| 00:06.08 | Qwell | did you measure how much you had? |
| 00:07.15 | elixer | nope |
| 00:07.22 | elixer | just the only clean spoon i had :-) |
| 00:34.25 | russellb | how is fundamentally wrong? |
| 00:34.33 | russellb | and was it metric, or silly american units? |
| 00:54.24 | *** join/#asterisk-dev mitcheloc (n=mitchel@adsl-68-120-230-95.dsl.irvnca.pacbell.net) |
| 01:00.19 | *** join/#asterisk-dev wunderkin (n=wunderki@ip72-223-86-126.ph.ph.cox.net) |
| 01:19.05 | blitzrage | metric left! |
| 01:28.46 | elixer | heh |
| 01:29.11 | elixer | 1 table spoon |
| 01:30.44 | elixer | so silly american units |
| 01:31.19 | Qwell | are there metric spoon units? O.o |
| 01:31.40 | elixer | i don't believe so |
| 01:31.57 | elixer | there's the american way, and the wrong way |
| 01:31.59 | elixer | heh |
| 01:32.08 | denon | there are tablespoons in metric |
| 01:32.39 | denon | my wife grew up on metric |
| 01:32.40 | elixer | well it was 1 US tablespoon |
| 01:32.53 | denon | and she says tablespoons there and here in the US are exactly the same |
| 01:33.20 | denon | 14ml of a half an ounce |
| 01:33.30 | denon | s/of/or/ |
| 01:37.13 | elixer | well either way... good yogurt. |
| 01:37.18 | elixer | :-) |
| 01:37.29 | Qwell | we're all such geeks |
| 01:37.54 | elixer | *crickets* |
| 01:37.57 | elixer | heh |
| 01:37.58 | outtolunc | a tablespoon is the exact amount that causes a child to run around the kitchen table twice after being given cough medicine (whereas a teaspoon only causes 1 lap around) <G> |
| 01:38.39 | Qwell | so, it gives a 50% reduction in table running? |
| 01:39.28 | elixer | well 1 tablespoon is 3 teaspoons... so it /should/ be a 33.3% reduction |
| 01:39.32 | elixer | something doesn't add up. |
| 01:39.36 | Qwell | I fail at math. |
| 01:39.41 | elixer | err |
| 01:39.44 | Qwell | no, you do |
| 01:39.46 | elixer | a 66.6% reduction |
| 01:40.06 | Qwell | 1 teaspoon == 1 lap. 3 teaspoons == 3 laps |
| 01:40.13 | Qwell | 1 tablespoon == 2 laps |
| 01:40.23 | elixer | ... carry the 1 ... |
| 01:40.24 | Qwell | thus, a reduction of 50% if administered at once :P |
| 01:40.26 | outtolunc | the child made wider laps due to speed increase <G> |
| 01:40.26 | elixer | ohhh |
| 01:40.28 | elixer | yeah, you're right |
| 01:40.58 | elixer | it is amazing i can even dress myself |
| 01:40.59 | elixer | heh |
| 01:43.29 | elixer | so what needs to be fixed/added? i'm ready to make something other than text changes |
| 01:43.32 | elixer | hit me |
| 01:43.52 | Qwell | bugs.digium.com :) |
| 01:43.57 | Qwell | pick your poison |
| 01:44.07 | elixer | everything already has a [patch] or is assigned |
| 01:45.48 | Juggie | assigned doesnt mean its fixed |
| 01:45.54 | Juggie | assigned just means that person is most likely to fix it |
| 01:46.24 | Juggie | but if you want to fix something say assigned to russellb, just ask here and say hey have you investigated/worked on bug #1234 yet |
| 01:46.44 | elixer | alrighty |
| 01:53.43 | Juggie | elixer, theres also a asterisk janitor project |
| 01:53.52 | Juggie | if you google for asterisk janitor you'll find it |
| 01:54.09 | elixer | k |
| 01:54.38 | elixer | is this list maintained? |
