| 01:14.31 | *** join/#asterisk-bugs infobot (~infobot@rikers.org) |
| 01:14.31 | *** topic/#asterisk-bugs is Asterisk Bugs Discussion -=- http://issues.asterisk.org/ -=- http://asterisk.org/developers/bug-guidelines - Discussions not related to reported issues should be in #asterisk -=- Join #asterisk-testing for help with testing issues marked Ready for Testing -=- http://xkcd.org/583/ -=- An issue tracker full of activity is a sign of a healthy project |
| 01:14.56 | Marquis42 | I started to get suspicous when the comment for the update_status function in app_queue did not even remotely match its actual functionality. So I started trying to figure out when that changed. |
| 01:15.19 | leifmadsen | interesting |
| 01:15.24 | leifmadsen | M12979 |
| 01:15.26 | MuffinMan | [closed] [Asterisk] Applications/app_queue 0012979: [patch] app_queue does not handle Custom: state interfaces reported by sigxcpu https://issues.asterisk.org/view.php?id=12979 |
| 01:15.33 | leifmadsen | you're looking at 1.6.2 or later? |
| 01:15.59 | Marquis42 | I was digging through the log on trunk, actually. |
| 01:16.17 | leifmadsen | ok that's what I'm looking at as well (trunk) and this code is in 1.6.2 or later |
| 01:16.21 | Marquis42 | I'm using 1.6.2, personally, for the call center that I'm having this issue with. |
| 01:16.21 | leifmadsen | (it was trunk at that point) |
| 01:16.28 | Marquis42 | Right |
| 01:16.31 | leifmadsen | ok, so yes it would affect your 1.6.2 version |
| 01:16.37 | leifmadsen | (1.6.0 and 1.6.1 had different changes) |
| 01:16.44 | leifmadsen | according to the log |
| 01:16.57 | Marquis42 | Yes, although I haven't looked at them specifically yet. |
| 01:17.06 | leifmadsen | ya and I wouldn't bother |
| 01:17.09 | leifmadsen | they are not supported |
| 01:17.14 | Marquis42 | indeed |
| 01:17.37 | Marquis42 | And really, 1.6.2 isn't quite soon either. But the issue is in 1.8 and trunk as well, so a fix is needed. |
| 01:17.39 | leifmadsen | interesting -- might be worth opening a new issue to discuss some of your thoughts and mark it as related to 12979 |
| 01:17.47 | Marquis42 | Will do. |
| 01:17.57 | leifmadsen | ya, if we can get it fixed soonish, then it can make it in for 1.6.2 and forward |
| 01:17.59 | leifmadsen | which would be ideal |
| 01:18.15 | Marquis42 | What do we have, 1 day? ;) |
| 01:19.02 | leifmadsen | actually ya, I just realized that ;) |
| 01:19.19 | leifmadsen | it'll be a BIT longer than 1 day because I figure we'll tag RC1 after this print which ends on Friday |
| 01:19.24 | Marquis42 | Yeah, I don't think I have the bandwidth (skills? :P ) to fix it before then. |
| 01:19.44 | leifmadsen | fair enough |
| 01:20.01 | leifmadsen | will be worth mentioning on asterisk-dev mailing list tonight/tomorrow for sure though |
| 01:20.16 | leifmadsen | if anything it should get fixed in 1.8 and forward |
| 01:20.22 | Marquis42 | Definitely, yes. |
| 01:20.29 | leifmadsen | I guess 1.6.2 would be a nice to have :) |
| 01:20.35 | Marquis42 | Yeah |
| 01:20.56 | leifmadsen | ok, I'm done sitting in front of the computer -- been here since 7am |
| 01:21.57 | leifmadsen | half a presentation done for Thursday morning.... 42 slides thus far |
| 01:21.57 | Marquis42 | The fix I have in mind is moderately invasive, though, rather than just restoring the exact code removed. Probably not a big issue for 1.8 since it'll be supported for some time yet, but I'm not sure how I'd feel about throwing something like that in a day or two before a release goes unsupported. Perhaps for 1.6.2 we *should* just revert the change? idk, just thinking out loud. |
| 01:22.06 | Marquis42 | heh |
| 01:22.24 | leifmadsen | Marquis42: sure -- we should target putnopvut directly tomorrow |
| 01:22.43 | Marquis42 | Sounds good. He's in Alabama with the Digium crew right? |
| 01:23.09 | *** join/#asterisk-bugs ccesario (~ccesario@201-42-156-58.dsl.telesp.net.br) |
| 01:23.52 | Marquis42 | Or more succintly, approximately 9:00am EST I can harass him right? ;) |
| 01:26.47 | leifmadsen | Marquis42: probably 10:00am EDT |
| 01:27.04 | leifmadsen | he's in CDT (Digium yes) |
| 01:27.42 | Marquis42 | OK, thanks. |
| 01:27.50 | Marquis42 | Now back to your slides! :) |
| 01:28.02 | Marquis42 | (oh, and thanks for letting me bounce all this off you) |
| 02:02.43 | *** join/#asterisk-bugs miguel3239 (~miguel323@17.140.211.66.inaddr.G4.NET) |
| 03:29.27 | *** join/#asterisk-bugs elguero (~miguel323@17.140.211.66.inaddr.G4.NET) |
| 04:01.11 | *** join/#asterisk-bugs Hyper-Core (~lol@71.31.117.174) |
| 04:13.35 | *** part/#asterisk-bugs Hyper-Core (~lol@71.31.117.174) |
| 05:41.47 | *** join/#asterisk-bugs oej (~olle@2001:470:1f15:d79:d918:7969:cb20:2cd3) |
| 07:08.56 | *** join/#asterisk-bugs tzafrir_laptop (~tzafrir@local.xorcom.com) |
| 07:21.33 | *** join/#asterisk-bugs oej (~olle@ns.webway.se) |
| 08:55.20 | *** join/#asterisk-bugs infobot (~infobot@rikers.org) |
| 08:55.20 | *** topic/#asterisk-bugs is Asterisk Bugs Discussion -=- http://issues.asterisk.org/ -=- http://asterisk.org/developers/bug-guidelines - Discussions not related to reported issues should be in #asterisk -=- Join #asterisk-testing for help with testing issues marked Ready for Testing -=- http://xkcd.org/583/ -=- An issue tracker full of activity is a sign of a healthy project |
| 09:06.19 | *** join/#asterisk-bugs oej (~olle@ns.webway.se) |
| 09:06.46 | *** join/#asterisk-bugs oej (~olle@ns.webway.se) |
| 09:33.29 | *** join/#asterisk-bugs vfabi (~fabi@h2-92.faust.net.ua) |
| 11:29.55 | *** join/#asterisk-bugs bbryant (~brett@c-76-26-221-79.hsd1.sc.comcast.net) |
| 11:29.55 | *** mode/#asterisk-bugs [+o bbryant] by ChanServ |
| 13:33.38 | *** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:33.38 | *** mode/#asterisk-bugs [+o leifmadsen] by ChanServ |
| 14:00.45 | *** join/#asterisk-bugs The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:13.58 | *** join/#asterisk-bugs putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:13.58 | *** mode/#asterisk-bugs [+o putnopvut] by ChanServ |
| 14:22.26 | *** join/#asterisk-bugs vfabi (~fabi@h2-92.faust.net.ua) |
| 14:28.23 | *** join/#asterisk-bugs vfabi (~fabi@h2-92.faust.net.ua) |
| 15:26.11 | *** join/#asterisk-bugs elguero (~miguel323@12.187.84.162) |
| 18:19.03 | *** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 18:19.03 | *** mode/#asterisk-bugs [+o leifmadsen] by ChanServ |
| 18:35.00 | *** join/#asterisk-bugs bbryant (~brett@c-76-26-221-79.hsd1.sc.comcast.net) |
| 18:35.00 | *** mode/#asterisk-bugs [+o bbryant] by ChanServ |
| 18:49.02 | *** join/#asterisk-bugs JonathanRose (~jonathan@nat/digium/x-qviqsyktbxyghxkn) |
| 18:51.02 | *** join/#asterisk-bugs mvanbaak (~michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 19:06.47 | leifmadsen | Marquis42: o/ |
| 19:07.12 | Marquis42 | ohai |
| 19:07.50 | leifmadsen | putnopvut: ohai to you :) |
| 19:07.53 | leifmadsen | Marquis42: and you :) |
| 19:08.12 | putnopvut | HOQQDY |
| 19:08.15 | putnopvut | Or Hi |
| 19:08.17 | putnopvut | whichever :) |
| 19:08.21 | leifmadsen | :D |
| 19:08.44 | leifmadsen | M12979 |
| 19:08.46 | MuffinMan | [closed] [Asterisk] Applications/app_queue 0012979: [patch] app_queue does not handle Custom: state interfaces reported by sigxcpu https://issues.asterisk.org/view.php?id=12979 |
| 19:08.52 | Marquis42 | Oh yeah, app_queue fun! :) |
| 19:08.57 | leifmadsen | putnopvut: Marquis42 may have found a race condition in that code |
| 19:10.04 | putnopvut | Neat! |
| 19:10.07 | Marquis42 | heh |
| 19:10.35 | leifmadsen | I wonder if I should copy/paste what we discussed last night to a pastebin |
| 19:10.38 | leifmadsen | I think I will |
| 19:10.40 | Marquis42 | It seems to me, unless my analysis is wrong, that there now exists the possibility for an agent to be rung by two (or possibly more) calls simultaneously. |
| 19:10.50 | Marquis42 | Yeah, that might save us some time. |
| 19:11.50 | leifmadsen | http://pastebin.com/kwtYx0S8 |
| 19:11.57 | leifmadsen | putnopvut: that'll catch you up (whenever you have time) |
| 19:13.32 | putnopvut | omg wall of text :( |
| 19:14.03 | Marquis42 | I can summarize if you'd like. |
| 19:15.01 | putnopvut | nah it's cool. reading now. |
| 19:15.04 | Marquis42 | k |
| 19:16.58 | putnopvut | Marquis42: So what's your plan of attack? |
| 19:17.18 | putnopvut | Hearing the idea of having a global member list sounds kind of like what used to be done in 1.4 for some certain things. |
| 19:17.21 | Marquis42 | Well, obviously we don't want a big linked-list of members to iterate over every time like the old code... |
| 19:19.05 | Marquis42 | However, I *think* that have an ao2 container of them would be ok in general. It would at least be much less expensive. I was considering, although maybe this part would have to be in trunk, having the global container share the ao2 object for member info and splitting out the per-queue information (penalty, etc.). |
| 19:22.42 | Marquis42 | I'm certainly open to other ideas as well, that's just what was rattling around my head. But it seems to me that there needs to be some kind of module-wide barrier to prevent the simultaneous call issue. |
| 19:49.52 | putnopvut | Yeah, one of the biggest issues we've always had in queues is that if Member A is a member of Queues B and C, then both queues have COPIES of A rather than them both sharing a reference to a single copy of A. |
| 19:50.03 | putnopvut | That just inevitably leads to crap. |
| 19:50.15 | *** join/#asterisk-bugs mvanbaak (~michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 19:50.17 | Marquis42 | Right, hence my thought to unify the common information. |
| 19:53.09 | Entomologist | *** CLOSED (19158) [Channels/General] Command "Show Channels" reports incorrect number of active calls in comparison to active channels |
| 19:53.10 | Entomologist | Reported by: pguillem |
| 19:53.10 | Entomologist | https://issues.asterisk.org/view.php?id=19158 |
| 19:53.11 | Entomologist | ********************************************************* |
| 19:53.30 | Entomologist | *** CLOSED (19160) [Channels/General] Command "Show Channels" reports incorrect number of active calls in comparison to active channels |
| 19:53.30 | Entomologist | Reported by: pguillem |
| 19:53.31 | Entomologist | https://issues.asterisk.org/view.php?id=19160 |
| 19:53.31 | Entomologist | ********************************************************* |
| 19:54.00 | Entomologist | *** CLOSED (19157) [Codecs/NewFeature] AMR-NB codec. |
| 19:54.01 | Entomologist | Reported by: snapple42 |
| 19:54.01 | Entomologist | https://issues.asterisk.org/view.php?id=19157 |
| 19:54.02 | Entomologist | ********************************************************* |
| 19:59.00 | Entomologist | *** CLOSED (19154) [Channels/chan_sip/General] directmedia or reinvite not working when calling extension that's located an a different asterisk node |
| 19:59.01 | Entomologist | Reported by: kwk |
| 19:59.01 | Entomologist | https://issues.asterisk.org/view.php?id=19154 |
| 19:59.02 | Entomologist | ********************************************************* |
| 20:08.57 | putnopvut | It's an idea I'd be all for, and I imagine most users of queues would be too. It's just that it would definitely need scrutiny and tests written due to its invasiveness. |
| 20:10.04 | Marquis42 | Most assuredly, yes. |
| 20:12.51 | Marquis42 | I guess the question I have is more if the approach seems correct. Obviously the code needs to be written, reviewed, and tested. And, also, is it *too* invasive for 1.8? There is clearly a bug here that needs solving. |
| 20:23.54 | putnopvut | Hrm. I'm not really qualified to make that sort of call since I don't really work on Asterisk much these days. Just hearing the change, I'd guess that it would be too invasive a change for 1.8 and if a less invasive fix could be made for 1.8, I'd recommend that. |
| 20:24.01 | putnopvut | Of course, band-aid fixes suck :( |
| 20:41.35 | *** join/#asterisk-bugs tzafrir_laptop (~tzafrir@bzq-218-155-148.cablep.bezeqint.net) |
| 20:43.11 | Marquis42 | Yes they do. :( |
| 20:43.37 | Marquis42 | I'll see if I can think up something clever for 1.8, or maybe run my cunning plan past russellb and see what his take is. |
| 20:58.27 | JunK-Y | just ordered "Asterisk: The Definitive Guide" |
| 20:58.40 | JunK-Y | good job russellb ! |
| 23:35.38 | *** join/#asterisk-bugs tzafrir (~tzafrir@212.179.75.202) |