| 01:40.46 | *** join/#asterisk-dev droemel (~droemel@p4FCAD33E.dip.t-dialin.net) |
| 07:26.07 | *** join/#asterisk-dev bulkorok (~bulkorok@85.183.36.36) |
| 07:33.37 | *** join/#asterisk-dev oej (~olle@2001:16d8:cc57:1000::42:1003) |
| 07:33.37 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:43.43 | *** join/#asterisk-dev oej (~olle@2001:16d8:cc57:1000::42:1005) |
| 08:43.43 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:56.25 | *** join/#asterisk-dev klobucar (uid1012@gateway/web/irccloud.com/x-kqhbqpzurcedvseu) |
| 08:57.19 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:806a:8215:b0d6:c9e9) |
| 08:58.04 | *** join/#asterisk-dev GameGamer43 (uid5533@gateway/web/irccloud.com/x-nlqqcqiahqqekcxc) |
| 09:00.02 | *** join/#asterisk-dev tonyclewis (uid6025@gateway/web/irccloud.com/x-cxxclsirdxtsosxt) |
| 09:02.25 | *** join/#asterisk-dev tonyclewis (uid6025@gateway/web/irccloud.com/x-unalsvecwbdwwgyr) |
| 09:03.29 | *** join/#asterisk-dev GameGamer43 (uid5533@gateway/web/irccloud.com/x-hialtbiitcqrsuzh) |
| 10:12.34 | oej | Hello everyone! |
| 10:29.44 | wdoekes | gomorron |
| 10:34.10 | oej | God morgon! |
| 10:47.08 | *** join/#asterisk-dev klobucar (uid1012@gateway/web/irccloud.com/x-lygjkbjkbdmudznn) |
| 10:54.24 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 10:54.24 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 11:10.41 | *** join/#asterisk-dev bombev (~bombev@PPPoE-Static-40-132.UnicsBG.Net) |
| 11:12.23 | bombev | hi, is there any way to use my Skype Incoming calls to go thru my PBX server? because Skype for asterisk (digium) is no longer available |
| 12:28.09 | *** join/#asterisk-dev GameGamer43 (uid5533@gateway/web/irccloud.com/x-vmgojfhqwzvscitd) |
| 12:39.54 | *** join/#asterisk-dev tonyclewis (uid6025@gateway/web/irccloud.com/x-tcotacnlqivxcxxm) |
| 12:42.43 | seanbright | are timers in asterisk independent? meaning if i create 10 timers that fire 10 times a second, will they all fire at the same time or will it be based on when they were ast_timer_open'd? |
| 12:43.02 | seanbright | i guess it could depend on the underlying timer implementation... |
| 13:15.55 | *** join/#asterisk-dev serafie (~erin@76.73.167.231) |
| 13:19.39 | *** join/#asterisk-dev bweschke (~bweschke@c-68-44-59-234.hsd1.pa.comcast.net) |
| 13:29.50 | *** join/#asterisk-dev frawd (~francois@221.red-80-28-139.adsl.static.ccgg.telefonica.net) |
| 13:49.11 | *** join/#asterisk-dev Victor_Yure (~IceChat77@unaffiliated/victoryure/x-837844) |
| 14:00.16 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-srjsayuiwiruvimn) |
| 14:01.54 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 14:01.54 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 14:01.57 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 14:01.57 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 14:04.52 | *** join/#asterisk-dev mattozzao (c8aac4ba@gateway/web/freenode/ip.200.170.196.186) |
| 14:06.40 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-bhfttwippfequgpb) |
| 14:06.40 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:20.47 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:20.47 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:35.28 | file | bananas! |
| 14:39.06 | jsmith | mangos? |
| 14:41.54 | Juggie | Pineapple? |
| 14:42.27 | WIMPy | knock, knock |
| 14:42.39 | pabelanger | on heaven door |
| 14:43.05 | russellb | Kiwiwiwiwiwiwi! |
| 14:43.39 | WIMPy | Hells door would be good enough. But it's hard to find from the inside. |
| 15:15.46 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-xleqlsgqxpucsgbb) |
| 15:41.24 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 15:41.24 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 15:45.28 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 15:45.28 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 16:16.51 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:24.09 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 16:24.09 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 16:31.58 | oej | file: Time for another question? |
| 16:32.07 | file | cookies! |
| 16:32.28 | oej | teleports a goat and a cookie to file |
| 16:33.15 | file | eep |
| 16:33.25 | oej | So I'm working on this generator. |
| 16:34.00 | oej | With a SNOM phone, it's turned on and off with periods of 80 ms during a call, saving a few packets. |
| 16:34.15 | oej | SO my generator is going to be activated many times per second. |
| 16:35.01 | oej | I need to save away pieces of data that I need to deactivate the generator and restore the channel. Allocating and deallocating a datastore for eveery change seems like a waste of resources. |
| 16:35.18 | oej | I can allocate once and reuse and let the channel destruction remove the leftovers. |