| 01:54.48 | Qwell | that should be a janitor project... |
| 01:54.55 | Qwell | update the janitor project page, heh |
| 01:54.58 | elixer | heh |
| 01:59.38 | Juggie | well, if you are not happy w/ the list post to the asterisk-dev mailing list :) |
| 02:00.38 | Juggie | i have a brand new router here i am going to install DD-WRT on and my gf wants me to go to sleep |
| 02:00.43 | Juggie | i put my foot down :P |
| 02:01.19 | Qwell | Juggie: so...see you tomorrow :P |
| 02:01.31 | Juggie | hahah nah, we bought a pack of 11 board games from costco |
| 02:01.33 | Juggie | she is checking those out |
| 02:01.37 | Juggie | that will keep her busy for a while |
| 02:01.54 | elixer | and by "board games" you mean... |
| 02:04.09 | Juggie | monopoly, sorry, clue, crib, checkers, chess... etc. |
| 02:04.13 | Juggie | on a wodden board thingy |
| 02:04.22 | Juggie | its a nice set |
| 02:04.34 | elixer | nice |
| 02:04.47 | Juggie | the boards are all mounted on wood |
| 02:06.33 | Juggie | now, to try and get DD-WRT on this thing |
| 02:28.38 | *** join/#asterisk-dev ZX81 (n=matt@202.20.97.200) |
| 02:31.13 | russellb | yeahhhh ... the janitor projects page needs some love. |
| 02:31.52 | snuff-work | i think most easy tasks have been done.. |
| 02:32.46 | snuff-work | still doxygen for fixing required :) |
| 02:34.10 | russellb | i need to spend some time coming up with a new janitor project list |
| 02:34.29 | russellb | for example, converting structures to use the stringfields API |
| 02:34.44 | russellb | though it needs more explanation than that ... |
| 02:34.53 | russellb | as it isn't optimal for *all* strings ... |
| 02:35.14 | snuff-work | mmm specify which ones need doing :) |
| 02:35.21 | russellb | heh |
| 02:35.38 | d3wayne | o.o |
| 02:35.51 | russellb | d3wayne: !!!!!!! |
| 02:36.01 | d3wayne | hello :-) |
| 02:36.18 | snuff-work | *cough* coverity scan... no progress? |
| 02:36.51 | russellb | snuff-work: no, i have not emailed them |
| 02:36.57 | russellb | keep it up though, i just might do it one of these days. |
| 02:41.35 | russellb | d3wayne: what are you up to? |
| 02:41.46 | russellb | d3wayne: throwing a party after fixing that meetme bug? :) |
| 02:41.58 | Qwell | outtolunc: does it already have zaptel rules by default or something? |
| 02:42.05 | d3wayne | meetme party this weekend! |
| 02:42.08 | russellb | heh |
| 02:42.19 | russellb | i meant are you partying right now over it, heh |
| 02:42.26 | russellb | i would be. |
| 02:42.28 | d3wayne | I was going to tcl be-160's ami script |
| 02:42.46 | russellb | ooh, that tcl's |
| 02:42.52 | d3wayne | lol |
| 02:42.59 | d3wayne | I'll party right now....hold on |
| 02:43.03 | russellb | o.O |
| 02:43.12 | russellb | ha |
| 02:44.03 | Qwell | outtolunc: no zaptel packages or anything installed? |
| 02:44.11 | Qwell | I wonder if that's actually done upstream... |
| 02:44.26 | Qwell | that would be...interesting |
| 02:44.34 | russellb | indeed |
| 02:44.35 | Qwell | do the udev guys actually maintain rules? |
| 02:44.57 | russellb | given that the fedora guys have been so strict about getting zaptel in since it's not in the kernel |
| 02:45.00 | d3wayne | I think tomorrow is the last day of my cold |
| 02:45.07 | Qwell | d3wayne: w00t |
| 02:45.07 | russellb | d3wayne: rockin' |
| 02:45.09 | russellb | colds suck |
| 02:45.12 | russellb | i'm sick of it |
| 02:45.17 | Qwell | har, har, har |
| 02:45.18 | russellb | *zing* |
| 02:45.19 | Qwell | nice pun |
| 02:58.15 | Qwell | note to self; don't read qdb.us |
| 02:58.25 | Qwell | hmm, bad timing |
| 03:02.31 | d3wayne | hmmmm....the issue summary page says issue 20 is the longest open issue (1400+ days) |
| 03:02.43 | Qwell | issue 20? |
| 03:02.58 | d3wayne | its closed, but the resolution is open |
| 03:03.02 | d3wayne | M20 |
| 03:03.03 | MuffinMan | [closed] [Asterisk] Core/General 0000020: [patch] Update MOH to deal with more than just MP3s reported by JerJer http://bugs.digium.com/view.php?id=20 |
| 03:03.36 | Qwell | weird... |
| 03:09.53 | Juggie | i bought a laptop bag @ walmart its pretty nice, came with a usb hub and a mini mouse |
| 03:09.56 | Juggie | was 35$ |
| 03:10.43 | d3wayne | mini mouse? she's hot |
| 03:13.53 | *** join/#asterisk-dev corydon76-home (i=white@pdpc/supporter/sustaining/Corydon76-home) |
| 03:13.53 | *** mode/#asterisk-dev [+o corydon76-home] by ChanServ |
| 03:16.06 | *** join/#asterisk-dev Corydon76-dig (i=purple@pdpc/supporter/sustaining/Corydon76-home) |
| 03:16.06 | *** mode/#asterisk-dev [+o Corydon76-dig] by ChanServ |
| 03:35.36 | *** join/#asterisk-dev mtaht4 (n=m@219-106-62-200.enitel.net.ni) |
| 03:42.06 | *** join/#asterisk-dev mog (n=mog@c-71-207-200-130.hsd1.al.comcast.net) |
| 03:42.06 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 04:56.54 | *** join/#asterisk-dev netson99 (n=netson99@202.153.241.206) |
| 04:57.07 | *** join/#asterisk-dev CunningPike (n=CunningP@S010600095b33697f.vc.shawcable.net) |
| 05:07.34 | *** join/#asterisk-dev Juggie (n=Juggie@CPE001601df17fb-CM000a73a18a20.cpe.net.cable.rogers.com) |
| 05:07.45 | Juggie | doot doot |
| 05:29.43 | jtodd | No dooting here. All asleep. |
| 05:39.28 | denon | zis is KAOS! We do not "doot doot" here! |
| 05:39.34 | denon | (any Get Smart fans in the crowd?) |
| 05:43.30 | jtodd | All of us are Get Smart fans. |
| 05:43.37 | jtodd | Well, most of us are Get Smart Fans. |
| 05:44.16 | jtodd | Would you believe one of us has seen an ad for Get Smart when we were trying to watch Hogan's Heroes? |
| 05:45.28 | jtodd | (now, THAT is an obscure Get Smart joke. Sorry to all our .eu readers for lapsing into bad 60's television rerun trivia from North America.) |
| 05:45.29 | denon | would you believe .. *some* of us are get smart fans? |
| 05:45.42 | denon | ... would you believe .. 2 of us are Get Smart fans? |
| 05:45.48 | denon | would you believe I've seen ads for it .. |
| 05:46.14 | jtodd | Of course, you might say it's on topic since he had a shoe phone. /dev/zap/shoephone1 |
| 05:46.24 | denon | haha |
| 05:46.34 | denon | funnier still, when zapatos is spanish for shoe |
| 05:46.39 | jtodd | Indeed. |
| 05:47.43 | denon | supposedly there's a Get Smart movie coming out soon |
| 05:47.47 | denon | (I mean, another one) |
| 05:48.08 | denon | somehow I just don't think it'll be the same |
| 05:48.08 | jtodd | yeah, it probably will suck. Too bad. |
