| 00:04.18 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 00:27.14 | *** join/#asterisk-dev Ta^3 (~tacvbo@189.146.180.42) |
| 00:28.38 | *** join/#asterisk-dev moy (~moy@CPE001cdfec4cee-CM00080dab8485.cpe.net.cable.rogers.com) |
| 00:41.52 | moy | leifmadsen: ping |
| 01:15.36 | *** join/#asterisk-dev elguero (~miguel323@ns1.nashuacs.com) |
| 01:26.12 | Slugs_ | Does chan_h323 parse RDNIS? |
| 01:27.31 | Slugs_ | I know the RDNIS is being sent by the switch, it just does not look like it's being parsed |
| 01:36.31 | *** join/#asterisk-dev jtodd (r2del4oga1@asterisk/community-director-and-tie-dye-shirt-lover/jtodd) |
| 01:36.31 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 02:03.19 | Corydon76-dig | seanbright: ping ping ping ping ping ping ping |
| 02:03.40 | Corydon76-dig | seanbright: the kqueue(2) patch now works on Mac OS X |
| 02:08.45 | Corydon76-dig | Magic key: contrary to the documentation, an event added is NOT automatically enabled unless explicitly disabled... you have to enable it explicitly |
| 04:00.57 | *** join/#asterisk-dev jkroon (~jkroon@dsl-244-31-217.telkomadsl.co.za) |
| 04:08.42 | jkroon | http://pastebin.co.za/98388 |
| 04:09.05 | jkroon | got some kind of lockup. it's not related to locks, at least not locks in asterisk. |
| 04:35.28 | *** join/#asterisk-dev diegomad (mad@186.0.4.182) |
| 05:17.23 | Corydon76-dig | jkroon: those aren't lockups at all. You told it to dial out on DAHDI group 1 and dial 888, and wait forever for an answer |
| 05:17.58 | Corydon76-dig | If you meant for it to quit after a certain amount of time, you should have specified a timeout to Dial |
| 05:17.58 | jkroon | yea, but the call isn't going out. |
| 05:18.20 | jkroon | the expectation is for it to keep dialing forever and ever. |
| 05:18.27 | jkroon | did I post the dialplan there? |
| 05:18.31 | Corydon76-dig | As far as Asterisk is concerned, it's dialling out |
| 05:18.48 | Corydon76-dig | and waiting forever for an answer |
| 05:19.01 | jkroon | it's also receiving the call back (keep in mind span 1 is connected to 3, and 2 to 4) |
| 05:19.20 | jkroon | what happens if Dial() fails? ie, gets congestion or something like that? surely the Dial() should then also fail? |
| 05:20.01 | jkroon | the dialplan (monkeytest context) says to try dialing out over DAHDI, and if that fails, answer the call. |
| 05:20.03 | Corydon76-dig | That's correct, but currently, it's ringing through, and as long as nothing on the other end cancels the call, it will continue forever |
| 05:20.38 | jkroon | the box is isolated, all incoming calls goes into [monkeytest], span 1,2 is connected to 3,4. |
| 05:20.50 | Corydon76-dig | I understand that |
| 05:21.19 | jkroon | so it should either fail (as in congestion/failed/busy/something) or get answered almost instantaneously as far as I can tell. |
| 05:21.54 | Corydon76-dig | Without seeing what's in your incoming context, I cannot say |
| 05:22.21 | jkroon | with a single originate it'll run for HOURS, originating two, it'll go bye bye |
| 05:22.23 | jkroon | http://pastebin.co.za/98388 |
| 05:22.26 | Corydon76-dig | but it's stuck at 888@incoming:1 |
| 05:23.01 | Corydon76-dig | Again, what's in your [incoming] context? |
| 05:23.24 | jkroon | let me post that for you. |
| 05:24.35 | jkroon | http://pastebin.co.za/98409 |
| 05:25.35 | Corydon76-dig | Probably some sort of contention for the context lock |
| 05:25.55 | Corydon76-dig | Could probably be alleviated by switching to rwlocks |
| 05:26.24 | jkroon | ok, core show locks shows that no locks are held at the time asterisk stops |
| 05:27.03 | Corydon76-dig | Yes, but just because there's no locks doesn't mean there's no contention |
| 05:27.09 | Corydon76-dig | What version are you using? |
| 05:27.33 | jkroon | tried with 1.6.1.18 and 1.6.2.6 |
| 05:27.45 | jkroon | that was taken with 1.6.2.6 |
| 05:27.58 | Corydon76-dig | Oh, weird |
| 05:28.12 | jkroon | ok, what info should all go into a bug report? |
| 05:29.12 | jkroon | what should be attachments and what should be in line? |
| 05:29.27 | *** join/#asterisk-dev jtodd (q7jvj76r5g@asterisk/community-director-and-tie-dye-shirt-lover/jtodd) |
| 05:29.27 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 05:29.57 | Corydon76-dig | jkroon: no idea |
| 05:30.09 | Corydon76-dig | jtodd: are you in HSV or back out? |
| 05:30.20 | jkroon | ok, i'll try and make a sane call and just file it and hope for the best. |
| 05:31.02 | Corydon76-dig | Generally anything longer than about 30 lines should go as an attachment |
| 05:32.14 | jkroon | and i tend to write essays. |
| 05:35.09 | jkroon | hmm, asterisk or chan_dahdi ... i think I need to get someone to first test this with a different card, and also using IAX/2 and/or SIP. |
| 05:41.31 | jkroon | ah! If I limit the outbound channels to a group of two of the PRIs the usual indicative errors/warnings doesn't show up. so it's something more down the lines of "if we intend to send a call out on a channel and one happens to come in on that channel" we might deadlock ... that also explains why a single "monkey" won't kill it, because the channels is a multiple of 2, if there is one available to send out on it'll always come back |
| 05:41.32 | jkroon | in on another one. |
| 05:41.38 | jkroon | thus avoiding the problem. |
| 05:53.10 | *** join/#asterisk-dev pyite (~dschreibe@unaffiliated/pyite) |
| 05:57.55 | jkroon | now a more interesting question ... did the two MoH instances get connected to each other or did the system simply block? I suspect block seeing that there are two "Down" channels... |
| 06:55.18 | *** join/#asterisk-dev Netgeeks (~chris@173.11.68.155) |
| 10:36.57 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 10:36.57 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 10:38.11 | kpfleming | welcome to saturday |
| 10:38.25 | snuff-home | doing much today? |
| 10:38.30 | kpfleming | yes! |
| 10:38.58 | kpfleming | wife and i are headed to volunteer at a Habitat for Humanity house that needs interior demolition before they start renovating it |
| 10:39.15 | snuff-home | oooh.. demolition!!! |
| 10:40.43 | kpfleming | great way to relieve a week of work stress |
| 10:43.07 | snuff-home | sounds like a good relief valve :) |
| 12:30.18 | *** join/#asterisk-dev mpe (~mpe@0x4dd624b2.adsl.cybercity.dk) |
| 13:34.35 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 13:54.38 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 14:15.53 | Slugs_ | Does chan_h323 parse RDNIS? |
| 14:15.59 | Slugs_ | I know the RDNIS is being sent by the switch, it just does not look like it's being parsed |
| 14:55.55 | seanbright | Corydon76-dig: both tests now pass on openbsd |
| 15:16.44 | jtodd | Corydon-76-dig: Not in HSV at the moment, but will be there tomorrow around noon-ish. |
| 15:37.48 | *** join/#asterisk-dev Netgeeks (~chris@173.11.68.155) |
| 16:04.00 | *** join/#asterisk-dev dwayne (~dwayne@c-71-207-214-190.hsd1.al.comcast.net) |
| 16:05.23 | *** join/#asterisk-dev mpe (~mpe@0x4dd624b2.adsl.cybercity.dk) |
| 16:25.14 | leifmadsen | moy: pong! |
| 16:26.20 | moy | now that's what I'd call network delay |
| 16:27.10 | moy | just a quick question about the translations ... I am not familiar with conf-roll-call-complete, what is that for? what is a roll call? |
| 16:55.25 | *** join/#asterisk-dev diegomad (~mad@190.146.200.120) |
| 16:56.56 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 16:56.56 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 18:22.20 | Corydon76-dig | seanbright: sweet |
| 18:49.35 | *** join/#asterisk-dev jkroon (~jkroon@dsl-244-31-217.telkomadsl.co.za) |
| 18:50.19 | jkroon | Kobaz, got a cleaner lockup atm. still used two threads, but I think I've managed to confirm my theory about it being a race on the directions of ISDN (something about outbound calls). |
| 18:50.21 | jkroon | http://pastebin.co.za/99099 |
| 18:51.49 | jkroon | so basically span 1 is connected to span 2, which means that in the above seeing that DAHDI/2-1 is dialing (outbound line), DAHDI/33-1 should be in some kind of inbound context. however, according to that pastebin my understanding is that the channel has been "up" for quite some time. is my assessment at least correct? can anybody else also possibly look and confirm? |
| 19:40.56 | *** join/#asterisk-dev akshay07b (~b49531e1@gateway/web/freenode/x-bekbamopkngwdgsg) |
| 20:17.06 | *** join/#asterisk-dev atis_work (~atis_work@193.238.212.171) |
| 21:29.42 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 22:17.06 | *** join/#asterisk-dev moy (~moy@74.12.123.160) |
| 22:20.28 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |
| 22:45.46 | *** join/#asterisk-dev mpe (~mpe@0x4dd624b2.adsl.cybercity.dk) |
| 23:55.02 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |