| 00:03.18 | *** part/#asterisk-dev Deeewayne (n=dwayne@nat/digium/x-1ca33954aa476989) |
| 00:06.49 | jeremy_g | one party looses audio after i do a bridge () |
| 00:07.44 | jeremy_g | any known issues with bridge () app |
| 00:08.36 | *** join/#asterisk-dev shazaum (n=shazaum@unaffiliated/shazaum) |
| 00:09.14 | snuff-work | i can't say i've ever used it |
| 00:16.00 | lmadsen | does the bug tracker have any open issues? |
| 00:16.09 | lmadsen | that is really the best spot to answer your question |
| 00:18.37 | snuff-work | wonders if he should self pimp his patch again.. |
| 00:18.47 | snuff-work | btw lmadsen my email make sense? |
| 00:20.16 | lmadsen | I haven't ready it quite yet unfortunately |
| 00:21.10 | snuff-work | mm i think u mean 'read it' but all good :) |
| 00:24.41 | jeremy_g | lmadsen:how do i search the bugtracker for bridge ()? |
| 00:24.48 | jeremy_g | any urls |
| 00:24.57 | jeremy_g | damn this question is a bit stupid |
| 00:26.24 | jeremy_g | typing bridge shows lots of crap |
| 00:26.53 | jeremy_g | suggest a search string for bridge |
| 00:32.52 | lmadsen | try Bridge() ? |
| 00:33.23 | lmadsen | or just look through the bug topics |
| 00:51.56 | jeremy_g | is this for real, http://www.asterisk.org/forum/viewtopic.php?t=20704&highlight=&sid=43e528feb07041af23dbcc44f8c00349 |
| 00:52.03 | jeremy_g | a bug or mis-config |
| 00:53.29 | Corydon76-dig | jeremy_g: #asterisk |
| 01:19.07 | *** join/#asterisk-dev ZaVoid (n=zavoid@nj-76-6-39-193.dhcp.embarqhsd.net) |
| 01:22.33 | *** join/#asterisk-dev outtolunc (n=me@c-71-198-197-201.hsd1.ca.comcast.net) |
| 01:23.52 | *** join/#asterisk-dev ZaVoid (n=zavoid@nj-76-6-39-193.dhcp.embarqhsd.net) |
| 01:31.04 | *** join/#asterisk-dev jtodd (n=jtodd@blob.fox-den.com) |
| 01:33.54 | *** join/#asterisk-dev ZaVoid (n=zavoid@nj-76-6-39-193.dhcp.embarqhsd.net) |
| 01:49.45 | *** join/#asterisk-dev Mw3 (n=mw3@ip59934bd1.rubicom.hu) |
| 02:09.40 | *** join/#asterisk-dev Deeewayne (n=dwayne@c-76-29-245-9.hsd1.al.comcast.net) |
| 02:09.40 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 02:28.58 | *** join/#asterisk-dev jbot (i=ibot@rikers.org) |
| 02:28.58 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and Tier 3.14159265 support is in #asterisk -=- Leif Madsen is now the Open Source Release and Issue Tracker Manager. Blame all problems you find on him. That is all. -=- http://xkcd.com/303/ |
| 02:36.21 | *** join/#asterisk-dev shazaum (n=shazaum@unaffiliated/shazaum) |
| 02:43.24 | *** join/#asterisk-dev felipe_ (n=felipe@my.nada.kth.se) |
| 03:04.15 | *** join/#asterisk-dev Damin (n=damin@nucleus.nacs.net) |
| 03:07.41 | *** join/#asterisk-dev djbelieny (i=48303da5@gateway/web/ajax/mibbit.com/x-f0905d50b0bf98c5) |
| 03:07.59 | djbelieny | hello world ! |
| 03:08.12 | snuff-work | foo=bar |
| 03:13.34 | *** join/#asterisk-dev russellb_ (n=russell@asterisk/digium-open-source-team-lead/russellb) |
| 03:13.34 | *** mode/#asterisk-dev [+o russellb_] by ChanServ |
| 03:34.32 | djbelieny | hey Russell can I discuss some things I am doing with mix_monitor.c with you? I need some expert's advise :) |
| 03:34.47 | djbelieny | *app_mixmonitor.c |
| 03:34.54 | djbelieny | Asterisk 1.6 |
| 03:37.21 | russellb_ | to be honest, i am so exhausted right now, i probably don't have much to offer |
| 03:37.31 | russellb_ | and i have some school work i'm trying to finish before i finally pass out |
| 03:37.52 | russellb_ | you can try this channel again tomorrow during US business hours, it is very active then |
| 03:37.58 | russellb_ | or try the asterisk-dev mailing list |
| 03:38.04 | russellb_ | (assuming the question is about the code) |
| 03:39.37 | djbelieny | ok... no problem... been here or a few hours today and nothing... :D no problem at all... try again tomorrow.... (yep it is about code...) |
| 03:40.25 | djbelieny | have a good night ... |
| 03:41.51 | *** part/#asterisk-dev djbelieny (i=48303da5@gateway/web/ajax/mibbit.com/x-f0905d50b0bf98c5) |
| 03:43.44 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 04:18.38 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 04:23.08 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 04:31.06 | *** join/#asterisk-dev CunningPike (n=arodgers@S01060014bf81366b.vc.shawcable.net) |
| 04:56.07 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 04:59.21 | *** join/#asterisk-dev BBHoss (n=bbhoss@68.159.148.173) |
| 04:59.54 | *** part/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 05:59.42 | *** join/#asterisk-dev snuff-work (n=aaa@120.17.23.149) |
| 05:59.42 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 06:07.11 | *** join/#asterisk-dev fiddur (i=fiddur@c042.rit.se) |
| 06:11.35 | *** join/#asterisk-dev bkw_ (n=brian@freeswitch/developer/bkw) |
| 06:45.17 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 06:45.17 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:04.46 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:37.18 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 07:37.18 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:52.44 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 07:52.44 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:17.14 | *** join/#asterisk-dev double_cheesburg (n=chatzill@ip70-162-238-155.ph.ph.cox.net) |
| 08:21.18 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@local.xorcom.com) |
| 08:21.18 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 08:23.31 | *** join/#asterisk-dev tamiel (n=tamiel@213.30.183.226) |
| 08:35.13 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 08:35.13 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:44.17 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 08:44.17 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:58.20 | *** join/#asterisk-dev lbt (n=david@78.32.229.233) |
| 09:11.01 | *** join/#asterisk-dev _ys (n=ys@91.151.196.6) |
| 09:19.38 | oej | Good morning folks! |
| 09:55.35 | *** join/#asterisk-dev Dj-Neo (n=Jarrings@33.215-242-81.adsl-dyn.isp.belgacom.be) |
| 10:16.56 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 10:37.35 | wedhorn | mvanbaak: you heard anything about SPCP? |
| 10:42.36 | *** join/#asterisk-dev snuff-work (n=aaa@120.17.1.106) |
| 10:42.41 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 11:05.26 | oej | SPCP? That's new. |
| 11:13.43 | mvanbaak | new to me as well |
| 11:16.11 | snuff-work | ponders the world of sun cluster.. |