| 05:48.23 | denon | the first movie was pretty good .. but then again, it was the real characters for max and 99 |
| 05:50.41 | denon | jtodd: btw, Get Smart reached out much further than north america |
| 05:52.31 | jtodd | Hm. Surprising. I'm always shocked to see American TV shows in reruns whenever I'm overseas. Get Smart was amusing, but I think a lot of US humor doesn't translate well (for cultural and language reasons.) |
| 05:52.44 | denon | well, UK countries |
| 05:52.49 | denon | not sure if it was in non-english areas |
| 05:55.06 | jtodd | I have referred to certain parts of Asterisk code and/or devices as "cones of silence" when they malfunction. Only some people get the joke. |
| 06:01.22 | denon | heheh |
| 06:01.31 | denon | personally, I think the cone of silence worked well |
| 06:01.51 | denon | of course, it had the added feature of letting the tv audience in on the secret |
| 06:03.50 | Juggie | am i missing something or can syslog not distinguish logging traffic based on the host which sent it |
| 06:13.31 | *** join/#asterisk-dev dlynes_laptop (n=dlynes@d154-20-9-152.bchsia.telus.net) |
| 06:42.59 | *** join/#asterisk-dev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 06:48.54 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 06:51.47 | yannj_fr | hi everybody |
| 06:57.16 | jmls_coffeeeee | hey |
| 07:15.11 | jmls_coffeeeee | me has a trio of different segfaults now ;) |
| 07:15.13 | jmls_coffeeeee | m10710 |
| 07:15.14 | MuffinMan | [new] [Asterisk] Core-General 0010710: asterisk segfault in ast_list_traverse_safe_begin reported by jmls http://bugs.digium.com/view.php?id=10710 |
| 07:25.40 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 07:34.42 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 07:43.48 | *** join/#asterisk-dev lbow (n=lbow@dsl-146-6-152.telkomadsl.co.za) |
| 07:54.47 | *** join/#asterisk-dev qdk (n=qdk@213.150.62.32) |
| 08:44.23 | *** join/#asterisk-dev IgorG (n=FeedomPa@195.162.32.126) |
| 08:52.34 | *** join/#asterisk-dev Corydon76-home (i=three@pdpc/supporter/sustaining/Corydon76-home) |
| 08:52.34 | *** mode/#asterisk-dev [+o Corydon76-home] by ChanServ |
| 09:08.09 | *** join/#asterisk-dev mvanbaak (i=michiel@vanbaak.xs4all.nl) |
| 09:18.57 | *** join/#asterisk-dev qdk_ (n=qdk@213.150.62.32) |
| 09:31.29 | *** join/#asterisk-dev pnlarsson (n=pnlarsso@c83-248-12-187.bredband.comhem.se) |
| 09:58.23 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 09:59.22 | puzzled | hi |
| 10:27.45 | *** join/#asterisk-dev phearless (n=phear@host217-34-75-65.in-addr.btopenworld.com) |
| 11:15.25 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 11:26.51 | *** join/#asterisk-dev IgorG (n=FeedomPa@195.162.32.126) |
| 11:30.40 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 11:31.01 | *** join/#asterisk-dev marta (n=aaa@myskin.iet.unipi.it) |
| 11:31.04 | marta | hi |
| 11:46.08 | yannj_fr | hi |
| 11:50.50 | *** join/#asterisk-dev putnopvut (n=putnopvu@user-24-214-124-177.knology.net) |
| 11:50.50 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 11:56.22 | *** join/#asterisk-dev Shido6 (n=shido6@204.126.120.132) |
| 12:01.38 | *** join/#asterisk-dev thewiizle (n=nick@87.127.85.42) |