| 16:35.27 | oej | Or I can expand the ast_channel structure. |
| 16:35.31 | oej | What do you think? |
| 16:35.44 | file | don't expand the ast_channel structure |
| 16:36.03 | oej | So create a channel datastore and keep it hanging like old fruit? |
| 16:37.01 | file | if you want to persist data like that, then yes |
| 16:37.57 | oej | And I can trust the channel destruction to destroy all datastores.. Ok. |
| 16:38.06 | oej | Again, we have reached an agreement. |
| 16:38.27 | oej | Now, why not change the ast_channel? Just because you don't want to, or do we have a reson for it? |
| 16:38.48 | oej | In the early days, we were afraid of memory usage, but that can't be an issue on most systems today. |
| 16:39.54 | file | well that plus everything has to be recompiled in earlier versions, have to bump the cleancount to since you just changed the universe |
| 16:40.12 | oej | Earlier versions won't be affected by this new functionality. |
| 16:40.22 | oej | It's only going in to new releases. |
| 16:40.26 | oej | So that can't be an issue. |
| 16:40.34 | file | and datastores were created for this exact reason - so anything can attach arbitrary data and we don't have to keep changing the channel structure |
| 16:40.38 | oej | I do understand why we don't do it in existing releases |
| 16:40.46 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 16:40.46 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 16:41.27 | oej | But finding a datastore and using it is much slower than just adding an int to the channel structure and use that. |
| 16:41.52 | file | it's a trade off |
| 16:41.56 | oej | THis is far different than voicemail accounts and such that affect dialplan. |
| 16:42.09 | oej | This is going to be accessed many times per second during a call. |
| 16:42.29 | oej | Think about it and I'll start with a datastore. |
| 16:43.05 | oej | datastores are perfect for apps like minimv that wants to hang additional data on the channel. |
| 16:44.41 | file | 'tis just my opinion on things |
| 16:46.06 | oej | I'll start with your suggestion and then see if I can convince you about other solutions :-) |
| 16:46.20 | oej | Back later. |
| 16:46.55 | *** join/#asterisk-dev oej (~olle@2001:16d8:cc57:1000::42:1004) |
| 16:46.55 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 16:48.27 | *** join/#asterisk-dev malcolmd_ (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 16:48.27 | *** mode/#asterisk-dev [+o malcolmd_] by ChanServ |
| 17:14.41 | *** join/#asterisk-dev bweschke (~bweschke@74-94-3-209-Philadelphia.hfc.comcastbusiness.net) |
| 17:29.36 | *** join/#asterisk-dev pnlarsson (~niklas@fw1.gml.g.icnet.infracom.se) |
| 17:46.20 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 17:46.23 | *** part/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 18:04.16 | putnopvut | Here's a fun-time activity for you. Open the following link (it's a backtrace from a crash jsmith had recently): https://www.dropbox.com/sh/nc92pver8ixcytg/JTv2AWICxo/asterisk_backtrace_20121114_2080 |
| 18:04.30 | jsmith | It's a fun one :-) |
| 18:04.31 | putnopvut | Now find the crashing thread, and find the call to ast_queue_frame() within the crash thread. |
| 18:04.59 | putnopvut | Now grep through the file for the address of the channel being passed to ast_queue_frame() |
| 18:05.18 | putnopvut | You'll find a completely separate thread where that same address corresponds to a manager session. |
| 18:05.27 | putnopvut | Big ol' wtf right there. |
| 18:05.38 | jsmith | So you're saying I can't queue up manager sessions as frames? |
| 18:05.45 | jsmith | What kind of PBX is this?!? |
| 18:05.51 | jsmith | feigns surprise |
| 18:05.52 | jsmith | :-p |
| 18:06.05 | putnopvut | Nah, that's fine. You just can't queue up frames on manager sessions :) |
| 18:09.43 | *** join/#asterisk-dev citywok (~kvirc@67-134-194-33.dia.static.qwest.net) |
| 18:19.40 | citywok | mjordan: reember my segfault MWI issue? I'm experiencing lockups now (not segfaults). no core dump b/c i have to kill -9 it, but here's core show locks: http://pastebin.com/DWqbVAgy -- any ideas? should i post it as a new issue? what should i call it? thanks! |
| 18:40.42 | bweschke | putnopvut/jsmith: valgrind time? |
| 18:41.19 | jsmith | bweschke: On a production server with several hundred agents? Surely you jest... :-) |
| 18:41.36 | jsmith | bweschke: That being said, I'd love help learning how to setup valgrind properly in a lab environment |
| 18:43.02 | leifmadsen | sputters and coughs something about devops |
| 18:43.03 | bweschke | jsmith: sorry… missed the part about production server in several hundred agents in the thread… yep… I jest there and don't call me surely. |
| 18:45.19 | jsmith | leifmadsen: I'm all for devops -- but I inherited a system I didn't design. Said system has problems, and I'm doing my best to try to solve said problems. |
| 18:47.33 | bweschke | jsmith: is said system using chan_agent too? |
| 18:48.16 | jsmith | bweschke: Of course :-/ |
| 18:48.40 | jsmith | bweschke: Over 200 queues, around 400 agents last time I checked -- it's ugly. |
| 18:49.13 | bweschke | jsmith: yep. sounds like a ticking timebomb |
| 18:49.21 | bweschke | jsmith: sorry man |
| 18:49.44 | jsmith | C'est la vie... we'll figure it out, or go down trying |
| 18:50.39 | mjordan | citywok: what version of 1.8? And that would be a new issue - but one we've seen before, so I'm surprised it turned back up |
| 18:50.58 | citywok | a 2 week old copy of svn |
| 18:51.05 | mjordan | citywok :-( |
| 18:51.21 | mjordan | citywok: Please file an issue with that core show locks and - if you can a gdb bt full - asap |
| 18:51.25 | citywok | i just updated to the latest svn at the crash |
| 18:51.36 | citywok | mjordan: no core dump, kill -9 doesn't give one afaik? |
| 18:51.45 | mjordan | citywok: attach to the process with gdb |
| 18:51.50 | mjordan | gdb -p [pid] |
| 18:52.04 | citywok | oh, i didn't know you could do that. dang. okay, next time it crashes i'll do that. |
| 18:52.10 | citywok | this is the 3rd time in about 2 weeks |
| 18:52.15 | mjordan | the core show locks is enough to get started on |
| 18:52.37 | putnopvut | That is a big ol' core show locks right there. |
| 18:52.37 | citywok | what should i name the ticket? i don't really know. lol. |
| 18:52.49 | citywok | there were 30 or 40 channels trying to dial |
| 18:52.53 | mjordan | deadlock when servicing MWI event |
| 18:52.56 | citywok | an agent hit dial, it would try and spool the call |
| 18:53.17 | citywok | nothing happened... so they hit dial again... and then we have a bunch of dummy calls |
| 18:53.37 | putnopvut | scrolling...scrolling...who has the damn event lock?! |
| 18:54.05 | mjordan | putnopvut: the lock is held during an event callback |
| 18:54.06 | putnopvut | oh, the top thread |
| 18:54.10 | mjordan | yeah. |
| 18:54.27 | putnopvut | Ah, it's just the top two threads involved in the deadlock. |
| 18:54.38 | mjordan | holding locks during an event callback - where you don't know where you're going to go off to - has caused issues in the past. |
| 18:55.55 | putnopvut | Yeah, probably should find the relevant callbacks, unlock, call the callbacks. |
| 18:56.08 | putnopvut | writes a patch. |
| 18:56.39 | leifmadsen | jsmith: oh I hear ya -- but your statement of wanting to know how to build a development system for testing is done via a devops model :) (believe me, I know what it's like to be fighting an issue without a net) |
| 18:56.47 | citywok | mjordan: https://issues.asterisk.org/jira/browse/ASTERISK-20682 -- locks & threads included |
| 18:57.22 | jsmith | leifmadsen: I know how to build a dev system -- I was specifically looking for more detailed information on setting up valgrind to work well with Asterisk |
| 18:57.41 | mjordan | putnopvut: you'd have to copy the subscription information |
| 18:57.50 | mjordan | putnopvut: but yes, that would solve a lot of nasty deadlock problems |
| 18:57.59 | mjordan | citywok: thanks |
| 19:02.17 | citywok | mjordan: thanks for your help :) -- stupid lockups are annnnnnoying. lol |
| 19:07.51 | *** join/#asterisk-dev wedhorn (~wedhorn@123-243-209-34.static.tpgi.com.au) |
| 19:13.52 | *** join/#asterisk-dev oej (~olle@h87-96-134-129.dynamic.se.alltele.net) |
| 19:13.52 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 19:50.51 | putnopvut | citywok: patch attached |
| 19:51.46 | citywok | that was quick, thanks guys! -- I'll apply it tonight. |
| 19:52.07 | putnopvut | Heh. It won't always be that way, but sometimes an issue really grabs me and I must fix it. |
| 19:52.27 | citywok | Yea, I'm aware. Hence the thanks for doing it so quickly! :p |
| 19:52.34 | citywok | gracias sir |
| 19:52.40 | putnopvut | de nada |
| 19:53.35 | citywok | btw in the past i've had epic numbers of invalid ast_refcount errors being spit out to the console/logs |
| 19:53.41 | citywok | i didn't check on this crash, probably shoulda |
| 19:54.03 | citywok | oh, nm, i think that's what happened when MWI caused a segfault, not a deadlock |
| 21:09.55 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 21:09.55 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 23:04.11 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-bhfttwippfequgpb) |
| 23:34.46 | *** join/#asterisk-dev moy (~moy@unvlon55-1176057127.sdsl.bell.ca) |
| 23:39.00 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 23:39.00 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |