| 00:14.00 | *** join/#asterisk-dev obnauticus (n=obnautic@c-67-160-183-109.hsd1.or.comcast.net) |
| 00:42.38 | *** join/#asterisk-dev seanbright (n=seanbrig@c-69-251-175-43.hsd1.md.comcast.net) |
| 00:44.00 | *** join/#asterisk-dev citats (n=james@mrplow.gnuinternet.com) |
| 00:44.00 | *** join/#asterisk-dev JT (n=j@unaffiliated/jt) |
| 01:03.37 | *** join/#asterisk-dev puzzled_ (n=patrick@puzzled.xs4all.nl) |
| 01:17.48 | *** join/#asterisk-dev jpeeler (n=jpeeler@adsl-065-005-230-151.sip.lft.bellsouth.net) |
| 01:24.45 | *** join/#asterisk-dev outtolunc (n=me@c-67-174-216-60.hsd1.ca.comcast.net) |
| 01:39.59 | *** part/#asterisk-dev ManxPower (n=manxpowe@209.16.72.139) |
| 01:45.07 | *** join/#asterisk-dev seanbright (n=seanbrig@c-69-251-175-43.hsd1.md.comcast.net) |
| 02:11.22 | *** join/#asterisk-dev g0mb0 (n=root@external.micom.mng.net) |
| 02:29.32 | *** join/#asterisk-dev ta^3 (n=tacvbo@189.146.195.7) |
| 02:31.24 | *** join/#asterisk-dev obnauticus (n=obnautic@c-67-160-183-109.hsd1.wa.comcast.net) |
| 02:51.34 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 02:51.34 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 03:46.24 | *** join/#asterisk-dev ManxPower (n=manxpowe@ip72-204-168-61.no.no.cox.net) |
| 04:11.45 | *** join/#asterisk-dev jameswf-home (n=james@ip72-204-221-181.ph.ph.cox.net) |
| 04:25.49 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 04:25.49 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 06:32.15 | mvanbaak | Corydon76-dig: I'll be doing this on debian. it already works here at home |
| 06:32.43 | mvanbaak | but thanks for the headsup |
| 06:39.53 | *** join/#asterisk-dev steliosk (n=Stelios@athedsl-117347.home.otenet.gr) |
| 06:48.21 | Corydon76-dig | mvanbaak: Debian patches are the reason for the crash on Ubuntu |
| 07:12.56 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:47.01 | *** join/#asterisk-dev lateil (n=oz@89.106.0.21) |
| 08:10.32 | *** join/#asterisk-dev JT (n=j@unaffiliated/jt) |
| 08:10.44 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 08:18.11 | *** join/#asterisk-dev oej (n=olle@soll4-125.cust.blixtvik.net) |
| 08:18.11 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:29.43 | mvanbaak | meh |
| 08:29.44 | mvanbaak | nice |
| 08:36.09 | g0mb0 | sergee, are you there |
| 08:36.11 | g0mb0 | ? |
| 08:50.24 | *** join/#asterisk-dev IgorG (n=IgorG@host-195-162-53-193.pppoe.omsknet.ru) |
| 08:54.28 | tzafrir | Corydon76-dig, hmmm... what Debian patches? |
| 09:39.42 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 10:37.58 | *** join/#asterisk-dev qdk_ (n=qdk@85.235.253.139) |
| 10:43.06 | *** join/#asterisk-dev RoyK (n=roy@fw.fortel.no) |
| 11:07.02 | *** join/#asterisk-dev _ys (i=yuri@91.151.196.254) |
| 11:45.58 | jmls | dang. anyone seen putnopvut ? |
| 11:47.37 | snuff-home | nope |
| 11:48.29 | jmls | mm. think I've found another bug in realtime queues |
| 11:51.50 | snuff-home | ah.. not bad :) |
| 11:56.17 | jmls | snuff-home: it depends on what "not bad" means to people ;) |
| 11:56.51 | jmls | I think that you can have calls waiting in a queue, available agents, but the calls won't go through ... |
| 11:58.37 | snuff-home | mm.. not so great.. |
| 11:59.07 | atis_work | jmls, it depends |
| 11:59.25 | atis_work | if call joins, and then you add realtime agent in queue, it won't be loaded after |
| 12:00.03 | atis_work | you will have to do "show queue 123" or something like that to refresh agents |
| 12:00.05 | jmls | yeah, that's what I'm hitting. |
| 12:00.23 | atis_work | i'v found that a month ago :p |
| 12:00.34 | jmls | However, I'm not adding agents, just not returning them if they are paused. Think of a view on "available" agent |
| 12:00.57 | atis_work | jmls, for that you have "AgentPause" and "AgentUnpause" AMI actions |
| 12:01.08 | jmls | not in realtime. yuk. |
| 12:01.16 | atis_work | yes, in realtime too |
| 12:01.41 | jmls | that's just so hackish. I can lead to differing states between realtime and asterisk members |
| 12:01.49 | jmls | (it can lead) |
| 12:02.06 | atis_work | there's no difference |
| 12:02.21 | jmls | so asterisk has a paused member, while realtime has the same member not paused |
| 12:02.30 | atis_work | no, asterisk will update realtime table |
| 12:02.47 | jmls | that's not what I've seen |
| 12:03.04 | atis_work | it's working for me :p |
| 12:03.45 | atis_work | jmls, http://www.pastebin.org/21308 |
| 12:04.58 | atis_work | jmls, check again - updated with log |
| 12:05.43 | atis_work | ok, log is from system with bunch of my custom patches, i like to update lastcall on pause/unpause |
| 12:05.46 | atis_work | but anyway - it works |
| 12:06.28 | atis_work | oh, and you need to have uniqueid |
| 12:06.44 | atis_work | and you will probably first get select |
| 12:07.10 | atis_work | M11896 |
| 12:07.14 | MuffinMan | [assigned] [Asterisk] Applications/app_queue 0011896: Optimized update_realtime_member_field function reported by atis http://bugs.digium.com/view.php?id=11896 |