| 12:01.39 | thewiizle | lo |
| 12:03.53 | jmls | hi |
| 12:03.59 | thewiizle | hi |
| 12:08.24 | *** join/#asterisk-dev Laureano (n=chatzill@200.59.172.38) |
| 12:08.30 | Laureano | Hello |
| 12:12.59 | *** join/#asterisk-dev mtaht4 (n=m@219-106-62-200.enitel.net.ni) |
| 12:14.29 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:19.14 | jmls | lots of hi's, no coding ... ;) |
| 12:19.25 | jmls | or is everyone coding, therefore no chat ... |
| 12:19.47 | snuff-away | mmm.. i would code.. if i could help |
| 12:20.01 | file | geek news, not actual news though |
| 12:20.18 | snuff-away | heh slashdot? |
| 12:20.37 | file | and digg |
| 12:20.46 | file | but digg makes my firefox unhappy >_< |
| 12:20.54 | file | 19.3% CPU usage! |
| 12:21.00 | snuff-away | mm.. i like my inquirer.net |
| 12:21.05 | snuff-away | mm that is bad |
| 12:21.37 | snuff-away | open office hates a few of the spreadsheets i have to open :( |
| 12:23.54 | *** join/#asterisk-dev Corydon76-dig (i=five@pdpc/supporter/sustaining/Corydon76-home) |
| 12:23.54 | *** mode/#asterisk-dev [+o Corydon76-dig] by ChanServ |
| 12:24.14 | snuff-away | i know someone commited my 'solaris' fixes from yesterday, the 'fcntl' stuff i added.. |
| 12:25.54 | putnopvut | snuff-away: was there a problem with the fixes? |
| 12:25.57 | snuff-away | if anyone can look over it that knows fcntl well my fix was more so it would compile |
| 12:27.42 | snuff-away | it seems ok.. but i'm no fcntl expert |
| 12:29.23 | *** join/#asterisk-dev eliel (n=eliel@200.5.113.234) |
| 12:30.16 | snuff-away | just looked over the few examples i could find |
| 12:31.28 | putnopvut | Before I committed those solaris fixes, I asked about it in here, and I got the okay to commit. |
| 12:31.39 | *** join/#asterisk-dev michael-i (n=michael-@141.41.40.55) |
| 12:31.44 | putnopvut | So if something doesn't work properly, don't feel responsible for it. |
| 12:33.16 | putnopvut | But I think those fixes are just fine, though. |
| 12:33.41 | snuff-away | yey thanks for the support putnopvut :) |
| 12:33.53 | snuff-away | the other fixes were nice n easy |
| 12:56.54 | *** join/#asterisk-dev miguel3239 (n=elguerom@ns1.nashuacs.com) |
| 13:03.06 | *** join/#asterisk-dev ManxPower (n=manxpowe@164.sub-70-222-209.myvzw.com) |
| 13:19.15 | *** join/#asterisk-dev nicchap (n=nicchap@204.101.222.130) |
| 13:21.37 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 13:27.37 | *** join/#asterisk-dev anonymouz666 (n=anonymou@189.25.74.159) |
| 13:33.29 | *** join/#asterisk-dev bbryant (n=12243@c-68-59-20-153.hsd1.sc.comcast.net) |
| 13:33.29 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 13:39.48 | *** join/#asterisk-dev Corydon76-home (i=indigo@pdpc/supporter/sustaining/Corydon76-home) |
| 13:39.48 | *** mode/#asterisk-dev [+o Corydon76-home] by ChanServ |
| 13:42.25 | bbryant | Corydon76-home, do you remember the bug number who's poster was getting neutral karma? |
| 13:43.10 | caio1982 | 10220 |
| 13:43.22 | caio1982 | Database query failed. Error received from database was #1054: Unknown column 'value' in 'field list' for the query: UPDATE mantis_user_karma_scores SET value = -3 WHERE user_id = 6552 |
