| 08:04.30 | *** join/#asterisk-bugs infobot (~infobot@rikers.org) |
| 08:04.30 | *** topic/#asterisk-bugs is Asterisk Bugs Discussion -=- https://issues.asterisk.org/jira -=- http://xkcd.org/583/ |
| 08:26.37 | *** join/#asterisk-bugs irroot (~irroot@196-215-124-168.dynamic.isadsl.co.za) |
| 08:44.43 | *** join/#asterisk-bugs irroot (~irroot@196-215-124-168.dynamic.isadsl.co.za) |
| 11:31.41 | *** join/#asterisk-bugs irroot (~irroot@41.221.5.170) |
| 11:57.51 | *** join/#asterisk-bugs acidfoo (~nib@modemcable094.94-70-69.static.videotron.ca) |
| 12:05.26 | *** join/#asterisk-bugs DocAwesome (~DocAwesom@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:05.26 | *** mode/#asterisk-bugs [+o DocAwesome] by ChanServ |
| 12:19.25 | *** join/#asterisk-bugs DocAwesome (~DocAwesom@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:19.25 | *** mode/#asterisk-bugs [+o DocAwesome] by ChanServ |
| 13:05.08 | *** join/#asterisk-bugs serafie (~erin@nat/digium/x-lrmlohvmjrsizoxj) |
| 13:40.53 | leifmadsen | ASTERISK-18420 |
| 13:40.56 | DocAwesome | [ASTERISK-18420] [Status: Triage] [regression] Inbound ISDN Overlap dial breaks with Asterisk Version >= 1.8.4.0 - https://issues.asterisk.org/jira/browse/ASTERISK-18420 |
| 13:41.00 | leifmadsen | nice still works! :) |
| 13:53.06 | pabelanger | tips his hat to leifmadsen |
| 13:53.13 | pabelanger | well played sir |
| 14:04.25 | leifmadsen | pabelanger: w00t |
| 14:04.59 | leifmadsen | pabelanger: I wrote a small bash script this morning to check via cronjob every 5 mins to make sure the bot was running |
| 14:05.33 | pabelanger | leifmadsen: what is it written in? |
| 14:06.07 | Qwell | blinks |
| 14:06.11 | Qwell | my guess - bash! |
| 14:06.32 | pabelanger | heh, I meant the bot |
| 14:06.33 | leifmadsen | :) |
| 14:06.36 | leifmadsen | bot is in python |
| 14:06.44 | pabelanger | twisted? |
| 14:06.46 | leifmadsen | it's in repotools |
| 14:06.50 | leifmadsen | I don't think it uses twisted |
| 14:06.55 | leifmadsen | jiraircbot |
| 14:07.11 | leifmadsen | uses SimpleServerIRCBot or something like that |
| 14:08.35 | pabelanger | Blah, we need to convert that bad boy to use twisted! http://twistedmatrix.com/documents/current/core/howto/clients.html#auto5 |
| 14:10.59 | leifmadsen | we? |
| 14:11.01 | leifmadsen | :) |
| 14:13.52 | pabelanger | I see what you did there |
| 14:17.11 | leifmadsen | :) |
| 14:31.12 | *** join/#asterisk-bugs The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:42.21 | *** part/#asterisk-bugs jsmith (~jsmith@fedora/jsmith) |
| 17:20.01 | *** join/#asterisk-bugs singler (~singler@78-60-139-121.static.zebra.lt) |
| 18:16.26 | JasonL | anyone have any ideas on ASTERISK-18487? It just keeps happening |
| 18:16.28 | DocAwesome | [ASTERISK-18487] [Status: Triage] Daily deadlock issue - https://issues.asterisk.org/jira/browse/ASTERISK-18487 |
| 18:21.28 | pabelanger | JasonL: try using 1.8.7.0-rc1 |
| 18:22.36 | pabelanger | it contains a few fixes for a performance regression, plus timing modules |
| 18:23.37 | JasonL | i installed it today, and same thing just happened |
| 18:24.54 | pabelanger | JasonL: how are you starting asterisk? |
| 18:25.50 | JasonL | asterisk -g |
| 18:26.10 | pabelanger | JasonL: what is the output of *CLI> core show locks |
| 18:28.46 | JasonL | I don't have debug threads enabled because the system can't seem to keep up with the load if it's turned on |
| 18:29.24 | JasonL | I did confirm that 1.8.7.0-rc1 fixes huge memory leaks in app_queue, which I thought was the problem, but it still happened |
| 18:29.30 | pabelanger | Ya, that is going to make it hard to figure out which thread is hold another |
| 18:29.49 | JasonL | any suggestions on how I should proceed to debug? |
| 18:30.04 | pabelanger | How easy is it to reproduce? |
| 18:30.22 | JasonL | i have never been able to reproduce myself as I dont really know whats causing it.. |