| 12:07.44 | atis_work | this helps avoid extra select, but i hadn't had time yet to polish it |
| 12:09.37 | *** join/#asterisk-dev mosty (n=mostyn@60-241-198-194.static.tpgi.com.au) |
| 12:13.33 | atis_work | jmls, working? |
| 12:13.50 | atis_work | you could drop me a log, i would pinpoint the problem quickly |
| 12:17.22 | jmls | I'm just changing my apps to "show" paused queue members, so I'll see what happens. |
| 12:17.58 | jmls | I'm using sipp to place 50 simultaneous calls into my queue (with "virtual" agents) to stress test ;) |
| 12:18.27 | *** join/#asterisk-dev fakhir (n=fakhir@unaffiliated/fakhir) |
| 12:19.08 | atis_work | hmm, that's strange to test feature with load-test tools |
| 12:19.45 | snuff-home | mm.. chan_jingle is interesting.. basically whole lot of funcs def'ed as int.. but only ever returns 1 for 3/4rds of function |
| 12:19.53 | snuff-home | +s |
| 12:20.36 | snuff-home | aka could be made void.. :) |
| 12:21.05 | atis_work | chan_jingle? rings christmas-bells attachet to casing? :D |
| 12:25.55 | snuff-home | mm.. kind of.. in a virtual way |
| 12:28.00 | atis_work | direct connection to santa-claus :) |
| 12:28.26 | atis_work | asterisk will fulfill kid's dreams :) |
| 12:28.51 | snuff-home | mm well time for me to dream.. |
| 12:29.03 | snuff-home | dont feel like stayin awake |
| 12:45.28 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 12:57.02 | *** join/#asterisk-dev JT (n=j@unaffiliated/jt) |
| 12:57.02 | *** join/#asterisk-dev snuff-home (n=snuffy@60-240-117-7-nsw-pppoe.tpgi.com.au) |
| 12:57.02 | *** join/#asterisk-dev jmls (n=jmls@81.138.42.77) |
| 13:21.50 | *** join/#asterisk-dev paravoid (n=paravoid@zebra.cslab.ece.ntua.gr) |
| 13:30.01 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.110.133) |
| 13:32.18 | *** join/#asterisk-dev lateil (n=oz@89.106.0.21) |
| 13:32.32 | *** join/#asterisk-dev DarkRift (i=dark@bas10-montreal02-1177583194.dsl.bell.ca) |
| 13:38.29 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 13:38.36 | jmls | atis_work: yay! working |
| 13:38.49 | *** join/#asterisk-dev ctooley (n=ctooley@doc-72-47-33-80.maryville.mo.cebridge.net) |
| 13:40.09 | jmls | using sipp to initialize 15 calls per seconds (max 30 calls), extension dials out via a looped span back into a queue, where the virtual agents "take" the call and "do work" for the average length of our call |
| 13:40.43 | jmls | I can monitor the call to check call quality and use the queue stats to make sure that everything works |
| 13:40.53 | jmls | we have 10 agents online |
| 13:41.06 | atis_work | jmls, that's why i wrote pbx_test_framework |
| 13:41.10 | jmls | oh, and this is using realtime queues with curl as the engine |
| 13:41.20 | atis_work | congratulations on success |
| 13:41.43 | jmls | atis_work: I know, but we also need to test our application on the other side - to make sure that everything is done right, calls recorded properly etc etc |
| 13:42.23 | jmls | jbot: hug Corydon76-dig |
| 13:42.23 | jbot | ACTION hugs Corydon76-dig |
| 13:42.44 | atis_work | jmls, i can generate customer and agent calls, customers call to some short codes that match to Login queue, logout queue, etc.. customers can traverse IVRs |
| 13:42.57 | jmls | cool. |
| 13:43.34 | atis_work | huh, sorry - agents call to short codes.. |
| 13:44.09 | atis_work | both sides are on agi |
| 13:44.44 | *** join/#asterisk-dev jmesquita (n=jmesquit@201.7.117.114) |
| 13:46.20 | jmls | heh. What I need to do is to tie in the autodialler virutal agents with the queue virtual agents and I will be able to *thrash* the system ;) |
| 13:47.12 | atis_work | jmls: what? |
| 13:47.17 | oej | Afternoon |
| 13:47.47 | atis_work | oej: hey |
| 13:48.03 | jmls | atis_work: I have a dialler that also can use the virtual agents. So, assign them to the outbound team, assign some more agents to the inbound team and get the dialler to call the queue number ;) |
| 13:48.19 | jmls | hmm. 40 active calls, 40 mixmonitors and asterisk is running at 23% |
| 13:48.25 | jmls | cpu time, that is |
| 13:48.50 | atis_work | jmls: i don't get - what's that dialer. what is outbound team and inbound team.. |
| 13:49.13 | atis_work | jmls: could you be interested in this: http://ftp.iq-labs.net/pbx-test/pbx-test-0.1.0.tar.gz |
| 13:49.25 | jmls | atis_work: I'll have a look. thanks |
| 13:50.11 | atis_work | :) |
| 13:50.32 | jmls | it's as easy as it looks 8-) |
| 14:00.42 | *** join/#asterisk-dev jeffg (i=IJJZuRif@pdpc/supporter/active/jeffg) |
| 14:01.30 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 14:01.30 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 14:01.54 | lmadsen | well hello everybody |
| 14:18.46 | *** join/#asterisk-dev Corydon76-vcch (i=yellow@pdpc/supporter/bronze/Corydon76-home) |
| 14:18.46 | *** mode/#asterisk-dev [+o Corydon76-vcch] by ChanServ |
| 14:18.59 | *** join/#asterisk-dev bkw_ (n=brian@adsl-70-234-184-1.dsl.tul2ok.sbcglobal.net) |
| 14:36.49 | *** join/#asterisk-dev putnopvut (n=putnopvu@216.207.245.1) |