| 11:16.34 | Dj-Neo | Who can help me about dialplan ? |
| 11:17.15 | Dj-Neo | how it's possible to call *72 and provide all param directly like I call *72[EXT]#[FWd-To-Ext]# |
| 11:25.19 | oej | Dj-Neo Wrong channel - test asterisk-users |
| 11:25.30 | oej | Sorry mean #asterisk |
| 11:25.39 | Dj-Neo | ok ty |
| 11:29.15 | tzafrir_laptop | moy, here? |
| 11:30.38 | *** join/#asterisk-dev qdk (n=qdk@81.7.168.130) |
| 11:52.30 | *** join/#asterisk-dev shenhaoyu (n=shenhaoy@def92-8-82-236-12-218.fbx.proxad.net) |
| 12:02.15 | *** join/#asterisk-dev frawd (n=francois@88.Red-83-57-203.dynamicIP.rima-tde.net) |
| 12:02.24 | *** part/#asterisk-dev frawd (n=francois@88.Red-83-57-203.dynamicIP.rima-tde.net) |
| 12:25.49 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:41.45 | *** join/#asterisk-dev russellb_ (n=russell@asterisk/digium-open-source-team-lead/russellb) |
| 12:41.46 | *** mode/#asterisk-dev [+o russellb_] by ChanServ |
| 12:43.50 | oej | Morning, Russellb! |
| 12:45.36 | *** join/#asterisk-dev ZaVoid (n=zavoid@75.147.121.177) |
| 12:51.09 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 12:53.54 | russellb_ | g'morning oej ! |
| 12:54.50 | oej | Backporting is no fun. |
| 12:54.55 | russellb_ | heh |
| 12:55.04 | oej | I have most of astdb running on trunk now in the appleraisin branch |
| 12:55.10 | russellb_ | my guess would be that main/db.c isn't that different, though |
| 12:55.13 | oej | I got very good help from Corydon for that |
| 12:55.28 | oej | Not between 1.4 and trunk, no |
| 12:55.34 | oej | but realtime is. |
| 12:55.37 | russellb_ | ohhhh |
| 12:55.55 | oej | The required operations, store and delete, doesn't exist in 1.4, so there was no way we could have done this earlier. |
| 12:55.57 | russellb_ | those crazy asterisk developers ... always changing things ... |
| 12:56.04 | oej | Yes, isn't that bad? |
| 12:56.24 | russellb_ | so you're backporting store and delete, too? |
| 12:56.28 | oej | So I need to backport realtime stuff to get astdb/realtime running on 1.4 |
| 12:56.34 | russellb_ | big job, indeed |
| 12:56.54 | oej | Yes, I have to build failover for FreePBX. ANd it was either changing all of FreePBX or changing this. For me, this was the easier path... Weird. |
| 12:57.32 | russellb_ | understood |
| 12:57.38 | oej | ...and a bit more fun... |
| 12:57.42 | russellb_ | :-) |
| 12:58.15 | oej | However, after I tested this, I would like to get some of your and Corydon's eyes over the code as I don't have much experience of realtime development. |
| 12:58.34 | oej | I think it will be a nice addition to 1.6.x at some point :-) |
| 12:58.37 | russellb_ | of course, just let me (or someone) know when you're ready, and we can help you with posting it on reviewboard |
| 12:58.52 | oej | Cool. Gotta learn how to do that too... |
| 12:59.06 | russellb_ | There is a script "post-review" in repotools that helps |
| 12:59.18 | oej | Ok, will update that repo locally |
| 12:59.39 | russellb_ | In short, you apply your changes to a working copy of trunk, like you were about to commit it, and just run "post-review" |
| 12:59.51 | oej | Easy enough |
| 13:00.03 | russellb_ | that creates a review request as a draft (nobody can see it yet) |
| 13:00.13 | russellb_ | Then, you log in, and fill in the summary/description |
| 13:00.19 | russellb_ | and publish the request |
| 13:00.51 | snuff-work | mm.. |
| 13:01.08 | snuff-work | looks @ his ao2_jabber and its not spam-ness to the -dev list |
| 13:01.12 | russellb_ | Later, if you need to post updated versions of the patch, you get the new patch applied to a working copy of trunk, and run "post-review -r <reviewnumber>" |
| 13:01.48 | russellb_ | snuff-work: let me check something for you |
| 13:02.02 | snuff-work | k |
| 13:04.20 | russellb_ | should write all of that down somewhere ... |
| 13:04.57 | *** join/#asterisk-dev DarkRift (n=dark@65.92.170.52) |
| 13:06.36 | russellb_ | snuff-work: alright, i kicked it and i think my last update went to the list |
| 13:06.52 | oej | yes, please write that down :-) |
| 13:07.01 | russellb_ | snuff-work: http://lists.digium.com/pipermail/asterisk-dev/2009-February/036810.html |
| 13:07.03 | russellb_ | oej: will do |
| 13:07.18 | snuff-work | joy! |
| 13:07.45 | snuff-work | loves 3g internet dongles |
| 13:07.46 | russellb_ | snuff-work: all i did was reset the review group thing. You had it set to asterisk-dev already ... I set it to the same thing again and it worked |
| 13:07.48 | russellb_ | shrugs |
| 13:08.04 | russellb_ | stupid open source software ... |
| 13:08.06 | russellb_ | :-D |
| 13:08.15 | snuff-work | heh |
| 13:13.17 | oej | russellb: Want an issue to think about? A nice little problem... |
| 13:16.24 | russellb_ | prefers not thinking |
| 13:16.27 | russellb_ | but go ahead |
| 13:17.56 | snuff-work | mm i have another one.. |
| 13:17.59 | snuff-work | M14427 |
| 13:18.00 | MuffinMan | [new] [Asterisk] Channels/chan_iax2 0014427: [patch] Allow manager command to see if IAX link is trunk or not reported by snuffy (Karma: +65.25) http://bugs.digium.com/view.php?id=14427 |
| 13:18.02 | snuff-work | ;) |
| 13:18.06 | oej | This was not a bug tracker issue |
| 13:18.13 | oej | It is about my work |
| 13:18.24 | oej | astdb is part of the core and loads before we have any realtime engines |
| 13:18.34 | oej | So I can't initialize realtime when it loads |
| 13:18.40 | snuff-work | i figured.. i was just trying to highjack.. as my change tonight is cursed |
| 13:18.45 | oej | I have to do it at first operation |
| 13:19.09 | oej | But if a core module do anything with astdb, we're lost, since that will be stored in the old astdb that will be gone. |
| 13:19.23 | oej | So far, only modules like dundi, chan_iax and chan_sip seems to be using astdb, so we're fine. |
| 13:20.09 | oej | but it is a chicken-and-the-egg issue |
| 13:20.16 | russellb_ | sure is .. |
| 13:20.58 | russellb_ | As you say, I think we're safe for now, but it would get tricky if something in the core wanted to use it |
| 13:21.03 | oej | At some point, I need to look at the core and see if we can move the initialization of astdb to a later point and just deny operations |
| 13:21.05 | oej | Yes |
| 13:21.05 | russellb_ | That might be a good note to add to the header file |
| 13:21.25 | oej | Yep. Oh, do I, me, myself and I have to document stuff too? |
| 13:21.35 | russellb_ | you don't _have_ to :) |
| 13:21.50 | oej | well, I don't want lmadsen and jsmith chasing me |
| 13:21.57 | russellb_ | but it's a good observation, and it'd be a shame for it to be lost in time |
| 13:22.13 | oej | yes, like the explanation of REALT I M E |
| 13:22.18 | russellb_ | lol |
| 13:22.41 | tzanger | REALT I M E? |
| 13:22.56 | oej | yes, we totally forgot the explanation of that abbreviation |
| 13:22.59 | russellb_ | we were joking about what REALTIME meant yesterday |
| 13:23.13 | tzanger | ah |
| 13:23.14 | russellb_ | I don't think it was ever actually an abbreviation |
| 13:23.27 | oej | We do however agree on one thing - it's a lousy name |
| 13:23.32 | russellb_ | yes |
| 13:23.40 | russellb_ | but, if we can come up with a good abbreviation ... |
| 13:23.43 | russellb_ | it might make it better :) |
| 13:24.03 | tzanger | oh you don't mean realtime in terms of scheduling latency |
| 13:24.07 | tzanger | you mean things like asterisk relatime |
| 13:24.09 | tzanger | er realtime |
| 13:24.13 | russellb_ | tzanger: exactly the problem, heh |
| 13:24.16 | russellb_ | asterisk realtime |
| 13:24.21 | tzanger | it's not realtime... it's like dynamic or something |
| 13:24.25 | russellb_ | yeah .. |
| 13:24.26 | tzanger | asterisk dynamic peers |
| 13:25.11 | oej | It's not only peers |
| 13:25.24 | oej | It's a data object manipulation abstraction layer |
| 13:25.40 | oej | adomal! |
| 13:28.25 | oej | The Asterisk ADOMAL subsystem handles scalability and persistence |
| 13:28.47 | oej | Thanks to ADOMAL, I can operate my Asterisk on a Nokia 800 with 50.305 SIP peers |
| 13:33.39 | *** join/#asterisk-dev russellb_ (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 13:33.39 | *** mode/#asterisk-dev [+o russellb_] by ChanServ |
| 13:34.09 | *** join/#asterisk-dev shazaum (n=shazaum@unaffiliated/shazaum) |
| 13:42.44 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 13:42.44 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 13:44.33 | russellb | file: stop talking about kpfleming, he's right there |
| 13:44.42 | kpfleming | no, i'm over there ----> |
| 13:44.49 | file | you started it! |
| 13:44.56 | russellb | (>'')> |
| 13:44.59 | russellb | <("<) |
| 13:45.16 | russellb | <(")> |
| 13:45.28 | kpfleming | so... i found a complication in our translator mechanism yesterday |
| 13:45.56 | kpfleming | i know that surprises you |
| 13:46.00 | russellb | it does not |
| 13:46.13 | oej | is very surprised. But have no clue why. |
| 13:46.28 | kpfleming | lets say you have a translator that takes in SLIN and outputs, oh... G.279A |
| 13:46.52 | kpfleming | let's also say, that for some strange reason,there are standards that say you cannot put more than 80ms of G.729a into a single frame |
| 13:47.08 | kpfleming | since you are insane, and you want to conform to the standards, you try to do so |
| 13:47.20 | kpfleming | what happens when someone passes your translator a frame of SLIN that has 100ms of audio in it? |
| 13:47.36 | file | quite. |
| 13:47.43 | russellb | yes. |
| 13:48.09 | kpfleming | now... some very small parts of asterisk are prepared for a frame to have a 'next' frame (ast_write()) |
| 13:48.14 | kpfleming | but most of asterisk is not |
| 13:48.19 | kpfleming | so you can't return multiple frames |
| 13:48.34 | russellb | and you can't buffer it |
| 13:48.37 | kpfleming | nope |
| 13:48.47 | russellb | no easy fix. |
| 13:48.54 | kpfleming | this will also be a problem if we ever decide to implement G.729B |
| 13:49.16 | russellb | either you make it so you can return multiple frames, or we do the right thing and make it so translators can have more attributes than just input/output format bits |
| 13:49.28 | kpfleming | if the codec decides that a particular 20ms chunk of the SLIN is 'silence', and outputs a 2-byte SID frame, then you have to stop *right there*, you can't put any more audio into that frame |
| 13:49.45 | kpfleming | we may have to do *both* |
| 13:49.50 | russellb | i was about to say that |
| 13:49.57 | kpfleming | because the trigger for multiple frames is not just amount of audio |
| 13:49.59 | russellb | and even then, you'll end up needing multiple frame support probably |
| 13:50.04 | kpfleming | yep |
| 13:50.25 | russellb | cries at the thought of what that will do to frame timestamps |
| 13:50.34 | kpfleming | actually, it's OK |
| 13:50.45 | kpfleming | the incoming frame represented 100ms of time, the combined output frames would too |
| 13:51.05 | russellb | yeah, but that means codec_whatever has to account for changing the timestamps |
| 13:51.14 | russellb | whereas i don't think they touch them at all right now |
| 13:51.15 | kpfleming | yes |
| 13:51.19 | russellb | just a bit more complexity |
| 13:51.22 | kpfleming | right, translate.c does it |
| 13:51.28 | russellb | just copies it, right? |
| 13:51.48 | kpfleming | sort of.. i'm not really clear on how frame timestamps are handled right now, to be honest |
| 13:51.48 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:51.48 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 13:51.57 | kpfleming | it's not an area i've ever spent any time looking at |
| 13:51.58 | russellb | isn't totally clear either |
| 13:52.22 | lmadsen | is totally hazy |
| 13:52.29 | russellb | i think they're mostly just ... passed through |
| 13:52.46 | kpfleming | well, smoothers mess with them already |
| 13:52.51 | russellb | indeed |
| 13:52.52 | kpfleming | they have no choice |
| 13:53.30 | kpfleming | as i said... complication |
| 13:53.34 | russellb | quite |
| 13:53.54 | kpfleming | we've gotten away with it because everyone is a 'good citizen' and never passes us such huge frames |
| 13:53.55 | russellb | it's almost like we have some media architecture work to do |
| 13:54.00 | kpfleming | almost, yeah |
| 13:54.05 | file | crazy talk |
| 13:54.18 | kpfleming | so, that's my thought for this morning |
| 13:54.23 | russellb | very nice. |
| 13:55.04 | kpfleming | on a good note, i've pulled the open-coded smoother out of codec_g729a in favor of using one that already exists in asterisk :-) |
| 13:55.10 | russellb | yay |
| 13:55.22 | kpfleming | so that will be the... 2nd place that uses a smoother |
| 13:55.31 | oej | timestamps is essential for video |
| 13:55.41 | russellb | video scmideo |
| 13:55.42 | oej | And we don't handle it properly today, according to rumours. |
| 13:55.42 | kpfleming | they are for audio as well, really |
| 13:55.53 | oej | But I guess you can get away with more on audio. |
| 13:56.02 | russellb | yes, they're important for audio for jitterbuffers to work properly |
| 13:56.02 | kpfleming | it screws up jitter buffers on lossy networks |
| 13:56.09 | oej | But sending video over an IAX2 trunk destroys timestamps |
| 13:56.16 | kpfleming | audio as well |
| 13:56.16 | russellb | ah ... |
| 13:56.21 | kpfleming | IAX2 does not preserve timestamps |
| 13:56.22 | russellb | iax2 invents its own timestamps i think |
| 13:56.26 | russellb | jinx. |
| 13:56.32 | kpfleming | kpfleming and russellb share a brain |
| 13:56.36 | russellb | O.O |
| 13:56.52 | oej | Neural networking? |
| 13:57.02 | russellb | and we're like 30 miles apart |
| 13:57.03 | russellb | long cables |
| 13:59.05 | oej | Bluetooth fixes everything - long range bluetoothh (LRB) |
| 13:59.15 | russellb | prefers greentooth |
| 14:01.03 | kpfleming | prefers toothpaste and SoniCare |
| 14:01.27 | file | I see alive people. |
| 14:01.46 | russellb | .fortune |
| 14:01.50 | russellb | :( |
| 14:03.19 | russellb | ~tackle file |
| 14:03.19 | jbot | ACTION tackles file to the ground. |
| 14:03.32 | file | implodes |
| 14:03.59 | codefreeze-lap | ~fortune |
| 14:03.59 | jbot | i guess fortune is a kickass app available for Linux. Packages are in OZ and http://frob.us/modules.php?name=Fortune lets you use fortune over the web. |
| 14:08.56 | oej | ~appleraisin |
| 14:09.09 | oej | Ha! The bot that doesn't know a thing. |
| 14:10.15 | kpfleming | i am not thrilled when i log in for the first time in the morning and there are four skype chat rooms with content awaiting my reading pleasure |
| 14:19.52 | jsmith | oej: It's simple: watch this! |
| 14:20.14 | jsmith | jbot: appleraisin is an Asterisk project to allow you to replace the AstDB database with a realtime engine |
| 14:20.15 | jbot | okay, jsmith |
| 14:20.24 | jsmith | ~appleraisin |
| 14:20.24 | jbot | methinks appleraisin is an Asterisk project to allow you to replace the AstDB database with a realtime engine |
| 14:21.00 | snuff-work | ~killtheuser |
| 14:21.41 | jsmith | jbot: killtheuser is an Asterisk project to change the internals of the chan_sip channel driver, not to do any harm to our wonderful Asterisk users |
| 14:21.42 | jbot | okay, jsmith |
| 14:22.17 | oej | Oh, he's in learning mode |
| 14:22.43 | oej | ~metermaids |
| 14:23.12 | oej | jbot: metermaids is the entity that watches the state of parkinglots. Feel free to call them Lisa, lovely Lisa metermaid. |
| 14:23.13 | jbot | oej: okay |
| 14:25.12 | kpfleming | jbot: oej is a large blonde Swedish man who likes to drink tea. |
| 14:25.13 | jbot | okay, kpfleming |
| 14:25.36 | oej | Now, this is essential knowledge! |
| 14:25.48 | oej | sips a nice cup of tea. Where's the cookie? |
| 14:26.31 | oej | replaces SENTINEL with NULL since mvanbaak isn't watching |
| 14:36.10 | *** join/#asterisk-dev stevie_ramjet (n=putnopvu@64.122.31.62) |
| 14:36.10 | *** mode/#asterisk-dev [+o stevie_ramjet] by ChanServ |
| 14:40.03 | kpfleming | stevie_ramjet: ! |
| 14:40.31 | stevie_ramjet | kpfleming, what's happenin'? |
| 14:40.46 | kpfleming | we are not skiing |
| 14:41.31 | stevie_ramjet | ah, I'm not either...yet |
| 14:41.44 | stevie_ramjet | but that should be different in about two hours. |
| 14:43.29 | lmadsen | stevie_ramjet: !!! |
| 14:44.26 | stevie_ramjet | lmadsen, hey there. |
| 14:44.39 | lmadsen | Do any USA'ians know where your President is? |
| 14:45.08 | stevie_ramjet | I don't. |
| 14:45.17 | stevie_ramjet | But I certainly don't account for everyone. |
| 14:45.19 | lmadsen | tisk tisk |
| 14:46.22 | lmadsen | it's his first foreign trip! |
| 14:48.03 | kpfleming | montana? |
| 14:50.36 | moy | tzafrir_laptop: here now, if that helps, heh |
| 14:58.42 | *** join/#asterisk-dev frawd (n=francois@187.pool85-54-33.dynamic.orange.es) |
| 14:58.50 | *** part/#asterisk-dev frawd (n=francois@187.pool85-54-33.dynamic.orange.es) |
| 15:01.39 | tzafrir_laptop | oej, how well is sqlite's performance as an astdb replacement? |
| 15:02.07 | tzafrir_laptop | (assuming it will be the default, as it is the only one that needs no extra setup) |
| 15:02.13 | oej | I have no idea |
| 15:02.18 | oej | Using mysql :-) |
| 15:02.21 | oej | Please test |
| 15:04.07 | tzafrir_laptop | sqlite3 generally has a known issue in Linux for requiring a sync() after each transaction |
| 15:04.28 | tzafrir_laptop | Or else it can't guarantee database consistency |
| 15:06.05 | oej | That would be up to the sqlite realtime engine |
| 15:06.56 | oej | My main motivation was the ability to move astdb out of the box for failover. |
| 15:07.04 | oej | Sqlite is as bad as db3 there ... |
| 15:07.30 | *** join/#asterisk-dev ctooley (n=ctooley@mo-71-0-221-206.dhcp.embarqhsd.net) |
| 15:07.30 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 15:10.41 | tzafrir_laptop | oej, db1, actually |
| 15:10.42 | *** join/#asterisk-dev Mog (n=mog@c-68-62-170-242.hsd1.al.comcast.net) |
| 15:10.43 | *** mode/#asterisk-dev [+o Mog] by ChanServ |
| 15:11.05 | oej | yes, db1. Right. Sorry. |
| 15:12.09 | tzafrir_laptop | Well, the theory is that with multiple writers sqlite3 will not be as good as db1. But then again, I haven't tested it yet |
| 15:19.47 | *** join/#asterisk-dev Deeewayne (n=dwayne@nat/digium/x-a7816f5ef2f0333c) |
| 15:19.47 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 15:50.13 | oej | Hmm. My res_mysql doesn't talk to my asterisk properly. The parameters are all gone. |
| 15:56.27 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 15:59.46 | *** join/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 15:59.46 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 16:14.40 | oej | What would cause a .so module not to get any parameters properly? |
| 16:15.54 | russellb | not to get parameters? I'm not sure what you mean |
| 16:16.34 | oej | I don't get easy stuff like database name, it's just -10.2123123123 or something |
| 16:16.56 | oej | And I've checked that everything is declared the requested way. |
| 16:17.42 | oej | The mysql engine registers fine and connects to the database |
| 16:17.56 | oej | But I can't send any queries to it, all functions get bad parameter values |
| 16:19.59 | *** join/#asterisk-dev schmooze (n=schmooze@h69-129-142-83.mdsnwi.tisp.static.tds.net) |
| 16:22.11 | oej | Weird... |
| 16:24.44 | russellb | sounds like version skew |
| 16:24.52 | russellb | like, it was built against headers from a different asterisk version |
| 16:25.06 | oej | <PROTECTED> |
| 16:25.19 | oej | Both should be 1.4 |
| 16:25.23 | oej | I will need to check again |
| 16:27.02 | oej | branches/1.4 of addons is for asterisk/branches/1.4 -right? |
| 16:27.21 | file | yes. |
| 16:27.44 | oej | ...removing all include files... |
| 16:28.27 | oej | Still doesn't work. |
| 16:29.51 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 16:30.24 | oej | will have to test with a "clean" res_config_mysql. I propably caused this damage all by myself |
| 16:35.19 | oej | A plain res_config_mysql shows the same problem... I need food to handle this! |
| 16:41.38 | *** join/#asterisk-dev Mr_BOnd_007 (i=Mr_BOnD_@119.160.199.6) |
| 16:42.11 | *** join/#asterisk-dev frawd (n=francois@88.Red-83-57-203.dynamicIP.rima-tde.net) |
| 16:47.18 | *** part/#asterisk-dev frawd (n=francois@88.Red-83-57-203.dynamicIP.rima-tde.net) |
| 16:54.23 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 16:54.24 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 16:56.48 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.10.119) |
| 17:04.43 | *** part/#asterisk-dev Mog (n=mog@c-68-62-170-242.hsd1.al.comcast.net) |
| 17:05.44 | *** join/#asterisk-dev mog (n=mog@nat/digium/x-4cf030e6097eec75) |
| 17:05.44 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 17:06.30 | mnicholson | ok, so I want to have one thread running ast_io_wait() and handling io requests for multiple bluetooth devices, but I need different threads adding and removing devices from that io_context. The idea I came up with was to use two locked queues to signal the monitoring thread what to add and remove from the io_context. When ever another thread wanted to add/remove an fd to watch, it would add an entry into the add/remove queue. |
| 17:06.31 | mnicholson | <PROTECTED> |
| 17:07.17 | russellb | Tricky .. |
| 17:08.25 | file | ah |
| 17:08.44 | mnicholson | this is partly necessary because for each bluetooth adapter, I have to listen() for audio connections on a particular socket, and then service those connections when they come in |
| 17:09.12 | mnicholson | but the listening thread does not know when to stop listening in all cases |
| 17:09.23 | mnicholson | well when to stop servicing audio data |
| 17:14.30 | russellb | yeah, i think the add/remove queue idea works |
| 17:14.42 | russellb | there are some other things you could do, but it requires adding locking internally to the ast_io_context stuff |
| 17:14.48 | russellb | also, I don't think astobj2 is necessary here |
| 17:14.56 | russellb | just a simple queue as a list would work fine |
| 17:15.34 | file | removal is what slightly concerns me, since when the call to remove it returns it isn't actually removed |
| 17:16.06 | russellb | good point, definitely something to consider |
| 17:16.22 | russellb | you could implement such that it's a block request |
| 17:16.47 | russellb | add to queue, send a signal to wake up the thread, and sleep on a condition until it tells you that it has been removed |
| 17:17.09 | russellb | yet another case where ast_poll_and_cond_wait() would be awesome :) |
| 17:17.48 | kpfleming | write it now! do it! do it! |
| 17:17.53 | russellb | heh |
| 17:18.01 | kpfleming | but i would avoid using signals |
| 17:18.03 | russellb | punts to glibc |
| 17:18.09 | kpfleming | signals are a pain in the arse, use a pipe instead |
| 17:18.14 | russellb | ahh, yes, that works |
| 17:18.21 | russellb | add an alert pipe to the io context :) |
| 17:18.23 | file | there was talk about how to implement it on one of the lists I look at long ago |
| 17:18.28 | kpfleming | yes |
| 17:18.33 | kpfleming | it would have to be done on the kernel side, then in glibc |
| 17:18.34 | russellb | it's coming one of these days |
| 17:18.40 | file | kpfleming: right |
| 17:18.50 | russellb | but i'm not the guy to write it. |
| 17:19.36 | russellb | and then it won't be supported on non-Linux |
| 17:19.38 | russellb | and we will be sad |
| 17:20.36 | mnicholson | hmm, im not sure if there are any cases where it matters if something is removed right then or not |
| 17:20.36 | russellb | maybe it'll be in POSIX2017 |
| 17:20.58 | russellb | mnicholson: as long as it doesn't mean you have a race condition where that thread is going to end up looking at memory that has been free'd |
| 17:21.36 | mnicholson | i need to figure out a way to get the id for stuff that I add |
| 17:22.05 | mnicholson | actually, adding will always be done from the monitoring thread, so I only need a queue for removal |
| 17:22.44 | russellb | assuming asynchronous removals are necessary |
| 17:23.18 | mnicholson | when the channel is hung up, a removal is needed |
| 17:23.37 | mnicholson | unless the io_context automatically removes invalid fds |
| 17:23.44 | russellb | it will not |
| 17:23.55 | russellb | but poll() will start returning errors |
| 17:24.03 | russellb | if you close an fd that is polling on |
| 17:24.38 | mnicholson | hmm |
| 17:37.09 | mnicholson | so I could do this just by calling close() on the fd from the place that needs to close the socket and then checking for errors in my call back function and returning 0 if I have a read error |
| 17:37.26 | kpfleming | yes i use this technique in other places |
| 17:42.04 | mnicholson | much simpler |
| 17:44.43 | oej | your SIP auth secret is null-X-%-<ctrl-g>-<CR> |
| 17:49.21 | oej | I need five reasons why I can't call a function in a standard 1.4 addons res_config_mysql... |
| 17:49.33 | oej | Hmm. Might try with a standard asterisk 1.4 just for fun. |
| 17:49.50 | oej | Thanks for the good advice! |
| 17:55.05 | kpfleming | are you talking to yourself? |
| 17:55.43 | oej | Well, if no one else bothers to answer I might give myself some pretty good advice if I have the time |
| 17:56.00 | oej | But YOU on the other hand might be able to give me a clue on what's going on, kpfleming |
| 17:56.09 | kpfleming | maybe so |
| 17:56.16 | kpfleming | what is the issue? |
| 17:56.19 | oej | I have asterisk-1.4 with mods and a plain res_config_mysql.so from addons 1.4 |
| 17:56.29 | kpfleming | ok |
| 17:56.41 | oej | When asterisk calls res_config_mysql it works in functions without any parameters |
| 17:56.49 | kpfleming | ok |
| 17:56.55 | oej | But when running a realtime op, the database name and table is all gone. |
| 17:57.05 | mnicholson | hmm, is it ok to call ast_io_add() from within and ast_io_callback? |