| 18:30.30 | JasonL | it just happens on it's own 1-2 times daily |
| 18:32.44 | pabelanger | Well, like I said, core show locks is the first step. If you system cannot handle it with DEBUG_THREADS, either scale now the amount of calls on asterisk or upgrade the system? |
| 18:34.20 | JasonL | alright |
| 18:34.35 | JasonL | we added a second Xeon, to see if that would help, but nothing |
| 18:34.58 | JasonL | system is never using much CPU, it had 16GB of ram, dunno what else to do to upgrade it, hehe |
| 18:35.04 | JasonL | has* |
| 18:36.34 | pabelanger | Is this a VM? |
| 18:40.22 | JasonL | no it's an HP server |
| 18:40.29 | JasonL | Running Debian 64bit |
| 18:43.45 | JasonL | would i get anywhere opening a support ticket with Digium? |
| 18:44.20 | JasonL | if I buy a support package |
| 18:45.07 | pabelanger | possible, you'll have to ask somebody in Digium sales |
| 18:47.00 | JasonL | something else to add, when the problem is happening, if I do a "queue show" nothing happens |
| 18:47.12 | JasonL | and I am using asterisk real time queues |
| 18:47.27 | JasonL | would going file based make a difference? |
| 18:48.44 | pabelanger | Well, you'd be using a difference code path |
| 18:48.52 | pabelanger | So it is possible |
| 18:53.38 | *** join/#asterisk-bugs oej (~olle@2001:470:1f15:d79:1426:b5c4:b598:4131) |
| 18:55.17 | *** join/#asterisk-bugs oej_ (~olle@ns.webway.se) |
| 18:57.24 | *** join/#asterisk-bugs irroot (~irroot@197.173.241.226) |
| 18:58.49 | JasonL | pabelanger: thank you for your help |
| 19:47.03 | *** join/#asterisk-bugs oej (~olle@ns.webway.se) |
| 20:20.56 | *** join/#asterisk-bugs malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 20:20.56 | *** mode/#asterisk-bugs [+o malcolmd] by ChanServ |
| 20:23.34 | irroot | ASTERISK-18492 ASTERISK-18491 these are related and resolved |
| 20:23.37 | DocAwesome | [ASTERISK-18492] [Status: Triage] Deadlock in channel.c / chan_sip.c - https://issues.asterisk.org/jira/browse/ASTERISK-18492 |
| 20:23.37 | DocAwesome | [ASTERISK-18491] [Status: Triage] Deadlock on chan_sip / MASTER_CHANNEL - https://issues.asterisk.org/jira/browse/ASTERISK-18491 |
| 20:33.35 | JasonL | irroot: what version are those resolved in? |
| 20:34.16 | irroot | 1.8.7-rc1 + |
| 20:34.29 | irroot | the patch was added after 1.8.6 |
| 20:34.33 | JasonL | ok, sounds like the issue i was having, but I had it again today with 1.8.7-rc1 |
| 20:35.26 | irroot | svn commit 332320 |
| 20:35.57 | irroot | if you core show locks and there locks with a ast_read that stays arround this is it |
| 20:36.21 | irroot | i looked at your output there was no "timerfd" in there on search |
| 20:37.14 | irroot | Numerous isues have been reported for deadlocks that are caused by |
| 20:37.16 | irroot | a blocking read in res_timing_timerfd on a file descriptor that will |
| 20:37.18 | irroot | never be written to. This patch adds some checks to make sure that |
| 20:37.19 | JasonL | ya i just searched that :) |
| 20:37.20 | irroot | the timerfd is both valid and armed before calling read(). |
| 20:37.45 | JasonL | so its not what I am experiencing then |
| 20:58.53 | *** join/#asterisk-bugs malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 20:58.53 | *** mode/#asterisk-bugs [+o malcolmd] by ChanServ |
| 21:48.29 | *** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 21:48.29 | *** mode/#asterisk-bugs [+o leifmadsen] by ChanServ |
| 22:08.08 | *** join/#asterisk-bugs elguero (~miguel323@2001:470:1f06:12c4::2) |
| 22:32.08 | *** join/#asterisk-bugs seraphie (~erin@75.76.38.159) |
| 23:10.12 | *** join/#asterisk-bugs Juggie (~Juggie@unaffiliated/juggie) |
| 23:10.12 | *** mode/#asterisk-bugs [+o Juggie] by ChanServ |
| 23:10.42 | *** join/#asterisk-bugs Juggie (~Juggie@unaffiliated/juggie) |
| 23:10.42 | *** mode/#asterisk-bugs [+o Juggie] by ChanServ |