| 14:36.49 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:41.34 | jmls | atis_work: just had that mismatch I was telling you about. |
| 14:41.56 | jmls | One of my "agents" is "available" in my application, but "in use" as a member |
| 14:43.33 | jmls | of the queue |
| 14:44.47 | atis_work | jmls, does "show queue 123" helps? |
| 14:44.55 | jmls | nope |
| 14:45.55 | atis_work | and your app fetches data from queue_members family? |
| 14:46.53 | jmls | yup |
| 14:47.06 | atis_work | weird, "show queue xxx" should load fresh data from realtime |
| 14:47.16 | jmls | it's not the paused flag. asterisk thinks it's "in use" when it isn't |
| 14:47.22 | atis_work | oh |
| 14:47.29 | atis_work | so it's device state |
| 14:47.31 | jmls | ah hah. |
| 14:47.40 | atis_work | maybe hanging channel? |
| 14:47.42 | jmls | I just shut sipp down. All channels are now torn down |
| 14:47.47 | jmls | with the exception of |
| 14:47.52 | atis_work | the one |
| 14:47.54 | jmls | Local/Agent-06@agent Agent-06@agentq:3 Busy Busy() |
| 14:47.55 | atis_work | :p |
| 14:48.09 | atis_work | huh, lock in app_busy |
| 14:48.32 | atis_work | i've seen this sometimes |
| 14:48.40 | Juggie | seanbright: ping |
| 14:48.47 | jmls | definately not good |
| 14:48.51 | atis_work | or no.. probably no, i don't have Busy() in production.. |
| 14:49.11 | atis_work | but i've seen that channels sometimes don't terminate |
| 14:49.44 | jmls | I have busy, because I double-check when putting a call through to an agent. I check that the agent really isn't busy by querying my app. If they are busy, I execute "Busy" in the dialplan |
| 14:50.16 | atis_work | jmls, i know, that's the logic for queue members without agent channel |
| 14:50.31 | jmls | because in the time the queue says "right, let's call agent x", and placing the call, Agent X may have gone "not ready" and pissed off to lunch |
| 14:50.54 | atis_work | jmls, it shouldn't be generally necessary if you're using state_interface |
| 14:51.14 | jmls | and how would you do that in 1.4 ?? |
| 14:51.28 | atis_work | there was a backport of state_interface |
| 14:51.41 | jmls | ;) |
| 14:51.45 | atis_work | i'm gonna pull it to production this week |
| 14:51.57 | atis_work | i think it's stable |
| 14:52.02 | jmls | what I need to do is find out why the channel did not get torn down |
| 14:52.13 | jmls | anyone ? Any thing I should try ? |
| 14:52.39 | atis_work | jmls, if you're not using state_interface, queue will create channel for each agent, i think every second.. |
| 14:52.51 | atis_work | so they will get to Busy() and terminate |
| 14:52.57 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 14:52.57 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 14:53.00 | atis_work | but i don't like that bunch of channels created |
| 14:53.10 | jmls | oh sh*t. School run ! |
| 15:03.37 | *** join/#asterisk-dev elguero (n=elguero@ns1.nashuacs.com) |
| 15:06.19 | *** join/#asterisk-dev seanbright (i=seanbrig@65.207.74.18) |
| 15:09.00 | *** join/#asterisk-dev sruffell (n=sruffell@216.207.245.1) |
| 15:09.01 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 15:33.58 | *** join/#asterisk-dev CunningPike (n=arodgers@vpn.dnv.org) |
| 15:38.28 | *** join/#asterisk-dev jpeeler (n=jpeeler@216.207.245.1) |
| 15:44.09 | *** join/#asterisk-dev kannan (n=kannan@123.201.60.110) |
| 15:44.59 | *** join/#asterisk-dev jameswf-home (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 15:45.33 | *** join/#asterisk-dev flujan (n=flujan@200-160-115-020.static.spo.ctbc.com.br) |
| 15:47.31 | jmls | hey guys |
| 15:47.40 | jmls | I've got a channel that's hanging around |
| 15:48.06 | jmls | I know exactly the circumstances of the call, and have all the channel information here |
| 15:48.11 | jmls | anyone care to help ? |
| 15:48.19 | jmls | latest 1.4 svn by the way |
| 15:49.50 | oej | channel type? |
| 15:50.17 | jmls | local |
| 16:04.55 | jsmith | putnopvut: Regarding your latest checkin on your queue-random-announce branch... doesn't that make it hard for people add or remove announcements without restarting the queue? |
| 16:05.50 | atis_work | jmls: is that channel still up? |
| 16:06.03 | atis_work | mine are usually terminating after hour or so |
| 16:06.39 | atis_work | plus i added "soft hangup" button on web interface, so agent's can hangup their hanging channels.. |
| 16:06.46 | atis_work | but would be good if you dig that deeper :) |
| 16:12.18 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 16:12.18 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 16:12.52 | jmls | atis_work: yes |
| 16:13.14 | jmls | Elapsed Time: 1h39m57s |
| 16:35.31 | jmls | noone wants to look at my orphan channel |
| 16:43.09 | atis_work | jmls, if you can reproduce that, go away and post a bug |
| 16:43.20 | atis_work | i mean - go on, not away :p |
| 16:43.43 | atis_work | i have actually never digged that deep enough |
| 16:44.25 | atis_work | something wrong today with either my keyboard or my fingers.. |