| 17:57.27 | kpfleming | oej: what function specifically are you referring to? |
| 17:57.39 | oej | realtime_mysql is one example |
| 17:59.10 | kpfleming | and what realtime API call are you using that ends up calling realtime_mysql? |
| 18:00.23 | oej | Here's one example. I run ast_load_realtime() |
| 18:00.36 | oej | In a "database get" operation (astdb) |
| 18:00.44 | kpfleming | ok |
| 18:00.47 | mnicholson | looks like it's not safe |
| 18:00.59 | oej | it runs realtime_func |
| 18:01.44 | kpfleming | and then the database and table parameters are NULL, or they point to empty strings, or are garbage? |
| 18:01.47 | oej | realtime_func is mapped to realtime_mysql |
| 18:01.51 | oej | garbage |
| 18:02.08 | kpfleming | ok |
| 18:02.16 | kpfleming | i suspect this is a compiler optimization issue |
| 18:02.28 | kpfleming | one side is passing arguments in registers, the other side is expecting them on the stack |
| 18:02.34 | Corydon76-dig | oej: what bug is your patch posted under? |
| 18:02.46 | oej | This is no bug report at all |
| 18:02.52 | kpfleming | you need to ensure that the compiler flags for building the .o/.so files are as close as possible between res_config_mysql and the asterisk stuff |
| 18:03.16 | oej | Didn't you say that addons took all the flags from the main compilation? |
| 18:03.58 | kpfleming | it is supposed to |
| 18:04.02 | oej | :-) |
| 18:04.03 | kpfleming | but that does not mean it works |
| 18:04.10 | kpfleming | have to head out to lunch, bbiab |
| 18:04.19 | oej | ok |
| 18:04.29 | oej | Hmm. How do I check these flags that are suppressed? |
| 18:04.45 | Corydon76-dig | NOISY_BUILD=y make |
| 18:07.22 | Corydon76-dig | oej: my community backports for 1.4 use a special Makefile that correctly match build flags |
| 18:07.36 | oej | http://pastebin.ca/1342118 |
| 18:07.48 | oej | Yes, these doesn't seem similar |
| 18:08.51 | Corydon76-dig | the FORTIFY_SOURCE is probably making the difference |
| 18:09.39 | Corydon76-dig | maybe also the -march |
| 18:10.02 | Corydon76-dig | Pretty much everything else is just build warnings |
| 18:13.08 | oej | No |
| 18:13.19 | oej | Where's that makefile? |
| 18:17.28 | oej | I might want access to svncommunity for this backport of realtime stuff |
| 18:19.24 | oej | Found it |
| 18:22.58 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 18:25.00 | djbelieny | hello all.... I need some help on some mods i am doing with app_mixmonitor.c, anyone up for it ? |
| 18:25.31 | *** part/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 18:26.24 | *** join/#asterisk-dev shido6 (n=shido6@96-28-34-156.dhcp.insightbb.com) |
| 18:26.48 | *** join/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 18:27.55 | djbelieny | I am trying to change app_mixmonitor.c to have the option to play a tone to either or both parties of a monitored call after X milliseconds. |
| 18:28.29 | djbelieny | I am doing this since I cannot find other way to do it from an automixmonitor session |
| 18:28.44 | oej | hmm the viewvc on svncommunity doesn't have a "download" link |
| 18:30.12 | djbelieny | but when I play a beep using ast_streamfile from within a mixmon thread it actually sounds like a crazy duck. |
| 18:30.32 | djbelieny | anyone up for giving me some pointers ??? Noob here :) |
| 18:55.59 | *** join/#asterisk-dev ctooley (n=ctooley@mo-71-0-221-206.dhcp.embarqhsd.net) |
| 18:55.59 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 18:57.59 | Corydon76-dig | djbelieny: you can't just add extra frames to a channel. That won't work. |
| 18:58.27 | Corydon76-dig | djbelieny: You'd have to either mix them into the existing queued frames or replace the queued frames entirely |
| 18:59.19 | Corydon76-dig | djbelieny: but you'd need to do that within the bridge, really, so you'd need to signal the bridge to play a particular file, much like the L() option to Dial |
| 19:04.54 | djbelieny | ok.. I'll read the code for the dial and come back here... this is great info thank you |
| 19:07.13 | oej | russellb: Around? |
| 19:08.34 | Corydon76-dig | No, asquare |
| 19:08.55 | djbelieny | another thing... how would I make a wave into frames to mix them with the other frames |
| 19:08.58 | file | oej: there are quite a few people lunching right now, he will return momentarily |
| 19:10.07 | djbelieny | actually read the frames from the wav and mix them ? |
| 19:11.24 | file | djbelieny: read my last part of http://lists.digium.com/pipermail/asterisk-dev/2009-February/036777.html |
| 19:11.40 | djbelieny | Thank you. |
| 19:12.20 | file | it has the two ways I thought of for mixing audio into a channel as such |
| 19:14.02 | djbelieny | Reading it now :D |
| 19:15.51 | *** part/#asterisk-dev snuff-work (n=aaa@120.17.1.106) |
| 19:19.13 | oej | file: Lunching? At this time? |
| 19:19.16 | oej | :-) |
| 19:19.23 | file | I should lunch |
| 19:19.41 | oej | Corydon76-d: Your makefile caused very few compiler options to activate |
| 19:23.48 | Corydon76-dig | oej: did it work, though? |
| 19:28.42 | djbelieny | Hey lunch is always good :) |
| 19:35.09 | oej | No, corydon. There's some issue with the mysql linking |
| 19:35.21 | oej | The problem is that your makefile doesn't really work with addons |
| 19:35.31 | oej | I need the makeopts from addons and include for asterisk |
| 19:35.35 | kpfleming | ok, i'm back, you have everything solved, right? |
| 19:35.36 | oej | so I added ASTINC |
| 19:36.01 | oej | Trying Tilghman's Makefile from svncommunity... |
| 19:36.28 | *** join/#asterisk-dev codefreeze-lap (n=murf@72.21.67.40) |
| 19:36.28 | *** mode/#asterisk-dev [+o codefreeze-lap] by ChanServ |
| 19:36.48 | oej | Loading the module, Asterisk complains that it can't find mysql_encode something |
| 19:36.58 | oej | So there's an issue with the linking process in that Makefile |
| 19:37.00 | kpfleming | yeah, that won't link in mysql libraries, i suspect |
| 19:37.03 | kpfleming | right |
| 19:38.14 | oej | kpfleming: http://pastebin.ca/1342118 |
| 19:38.23 | oej | Shows the difference between compile in addons and asterisk 1.4 |
| 19:39.13 | oej | Tilghman suggested I add march and FORTIFY_SOURCE but that did not solve the issue |
| 19:41.51 | oej | <PROTECTED> |
| 19:47.29 | wedhorn | mvanbaak, oej: spcp seems to be the protocol used by the new cisco 5xx phones (like the cp-524g) |
| 19:47.42 | wedhorn | only info I've seem to found is https://www.myciscocommunity.com/docs/DOC-2597;jsessionid=7951E104236CF297541FF7BE069E712B |
| 19:47.43 | oej | Aha |
| 19:47.44 | Corydon76-dig | oej: you could just use the ODBC library |
| 19:47.51 | oej | :-) |
| 19:48.27 | wedhorn | spcp seems to be sccp with some sort of token exchange before registration |
| 19:48.49 | Corydon76-dig | Or bludgeon Monty into giving Asterisk an exception to the MySQL GPL license |
| 19:49.07 | wedhorn | the new phones look like they'll be cheap, us$130 for a 4 line |
| 19:49.08 | Corydon76-dig | Didn't you say he was Swedish? |
| 19:49.49 | oej | Working on it, Corydon. Monty was finnish btw. And he has left SUN... |
| 19:51.38 | Corydon76-dig | Ah, Finnish, Swedish. Y'all look alike to me |
| 19:51.50 | Corydon76-dig | ;-) |
| 19:59.24 | oej | Ah, Russians, Americans. Y'all look alike to us |
| 19:59.37 | oej | ;-) |
| 19:59.47 | seanbright | we smell better |
| 20:00.10 | oej | Ahh. That's the difference. THanks. |
| 20:00.25 | Corydon76-dig | oej: that's because some Russians are Americans... |
| 20:00.26 | seanbright | you're welcome :) |
| 20:00.48 | Corydon76-dig | and Finns and Swedes and... pretty much every nationality... |
| 20:01.11 | Corydon76-dig | America... Mutt Pride! |
| 20:03.28 | Corydon76-dig | bbiab... need to go to the repair shop and pick up my body disposal unit |
| 20:03.37 | kpfleming | ok, i'm off the phone now |
| 20:05.44 | kpfleming | so i would edit the addons Makefile (temporarily) to remove -fPIC |
| 20:05.54 | kpfleming | and add -pthread |
| 20:06.09 | kpfleming | you said you already tried add -march=i686? |
| 20:06.16 | kpfleming | oej: ^^^ |
| 20:07.45 | *** part/#asterisk-dev djbelieny (n=djbelien@72-48-61-165.dyn.grandenetworks.net) |
| 20:08.02 | oej | Back |
| 20:08.09 | russellb | Forward |
| 20:08.15 | oej | Sideways... |
| 20:08.24 | seanbright | Up-Up-Down-Down-Left-Right-Left-Right-B-A-Start |
| 20:11.44 | oej | kpfleming: No, still no parameter passing |
| 20:11.52 | oej | (Plain res_config_mysql) |
| 20:12.31 | oej | The module loads ok and connects to mysql |
| 20:13.23 | oej | http://pastebin.ca/1342211 |
| 20:13.38 | oej | Still using fPIC.. Weird. |
| 20:13.40 | oej | Let me check |
| 20:13.50 | denon | seanbright: wow, I haven't thought about that key sequence in 15 years |
| 20:13.53 | denon | or .. 20 |
| 20:13.57 | denon | or something |
| 20:14.05 | denon | dists off his old NES |
| 20:14.07 | denon | er dusts |
| 20:14.17 | oej | the -fPIC must come from the asterisk directory |
| 20:14.18 | seanbright | contra <3 |
| 20:25.36 | *** join/#asterisk-dev korihor (n=korihor@200.44.218.45) |
| 20:32.49 | oej | kpfleming: Getting parameters, but there's a difference in the va-lists somewhere |
| 20:39.31 | kpfleming | ok, so you're getting closer |
| 20:39.39 | oej | Yes, thanks for the help |
| 20:39.57 | kpfleming | is the version of asterisk you are compiling also the one that is installed (are its headers in /usr/include/asterisk)? |
| 20:40.15 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-179-75-202.static.bezeqint.net) |
| 20:40.15 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 20:40.39 | oej | yes |
| 20:40.55 | kpfleming | does res_config_mysql.c include asterisk.h as the very first header? |
| 20:43.01 | oej | yes |
| 20:43.55 | Corydon76-dig | Do you have more than one asterisk.h installed in your system headers? |
| 20:44.23 | oej | I removed all of them before install |
| 20:44.33 | oej | Or how do you mean? |
| 20:44.43 | oej | "system headers"? |
| 20:44.45 | Corydon76-dig | like /usr/local/include/asterisk.h |
| 20:44.49 | *** join/#asterisk-dev ScribbleJ (n=sj@c-67-172-6-141.hsd1.il.comcast.net) |
| 20:44.51 | oej | Aha |
| 20:46.00 | oej | Only one asterisk.h installed |
| 20:48.01 | Corydon76-dig | When you used my Makefile from community, did you define ASTSRC before running Make ? |
| 20:48.04 | kpfleming | so is it only your code that is having this trouble calling res_config_mysql? can you use realtime from chan_sip or something else OK? |
| 20:48.28 | oej | Well, the trouble is gone now thanks to your fixes. Now it's my code |
| 20:48.45 | Corydon76-dig | woot |
| 20:48.45 | kpfleming | yay! |
| 20:48.46 | oej | I did not test chan_sip or anything else though. |
| 20:55.06 | kpfleming | https://bugs.digium.com is alive, using a 30-day FreeSSL cert for testing |
| 20:55.14 | kpfleming | Qwell: ^^^ |
| 20:55.33 | Qwell | I will comment on that bug. |
| 20:57.40 | Corydon76-dig | starts a database transaction on oej |
| 20:58.00 | oej | waits for commit |
| 20:58.33 | Corydon76-dig | rollsback oej |
| 20:58.46 | oej | yes, I'm back. Parameters lost. |
| 20:58.52 | *** join/#asterisk-dev obnauticus (n=obnautic@about/windows/regular/obnauticus) |
| 21:01.43 | Damin | Hmm... |
| 21:02.32 | Damin | If you have nat=yes and externip=x.x.x.x, shouldn't Asterisk set all source packets as coming from x.x.x.x? |
| 21:02.56 | Damin | Or does that just munge the SIP via headers and RTP endpoints? |
| 21:03.32 | file | nat=yes is not used for when Asterisk is behind NAT, externip only modifies the messages |
| 21:03.33 | kpfleming | asterisk can't change the source IP address to an addres that doesn't live on the box |
| 21:04.16 | Damin | kpfleming: Here is the rub then.. |
| 21:04.40 | oej | There is a problem when Asterisk has multiple IP addresses on the same box |
| 21:04.50 | Damin | kpfleming: I'm worning on a failover mechanism, similar to what flip_1405 does, which basically is an active/passive failover mechanism.. |
| 21:05.17 | kpfleming | i have no clue what flip_1405 is or does |
| 21:05.19 | Damin | kpfleming: It implements a floating IP that moves between two servers.. |
| 21:06.06 | Damin | kpfleming: If the Slave deteces the Master is down, and can't nmap Asterisk's 5060 port on that box, then it takes over the IP, gratuitous arps for it, then starts Asterisk on the slave server. |
| 21:06.23 | kpfleming | ok |
| 21:06.27 | kpfleming | sounds like most HA solutions |
| 21:06.49 | *** join/#asterisk-dev telecos (n=sergio@87.219.167.0) |
| 21:07.09 | Damin | kpfleming: So how can I get Asterisk to send source packets from the eth0:0 address, rather than the eth0 IP.. |
| 21:07.40 | Damin | kpfleming: Some sort of combination of localnets maybe? Or perhaps I ought to move the Master/Slave IP addresses out of the same subnet.. |
| 21:08.06 | Damin | kpfleming: Right now, 207.166.192.52 is the Master IP, 207.166.192.53 is the Slave IP and 207.166.192.51 is the IP that floats between the two boxes.. |
| 21:08.28 | oej | And bindaddr is... |
| 21:08.43 | Damin | If I move the Master and Slaves to a different subnet, say a private one, I wonder if that would work.. |
| 21:08.58 | Damin | oej: I set bindaddrr=207.166.192.51 |
| 21:09.24 | oej | But that IP might not be available when asterisk starts? |
| 21:09.29 | Damin | oej: But when I do that, I can no longer test against 127.0.0.1 to see if the local Asterisk is responding or not.. |
| 21:09.40 | Damin | oej: It should always be available when Asterisk starts.. |
| 21:09.43 | kpfleming | that is correct |
| 21:09.48 | kpfleming | so you have found one of the known problems in chan_sip |
| 21:10.04 | kpfleming | it can't bind to multiple IPs, and if you don't do that, you can't control the source address |
| 21:10.19 | kpfleming | so you are in a catch-22 |
| 21:10.57 | Damin | Hmm. soo binaddr=x.x.x.x,y.y.y.y is not valid then either.. |
| 21:10.58 | kpfleming | one trick some people use is to bind to the floating IP, then put in an iptables rule (or similar) to map a different port number on the primary IP over to the secondary IP's port 5060 |
| 21:11.03 | kpfleming | not in chan_sip |
| 21:11.09 | file | kpfleming: I was just about to say that |
| 21:11.11 | kpfleming | chan_iax2 can bind to multiple IPs, but not chan_sip |
| 21:12.10 | Damin | I wonder what would happen if I moved the Master/Slave addresses over to a 192.168.x.x subnet.. |
| 21:13.09 | Damin | I suppose I could cross connect eth1 between the two boxes, put them on a private subnet... |
| 21:13.24 | Damin | Crap.. Somehow, I managed to get this working once before.. |
| 21:17.14 | Damin | If I use binaddr=207.166.192.51 it works, but now I have to figure out how to work around the failover issue.. |
| 21:17.16 | Damin | udp 0 0 207.166.192.51:5060 0.0.0.0:* |
| 21:18.00 | Damin | I think I'm going to use an arp lookup to see what MAC address responds to the floating IP. |
| 21:18.10 | Damin | If the MAC is local, then we don't have anything to do.. |
| 21:20.03 | kpfleming | or just nmap some other port... chan_iax2, AMI, etc. |
| 21:20.40 | Damin | kpfleming: That doesn't help if the SIP channel falls over.. ;) |
| 21:21.03 | kpfleming | true |
| 21:21.09 | kpfleming | so don't let it do that then |
| 21:21.22 | Damin | kpfleming: This box only loads about 35 modules.. I'm not sure chan_iax is even loaded.. ;) |
| 21:22.14 | oej | damin: Use manager to check if the sip channel is up... |
| 21:22.36 | oej | Not good enough but something |
| 21:26.57 | oej | astdb/realtime now working on 1.4, a few cli commands not implemented still... |
| 21:27.54 | oej | this two-hour hack took considerable more time than two hours. |
| 21:27.57 | Damin | Thanks for the ideas.. I'm going to go back to the drawing board and see what I can come up with.. |
| 21:28.06 | oej | Good night and thank you for all the help! |
| 21:28.31 | oej | damin: Working with similar issues with keepalived tomorrow. Let's keep in touch. |
| 21:29.05 | oej | The astdb/realtime stuff was exactly for a failover solution for freePBX installs. |
| 21:41.38 | mnicholson | bah, I will need my queue after all, I don't need it for bluetooth enabled phones, but I will need it for headsets |
| 21:41.52 | mnicholson | i could actually handle them from a different thread... |
| 21:44.36 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 22:21.37 | kpfleming | mnicholson: int zero = -1; |
| 22:21.40 | kpfleming | that would be better |
| 22:21.51 | mnicholson | kpfleming, i need it to timeout |
| 22:21.59 | kpfleming | i know, i was just trolling |
| 22:22.03 | mnicholson | oh :) |
| 22:22.37 | mnicholson | I am planning to rewrite most of that function soon |
| 22:36.03 | *** join/#asterisk-dev DarkRift (n=dark@bas1-sthubert21-1279566223.dsl.bell.ca) |
| 22:52.50 | Damin | Hmm.. |
| 22:55.03 | Damin | app_voicemail: Any way to have the system confirm the extension that you are forwarding a Voicemail to? For example, if you forward a message, it asks for the extension.. most other systems will then play back the name of the person that you selected and give you a chance to accept/deny it before forwarding. This is a small, but really important feature for mobile users, because they interact w/ the voicemail via DTMF and it's easy to muck up the DTMF |
| 22:57.05 | snuff-home | mm.. codefreeze-lap that thing i raised about ast_expr2.y.something and it not liking the word 'pure-parser'.. would it be easy/hard to change 'pure' to something else? |
| 22:57.33 | Corydon76-dig | Damin: did you mean to ask that as a development question? |
| 22:57.51 | Corydon76-dig | Because, yeah, we could develop a ton of stuff |
| 22:58.24 | codefreeze-lap | snuff-home: pretty hard. The "pure" part is a kinda (less than 2 years old) feature that allows thread-safety. |
| 22:58.45 | codefreeze-lap | uh, kinda "new" feature, I mean |
| 23:00.17 | *** join/#asterisk-dev shazaum (n=shazaum@unaffiliated/shazaum) |
| 23:01.51 | snuff-home | mmm.. i thought the value 'pure-parser' was just like a word.. |
| 23:07.49 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 23:08.19 | codefreeze-lap | snuff-home: no, it's an option directing what style of parser to generate. |
| 23:10.41 | snuff-home | ah k.. wonder why solaris doesn't like it |
| 23:10.52 | snuff-home | it used to never have issue |
| 23:11.58 | *** join/#asterisk-dev moy (n=chatzill@74.12.124.158) |
| 23:23.47 | *** join/#asterisk-dev russellb_ (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 23:23.47 | *** mode/#asterisk-dev [+o russellb_] by ChanServ |
| 23:29.45 | *** join/#asterisk-dev elitecoder (n=liq@apollo.bullethost.com) |
| 23:31.20 | elitecoder | I'm using asterisk to setup a phone system - is this the right channel? or is this developing asterisk itself? |
| 23:31.42 | ScribbleJ | YOu want #asterisk |
| 23:31.45 | ScribbleJ | This is the latter. |
| 23:31.47 | elitecoder | ahh ok |
| 23:31.49 | elitecoder | thanks |
| 23:31.58 | *** part/#asterisk-dev elitecoder (n=liq@apollo.bullethost.com) |
| 23:32.06 | kpfleming | d00d... that was elitecoder! |
| 23:32.11 | ScribbleJ | "Oh, sorry, this is abuse. You want #12B, just down the hall. ..... twit."" |
| 23:32.13 | *** join/#asterisk-dev qdk (n=qdk@0x573d8dd1.bynqu2.dynamic.dsl.tele.dk) |
| 23:53.24 | codefreeze-lap | snuff-home: hey, find out the version of bison on the system that's complaining |