtv | |
| [09:28:01] | stuartm: | Beirdo: should we be explicitly closing database connections when we're finished with them and that's just not happening in some places? |
| [09:29:42] | Beirdo: | heh |
| [09:29:48] | Beirdo: | I just had the same thought. |
| [09:30:13] | Beirdo: | I may need to explicitly close the db connection *from the same thread that opened it* |
| [09:30:37] | Beirdo: | we are closing them in the dtors right now |
| [09:30:59] | Beirdo: | but that will be run from the corecontext (UI) thread on shutdown |
| [09:31:09] | Beirdo: | hmmm |
| [09:31:17] | Beirdo: | let my try that quickly so I can sleep |
| [09:31:18] | Beirdo: | :) |
| [09:35:43] | Beirdo: | otherwise I'll just be lying there thinking about it for hours |
| [09:35:50] | Beirdo: | instead of the 10min it takes to try it |
| [09:39:49] | Beirdo: | OMG |
| [09:39:54] | Beirdo: | :) |
| [09:40:02] | Beirdo: | OK, I can now go to sleep happy. |
| [09:40:16] | Beirdo: | that worked just fine |
| [09:40:27] | stuartm: | yay |
| [09:41:00] | Beirdo: | soooo. Anything using the database not from CoreContext/UI thread... must close the DB connection explicitly or it will be a pain |
| [09:41:04] | Beirdo: | good to know |
| [09:41:22] | Beirdo: | especially as DB logging is on by default, and falls in that category |
| [09:41:54] | mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 252 seconds) | |
| [09:43:19] | Beirdo: | so tomorrow, I'll go look for the other cases (I think it's the programinfo updater and the scheduler and maybe datadirect) |
| [09:43:48] | Beirdo: | and then tidy up and commit what will at least cause it to happen less often, if not eradicate it. |
| [09:43:58] | Beirdo: | yay. And once again, good night :) |
| [09:44:26] | stuartm: | eit thread |
| [09:44:46] | stuartm: | probably, since I see those errors regularly appearing in the logs and not just on shutdown |
| [09:45:56] | stuartm: | oh, and good night ;) |
| [09:50:19] | stuartm: | oh cool, they've rolled out my upstream upgrade 2 months early |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.