| 16:50.06 | jsmith | jmls: I'd be happy to look at it, but I wouldn't be of any help. |
| 16:50.12 | jmls | "go away". nice. Now my feelings are hurt as well ;) |
| 16:50.26 | jsmith | jmls: I'd just point at it and say "Gee, looks like you have a stuck channel. How strange." |
| 16:50.30 | jmls | I was just wondering if there would be any useful info for someone to look at |
| 16:50.42 | jmls | jsmith: heh |
| 16:52.11 | tzafrir | jmls, if it's yous, it's not really orphan |
| 16:52.22 | tzafrir | s/yous/yours/ |
| 16:53.36 | jmls | tzafrir: it is. It was some poor channel I found hanging around with no parents or siblings, just a ,2 stuck all alone in the world. I've taken pity on it for the short term and have not yet killed it ... |
| 16:54.32 | atis_work | jmls, maybe you can just say "go away" fto it :) |
| 16:55.06 | atis_work | anyway i think somebody will ask you for valgrind |
| 16:55.17 | atis_work | and for now you can kill -11 and bt it |
| 17:00.51 | jmls | I have the core now, thanks. If anyone wants a bt on it, just let me know. |
| 17:01.31 | atis_work | jmls, just do "thread apply all bt".. that will give kilobytes of backtraces :) |
| 17:01.44 | atis_work | huh, "thread apply all bt full" |
| 17:18.54 | *** join/#asterisk-dev joetester (n=patrick@216.191.34.13) |
| 17:19.27 | joetester | file: ping? |
| 17:19.48 | file | pong |
| 17:20.17 | joetester | file: I have a chan_sip implementation question, for my own education. |
| 17:20.28 | file | ask. |
| 17:21.30 | joetester | Ok, so sip sessions are initiated by invites, and terminated by a bye right? In chan_sip, what happens if the one end loses power, crashes or otherwise disconnects without being able to send a bye packet? |
| 17:21.47 | joetester | Correct whatever mistake there is in my understanding, of course. |
| 17:22.31 | file | if session timers are in use then when it goes to make sure the remote side is still alive it would see it is not, and the dialog would be torn down |
| 17:22.44 | file | if rtptimeout was in use then no RTP would pass for some time, and the dialog would be torn down |
| 17:23.02 | anonymouz666 | rtptimeout does not work pretty well in some cases. |
| 17:23.11 | joetester | How do you check if the remote side is still alive? Re-invite? |
| 17:23.24 | anonymouz666 | or update method |
| 17:24.02 | joetester | Are session timers automagically in effect all the time? |
| 17:24.43 | *** join/#asterisk-dev RoyK (n=roy@ip-108-57-149-91.dialup.ice.no) |
| 17:24.56 | anonymouz666 | joetester: You should read Asterisk SIP Session Timer written by Raj Jain |
| 17:25.24 | joetester | anonymous666: what do you mean about rtptimout does not work pretty well in some cases? |
| 17:26.15 | joetester | You mean it may mistakenly detect the other end is not alive anymore? |
| 17:31.05 | anonymouz666 | well, in some cases even RTP is not passing through Asterisk... so rtptimeout can't help at all. |
| 17:31.17 | anonymouz666 | M10665 |
| 17:31.19 | MuffinMan | [closed] [Asterisk] Channels/chan_sip/General 0010665: [patch] SIP Session-Timers Support in Asterisk reported by rjain http://bugs.digium.com/view.php?id=10665 |
| 17:31.31 | anonymouz666 | please take a look at attached files, there is a PDF file |
| 17:40.35 | putnopvut | jmls: very belated response, sorry, but I really don't think that branch is any sort of finalized condition right now so things will get fixed as I go with it. |
| 17:48.33 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 17:51.23 | Qwell | http://pastebin.ca/919202 |
| 17:51.27 | Qwell | russellb: would you be completely against that? |
| 17:51.52 | Qwell | tzafrir: maybe you can take a look too, and see if you can tell me why dash sucks so much? :) |
| 17:52.11 | russellb | Qwell: lol! |
| 17:52.20 | russellb | Qwell: i think that's probably ok ... |
| 17:52.29 | Qwell | freaking dash |
| 17:53.55 | Qwell | Your shell sucks with ~ expansion (specifically, when doing "make install DESTDIR=~/path"). Please replace and try again. |
| 17:53.58 | Qwell | slightly better |
| 17:55.05 | *** join/#asterisk-dev qdk_ (n=qdk@0x573fe57e.bynqu2.broadband.tele.dk) |
| 17:55.32 | Qwell | russellb: I tried to just hack out the ~ by overriding DESTDIR, but it's not wanting to work. :( |
| 17:55.51 | Qwell | DESTDIR:=$(subst ~,$(shell echo ~),$(DESTDIR)) |
| 17:55.55 | tzafrir | ubuntu has dash as /bin/sh, and Debian plans to move to it as well |
| 17:56.00 | Qwell | but because of the way DESTDIR is being passed in, it isn't happy |
| 17:56.08 | Qwell | tzafrir: s/plans to/already has/ |
| 17:56.34 | Corydon76-dig | Yay, noncompliant shells! |
| 17:56.37 | Qwell | tzafrir: try this on ubuntu or a debian unstable box; make isntall DESTDIR=~/asterisk-1.4 |
| 17:56.47 | Qwell | install, rather... |
| 17:57.55 | tzafrir | dash here expands ~ |
| 17:58.02 | Qwell | not when you do it like the above |
| 17:59.08 | tzafrir | (here = Debian Etch) |
| 17:59.55 | tzafrir | If you run: make DESTDIR=~/soemthing, make should get it expanded already, right? |
| 18:00.05 | Qwell | right |
| 18:01.16 | Corydon76-dig | In other news, Ubuntu announced plans today to switch to the Minix kernel, saying the Linux kernel had become too slow and bloated. When pressed for comment, the Ubuntu developers wagged their tongues at the reporter, saying simply, "Pbbbbbt" |
| 18:01.29 | Qwell | boo, ksh fails too |
| 18:02.04 | tzafrir | Corydon76-dig, dash is posix compliant, and is generally faster than bash at running scripts |
| 18:02.18 | Corydon76-dig | tzafrir: bash is the standard now. |
| 18:02.24 | Qwell | because it gets to ignore half of the script ;) |
| 18:02.42 | tzafrir | Debian has had for a long time a policy that any script that is for "/bin/sh" should be for any posix shell |
| 18:02.51 | tzafrir | Ubuntu just used that policy |
| 18:02.55 | Corydon76-dig | tzafrir: and Minix isn't POSIX compliant? |
| 18:03.11 | tzafrir | Minix has lousy performance |
| 18:03.34 | tzafrir | dash serves to reduce startup time |
| 18:03.34 | Corydon76-dig | but it's small and tight! |
| 18:03.47 | russellb | i think arguing about shells is really aside from the point here :) |
| 18:03.52 | Corydon76-dig | Who needs Gigabit drivers, anyway? |
| 18:04.01 | russellb | if you say /bin/sh for a script, it damn well better be /bin/sh compatible, that's pretty hard to argue against |
| 18:04.10 | russellb | and really, that is aside from the point |
| 18:04.20 | russellb | the point here, /bin/dash itself is not /bin/sh compatible in this one detail |
| 18:04.32 | russellb | so we're determining whether that is in fact the case, and if so, we can report it |
| 18:04.42 | Qwell | well, ksh does the "wrong" thing too |
| 18:05.05 | Corydon76-dig | russellb: see "hyperbole" |
| 18:07.15 | tzafrir | A quick test shows that zsh and busybox ash also fail to expand 'echo A=~' as 'echo A=$HOME' |
| 18:07.23 | tzafrir | I haven't checked other shells |
| 18:07.38 | russellb | Corydon76-dig: ok. |
| 18:08.39 | Corydon76-dig | tzafrir: the "let's change shit; never mind what it breaks" is a pet peeve of mine |
| 18:09.28 | Qwell | A104139 |
| 18:09.33 | MuffinMan | [104139] Committed by qwell on 2008-02-26 14:09:13: Since all shells aren't as awesome as bash, we have to fail if somebody tries to use a literal "~" in DESTDIR. |
| 18:09.50 | Corydon76-dig | Heh |
| 18:10.14 | jameswf | who doesnt use bash?? bsd users bah on them anyway |
| 18:10.36 | Qwell | that's another thing that annoys me. It's very difficult to specify bash in a shell script |
| 18:11.23 | Qwell | #!/usr/local/something/silly/and/non/standard/bin/bsd/bash just doesn't cut it |
| 18:11.50 | tzafrir | Qwell, right. You need to set SHELL to make |
| 18:12.17 | tzafrir | hmm... sorry, wrong me |
| 18:12.30 | jmls | any cross-compilation gurus lurking ? |
| 18:13.19 | tzafrir | you can use env . And hope it's in the PATH. And give up passing parameters |
| 18:13.30 | Qwell | tzafrir: yeah, exactly |
| 18:15.35 | jmls | heh |
| 18:16.03 | jmls | A67026 |
| 18:16.04 | russellb | uh oh |
| 18:16.05 | MuffinMan | [67026] Committed by russell on 2007-06-04 13:02:31: Change the configure script to build a test program against libcurl to make |
| 18:16.16 | jmls | I'm trying to port curl and curl support into astlinux, and I can get curl compiled. However, the asterisk make is choking |
| 18:16.30 | jmls | ha! : /usr/src/astlinux/build_i586/staging_dir/bin-ccache/../lib/gcc/i586-linux-uclibc/4.1.2/../../../../i586-linux-uclibc/bin/ld: cannot find -lcurl |
| 18:16.51 | Corydon76-dig | libcurl is, uh, needed |
| 18:17.02 | jmls | libcurl is, uh, there |
| 18:17.23 | Corydon76-dig | Is it in one of the library directories? |
| 18:17.23 | jmls | build_i586/root/usr/include/curl/curl.h |
| 18:17.47 | Corydon76-dig | Okay, that's the include... what about the library |
| 18:18.03 | jmls | which file ? libcurl.so ? |
| 18:18.07 | Corydon76-dig | Correct |
| 18:18.25 | Corydon76-dig | or libcurl3.so, if it's named that |
| 18:18.40 | jmls | I can select the func_curl in make menuselect, so I assumed that asterisk can find it when configuring |
| 18:19.28 | jmls | build_i586/root/usr/lib/libcurl.so |
| 18:19.52 | jmls | build_i586/root/usr/lib/libcurl.so.4 |
| 18:20.41 | Corydon76-dig | Is that directory in your /etc/ld.so.conf ? |
| 18:21.17 | Corydon76-dig | or whatever ld uses in your cross-compile? |
| 18:22.15 | Corydon76-dig | Ah, that's why menuselect let you select it. The configure script can't do testing on your target platform, only on the compile platform |
| 18:22.39 | Corydon76-dig | It test-compiles on your compile platform, which is why it's selectable |
| 18:24.46 | russellb | it can certainly do a test compile for the target |
| 18:24.59 | russellb | it may be using the wrong AC macro and is having it compile for the host |
| 18:25.05 | russellb | which is probably a 1 line configure.ac fix |
| 18:25.33 | jmls | in asterisk or astlinux ? |
| 18:25.46 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 18:25.56 | russellb | someone check the autoconf docs for AC_COMPILE_IFELSE() to see what it says |
| 18:26.20 | russellb | also, check AC_PATH_TOOL() |
| 18:26.25 | russellb | look for host vs. target comments ... |
| 18:27.41 | russellb | i must go get food, or i will die |
| 18:28.09 | jmls | where would I find the autoconf docs ? |
| 18:28.29 | russellb | http://www.gnu.org/software/autoconf/manual/ |
| 18:28.37 | jmls | oh. ;) |
| 18:31.44 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.111.94) |
| 18:33.51 | jmls | and me |
| 18:34.53 | jmls | right. Found the code in configure.ac |
| 18:35.09 | jmls | but, I really haven't the faintest idea what I'm looking at. Sorry. |
| 18:35.10 | jmls | http://www.pastebin.ca/919260 |
| 18:35.47 | jsmith | I'm also a firm believer in doing if statements backwards: if (0==some_var) |
| 18:36.01 | jsmith | That way, if you forget the second equals sign, the compiler throws an error |
| 18:37.37 | outtolunc | jmls: what was the question again? |
| 18:38.18 | jmls | outtolunc: cross-compiling under astlinux causes asterisk to bork |
| 18:38.25 | jmls | with func_curl |
| 18:38.34 | jmls | usr/src/astlinux/build_i586/staging_dir/bin-ccache/../lib/gcc/i586-linux-uclibc/4.1.2/../../../../i586-linux-uclibc/bin/ld: cannot find -lcurl |
| 18:39.14 | outtolunc | whereis your curl.h? |
| 18:39.23 | jmls | build_i586/root/usr/include/curl/curl.h |
| 18:39.53 | outtolunc | is that in your <include> path ? |
| 18:40.02 | jmls | yes |
| 18:40.12 | jmls | russellb asked me to look at configure.ac, specifally AC_COMPILE_IFELSE |
| 18:40.21 | jmls | but I have no idea what I'm doing :( |
| 18:42.00 | outtolunc | try echo'ing ac_cv_curl_h and curl_version() just above line 22 |
| 18:42.20 | outtolunc | and or CPPGFLAGS |
| 18:42.23 | outtolunc | er-G |
| 18:45.51 | *** join/#asterisk-dev oej (n=olle@soll4-125.cust.blixtvik.net) |
| 18:45.51 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:54.06 | *** join/#asterisk-dev Netgeeks (n=chris@204.11.231.198.static.etheric.net) |
| 18:56.05 | mvanbaak | hhmm, braces are mandetory now eh ? |
| 18:57.07 | mvanbaak | oej: nice image you took from me with the XO |
| 18:57.26 | Netgeeks | history question here. Running 1.4.5 as an intermediary in a sip call call it node b. call comes in from remote node a , receieved by b, b calls node c and reinvites out of media between a and c. a sends a hangup to b. b instead of sending hangup to c, tries to reinvite the media back to itself from c, but sends no codecs. c sends back 488 to b. b does nothing. c either has to send hangup individually or timeout has to kill |
| 18:57.46 | mvanbaak | http://michiel.vanbaak.info/temp/IMG_5811.jpg |
| 18:57.49 | mvanbaak | safe for work |
| 18:57.57 | Netgeeks | if I disable reinvite, then no problems |
| 18:58.05 | seanbright | i think the non-brace style should be deprecated for 1.6, and then removed in 1.8 ;) |
| 18:58.06 | oej | mvanbaak: I just fail to see the child :-) |
| 18:58.22 | mvanbaak | hehehe, yeah |
| 18:58.32 | Netgeeks | mvanbaak: I played with one of those last week. the design is pretty neat, I didn't much like the os |
| 18:58.39 | mvanbaak | but your new project sounds great: one more laptop per geek |
| 18:58.45 | oej | Netgeeks: Sounds like a bug |
| 18:59.40 | Netgeeks | hey Olle: yeah, it does, just wondering if it had been fixed since 1.4.5. But I would assume you wold have been the fixer |
| 19:00.08 | oej | ...or file... Josh has been doing a lot of media and fax-related stuff |
| 19:01.45 | file | I would realllllllly try latest... |
| 19:02.06 | Netgeeks | File: I can't try latest, I get nailed by another bug |
| 19:02.52 | Netgeeks | I've got a really complicated dial plan that uses alot of variable space and macros and when I went to 1.4.11 from 1.4.5, I ran into stack corruption when using macros |
| 19:03.50 | file | well I can't remember that far back :) |
| 19:04.15 | Netgeeks | where is the patch notes (link please) so I can search |
| 19:04.25 | file | patch notes? |
| 19:04.40 | file | SVN keeps a complete history of every commit done along with relevant issue, plus bugs.digium.com has every bug reported |
| 19:04.41 | Qwell | tzafrir: around? |
| 19:04.56 | Netgeeks | yeah, for the latest 1.4 release - the one autegenerated from patches? |
| 19:04.58 | tzafrir | yes |
| 19:05.06 | Qwell | M12071 |
| 19:05.08 | MuffinMan | [new] [Asterisk] Channels/chan_zap 0012071: chan_zap fails to close file descriptors in case of an error reported by tzafrir http://bugs.digium.com/view.php?id=12071 |
| 19:05.14 | Qwell | what's the error that you get there when it fails? |
| 19:05.22 | Qwell | the incorrect signalling error, or whatever |
| 19:05.26 | JunK-Y | quick question, with TBCT, when you're making the outbound dial, it will automatiquely bridge the inbound and outbound channel, right? |
| 19:05.29 | tzafrir | could be |
| 19:05.45 | JunK-Y | so how we can do a conference with an outbound # ?? |
| 19:05.49 | tzafrir | in that specific example - incorrect signalling |
| 19:05.51 | file | Netgeeks: http://svn.digium.com/svn/asterisk/tags/1.4.18/ChangeLog that? |
| 19:05.55 | JunK-Y | in a way, we need to stay in the path |
| 19:06.01 | Netgeeks | file: yes, thanks |
| 19:06.13 | file | JunK-Y: what? |
| 19:06.29 | tzafrir | Qwell, in that example the error is for the first channel. Naturally this is not always the case |
| 19:07.39 | Qwell | not finding any error message with "incorrect sig" in it... |
| 19:08.02 | JunK-Y | file: im still getting: |
| 19:08.04 | JunK-Y | Handle Q.932 ROSE return error component |
| 19:08.05 | JunK-Y | Unable to handle return result on switchtype 3! |
| 19:08.10 | JunK-Y | when im making my outbound call |
| 19:10.56 | Netgeeks | file: okay, no patch in that changelog, I'll document this fully and add a bug report in mantis |
| 19:11.33 | file | I will ask you to try the latest... |
| 19:11.38 | Qwell | tzafrir: I can't find where that error is coming from.. |
| 19:12.17 | tzafrir | Qwell, the original error, or the error on "reload" (unload; load)? |
| 19:12.33 | Qwell | Signalling requested on channel %d is %s but line is in %s signalling |
| 19:12.35 | Qwell | That, maybe? |
| 19:12.36 | Netgeeks | okay, I'll not post a bug then |
| 19:12.57 | *** join/#asterisk-dev snuff-home (n=snuffy@60-240-117-7-nsw-pppoe.tpgi.com.au) |
| 19:12.59 | tzafrir | yes |
| 19:13.09 | JunK-Y | file: any idea why that TBCT doesnt work? |
| 19:13.19 | file | JunK-Y: because there might be a protocol issue? |
| 19:13.24 | tzafrir | Qwell, but this is not the topic of the bug. That is a configuration error |
| 19:13.34 | Qwell | tzafrir: yes, I know |
| 19:13.45 | Qwell | I was just trying to find the original error, to see why it wasn't closing the channel |
| 19:13.50 | tzafrir | ah, ok |
| 19:14.33 | JunK-Y | file: i realized that, but any hint to fix it? |
| 19:14.43 | file | JunK-Y: ...no |
| 19:16.10 | jsmith | JunK-Y: OK, here are my notes on TBCT |
| 19:16.32 | jsmith | 1) Make sure you have 'facilityenable=yes' in zapata.conf for those channels |
| 19:16.39 | JunK-Y | jsmith: transfer=yes too |
| 19:16.43 | JunK-Y | already done. |
| 19:16.47 | jsmith | 2) Make sure you have 'transfer=yes' in zapata.conf for those channels |
| 19:17.22 | jsmith | 3) Make sure you're using a recent version of libpri |
| 19:17.22 | JunK-Y | ive 1.4.3 |
| 19:18.07 | JunK-Y | put your hands in the air, puts ur hands in the air ! :) |
| 19:18.12 | jmls | Jesus, I thought that my code was unreadable. This autoconf stuff really ... is .... painful |
| 19:18.14 | Qwell | <PROTECTED> |
| 19:18.15 | Qwell | <PROTECTED> |
| 19:18.17 | Qwell | am I reading that wrong? |
| 19:18.30 | jmls | heh. I really have no idea ! |
| 19:19.12 | Qwell | if we fail to ast_calloc tmp, free(tmp) (and segfault)? |
| 19:20.24 | snuff-home | mm interesting.. |
| 19:48.42 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:23.49 | *** join/#asterisk-dev Iamnacho (n=Iamnacho@ip68-103-153-140.ks.ok.cox.net) |
| 20:25.11 | *** join/#asterisk-dev RoyK (n=roy@ip-108-57-149-91.dialup.ice.no) |
| 20:34.46 | *** join/#asterisk-dev kpfleming (n=kpflemin@216.207.245.1) |
| 20:34.46 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 20:45.31 | *** join/#asterisk-dev andreg (n=andreg@71-209-140-34.phnx.qwest.net) |
| 20:48.20 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:50.24 | *** join/#asterisk-dev mosty (n=mostyn@60-241-198-194.static.tpgi.com.au) |
| 20:56.54 | snuff-work | jmls: there was a patch to speed up mixmon or record.. caching 32k buffers then putting in file.. its on mantis somewhere |
| 20:59.01 | *** join/#asterisk-dev janinge (n=janinge@ninge.net) |
| 20:59.40 | joetester | I have a linux c++ related question? It has a sort of distant link to asterisk. |
| 20:59.54 | joetester | Or C/ |
| 21:00.00 | joetester | I mean C dammit. |
| 21:00.42 | tzafrir | so ask. It's quiet now |
| 21:01.09 | joetester | All right, if I want to alter a file timestamp from a C/C++ program, do I have to use system(touch blabla)? |
| 21:01.23 | joetester | Or is there a better way? |
| 21:01.47 | snuff-work | M11962 |
| 21:01.48 | MuffinMan | [new] [Asterisk] Resources/res_monitor 0011962: [patch] Simple file recording buffer - great performance gain reported by garlew http://bugs.digium.com/view.php?id=11962 |
| 21:02.34 | mosty | joetester, /bin/touch is a binary file, so there must be a lower level way to do it |
| 21:04.01 | joetester | I would assume so too... I'm not finding it though. |
| 21:04.53 | mosty | joetester, you could always get the source for /bin/touch - it's probably written in c |
| 21:06.13 | tzafrir | joetester, utime(2) ? |
| 21:06.46 | joetester | Wassat tzafrir? |
| 21:06.58 | seanbright | joetester: a C library routine to do what you want? |
| 21:07.02 | seanbright | `man utime` |
| 21:07.05 | joetester | Sweet. |
| 21:07.23 | tzafrir | (2) means from section 2 of the manual |
| 21:07.30 | tzafrir | man 2 utime |
| 21:07.46 | joetester | I have no utime at all...? |
| 21:07.56 | tzafrir | What distro is that? |
| 21:08.01 | joetester | debian? |
| 21:08.14 | tzafrir | aptitude instal lmanpages-dev |
| 21:08.21 | tzafrir | aptitude install manpages-dev |
| 21:08.50 | joetester | Sweeeeeet. |
| 21:09.39 | joetester | Thanks, seems like what I was looking for... exarctly. |
| 21:10.38 | joetester | How come I didn't have the dev manpages? |
| 21:11.06 | mvanbaak | because you did not install them |
| 21:11.08 | mvanbaak | ;) |
| 21:12.34 | joetester | Good to know, I'll be more careful in the future, and I will try not to humiliate myself too much. |
| 21:16.40 | Corydon76-dig | Also: apt-get install manpages-posix-dev |
| 21:17.07 | Corydon76-dig | and: apt-get install funny-manpages |
| 21:18.32 | Qwell | funny-manpages? |
| 21:20.16 | Corydon76-dig | uubp - Unix-to-Unix beer protocol |
| 21:21.27 | jeffg | mmm, uubp |
| 21:23.42 | joetester | w00t... worked perfectly tzafrir! |
| 21:23.55 | joetester | you rock! |
| 21:24.14 | Corydon76-dig | I think we need res_rescrog |
| 21:24.34 | Corydon76-dig | rescrog — change something, make it different |
| 21:24.51 | Corydon76-dig | BUGS: Repeated rescrogs done too quickly will lead to advanced technology beyond our comprehension. |
| 21:25.24 | Corydon76-dig | So that's the problem... we don't have bugs, we have advanced technology beyond our comprehension... |
| 21:25.32 | tzanger | Corydon76-dig: haha |
| 21:26.08 | tzanger | Corydon76-dig: that's my new msn name |
| 21:27.24 | jeffg | http://www.anvari.org/fortune/Best_Fortunes/311_clarke-s-third-law-any-sufficiently-advanced-technology-is-indistinguishable-from-magic.html |
| 21:27.57 | mvanbaak | oej made res_obsd ;) |
| 21:34.10 | codefreeze | Corydon76-dig: "man party" looks interesting. So does grope... |
| 21:36.58 | *** join/#asterisk-dev fakhir (n=fakhir@unaffiliated/fakhir) |
| 21:48.00 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 21:57.17 | *** join/#asterisk-dev janinge (i=j@ninge.net) |
| 22:09.02 | *** join/#asterisk-dev tzafrir (n=tzafrir@62.90.10.53) |
| 22:09.02 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 22:31.16 | Netgeeks | file: you around? |
| 22:33.46 | file | eh? |
| 22:35.27 | Netgeeks | 1.4.18 sorta works right, but still has what I would term - wierdness. |
| 22:35.54 | snuff-work | anyone has comment on that bug/feature 11962? |
| 22:36.03 | Netgeeks | it actually sends a valid codec in the reinvite after bye, so everything works right in the end, but it shouldn't be rinviting, it should just BYE |
| 22:36.10 | Netgeeks | works good enough for me tho |
| 22:36.26 | Netgeeks | so there is the answer to 'try the latest' |
| 22:38.02 | file | uh huh |
| 22:39.54 | Netgeeks | okay! |
| 22:43.16 | Netgeeks | but you would consider that wierd no? scenario is a invites c@b, b invites c and reinvites c to a, a to c. Call works... a sends bye to b, b acks a. b sends reinvite to c bringing the media back to itself, c OK's the sdp change, b acks c's okay, b sends bye to c, c ok's bye |
| 22:43.48 | Netgeeks | all that reinvite after the first bye is un-needed, no? |
| 22:44.11 | file | I can't answer without seeing traces and stuff... and i'm soon out of here. |
| 22:45.15 | Netgeeks | okay, no worries, I'll throw together traces, and such and drop it into pastebin some time you have time |
| 22:50.11 | mvanbaak | SILENCE !!! |
| 22:50.45 | Corydon76-dig | "I keel you" |
| 22:51.48 | *** join/#asterisk-dev sruffell (n=sruffell@216.207.245.1) |
| 22:51.49 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 22:52.17 | Corydon76-dig | Afternoon, Shaun |
| 22:52.55 | mvanbaak | :) |
| 22:53.24 | mvanbaak | 'Are you my virgins ????' |
| 23:05.39 | jmls | eek. |
| 23:05.41 | jmls | [Feb 26 21:47:57] WARNING[31871]: app_queue.c:746 create_queue_member: No location at interface '' |
| 23:05.45 | jmls | bam! segfault |
| 23:09.13 | jmls | bt full at http://www.pastebin.ca/919727 |
| 23:09.16 | Deeewayne | Corydon76-dig, ping |
| 23:15.35 | *** join/#asterisk-dev jmls_work (n=jmls@mail.tessera.co.uk) |
| 23:22.19 | jmls_work | eek |
| 23:22.23 | jmls_work | another segfault |
| 23:22.26 | jmls_work | exactly same place |
| 23:22.28 | jmls_work | #0 0x08100338 in ast_copy_string (dst=0xb683cfcd "", src=0x0, size=80) at /usr/src/asterisk/include/asterisk/strings.h:168 |
| 23:23.45 | jmls_work | this is latest 1.4 svn |
| 23:23.55 | *** join/#asterisk-dev sruffell (n=sruffell@216.207.245.1) |
| 23:23.55 | jmls_work | looks like I'll have to log a bug |
| 23:23.55 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 23:25.08 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 23:25.08 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 23:26.04 | jmls_work | mmm. running a patched app_queue.c |
| 23:26.11 | Corydon76-dig | Deeewayne: pong |
| 23:26.12 | jmls_work | let me revert and retest |
| 23:31.05 | Deeewayne | Corydon76-dig, gotcha in jabber :-) |
| 23:48.53 | Qwell | kpfleming: that's...it? |
| 23:49.11 | Qwell | silly...memory |