| 13:46.07 | bbryant | fixed |
| 13:47.21 | caio1982 | metakarma is when you get +1 karma for fixing someone's -3 karma? :) |
| 13:48.19 | *** join/#asterisk-dev jtexter3 (n=jtexter3@12.159.220.114) |
| 13:48.36 | bbryant | lol |
| 13:50.20 | *** join/#asterisk-dev jsmith-teaching (n=jsmith@atelka.info) |
| 13:50.31 | *** join/#asterisk-dev tsurko (n=tsurko@213.91.216.130) |
| 13:51.56 | *** join/#asterisk-dev Digitaldaz21 (n=me@87-194-8-193.bethere.co.uk) |
| 13:53.55 | *** join/#asterisk-dev Corydon76-dig (i=indigo@pdpc/supporter/sustaining/Corydon76-home) |
| 13:53.55 | *** mode/#asterisk-dev [+o Corydon76-dig] by ChanServ |
| 13:54.32 | Digitaldaz21 | Hi all, cna anyone help me please? |
| 13:59.15 | Corydon76-dig | Try #asterisk |
| 13:59.30 | Digitaldaz21 | Cheers! |
| 14:04.09 | *** join/#asterisk-dev _mihai_ (n=_mihai_@38.96.187.252) |
| 14:04.42 | *** join/#asterisk-dev _mihai_ (n=_mihai_@38.96.187.252) |
| 14:15.34 | *** join/#asterisk-dev elixer (i=elixer@65.207.74.18) |
| 14:16.53 | *** join/#asterisk-dev mog (i=mog@nat/digium/x-63525e74b7759a12) |
| 14:16.53 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 14:18.43 | *** join/#asterisk-dev mbranca (n=matteo@host-210-mi.linuxserver.it) |
| 14:24.03 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 14:24.03 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 14:26.52 | *** join/#asterisk-dev marta (n=aaa@87.19.202.200) |
| 14:26.57 | marta | re |
| 14:28.15 | *** join/#asterisk-dev blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 14:34.26 | *** join/#asterisk-dev dioedu (n=dioedu@c915230f.virtua.com.br) |
| 14:50.04 | *** join/#asterisk-dev netson99 (n=netson99@210.210.181.232) |
| 14:54.02 | *** join/#asterisk-dev putnopvu1 (i=putnopvu@nat/digium/x-8c5ad60de63b689d) |
| 14:54.49 | *** join/#asterisk-dev putnopvut (i=putnopvu@nat/digium/x-892e73c0f66c706a) |
| 14:54.49 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:00.25 | *** join/#asterisk-dev km- (n=pgrace@aeneas.fierymoon.com) |
| 15:00.46 | km- | can someone in the know explain to me what putting a zaptel channel in "conference announce" mode does? |
| 15:01.07 | km- | I see the ioctl for it but I'm failing to figure out what it does special to the zap hardware |
| 15:12.03 | russellb | i doubt it has an effect down to the hardware |
| 15:12.22 | russellb | probably just for the zaptel module |
| 15:12.27 | russellb | for what it does with the audio |
| 15:12.39 | russellb | conference announce ... perhaps talk only? |
| 15:13.02 | km- | app_rpt uses it for a zaptel channel |
| 15:13.11 | russellb | i see. |
| 15:13.17 | russellb | i have _no idea_ how that module works |
| 15:13.19 | km- | and I'm evaluating removing the zaptel dependency on tx/rx channels |
| 15:13.20 | km- | yeah. |
| 15:13.42 | km- | it's kinda funky how it works, and everyone swears up and down that it requires zaptel hardware, but nobody specifically says if that's just for the meetme aspect or if it's for the whole shebang |
| 15:13.45 | russellb | you can try asking the asterisk-dev mailing list the question |
| 15:13.56 | russellb | you'd get a bigger audience |
| 15:14.22 | km- | yeah.. |
| 15:14.23 | russellb | well, perhaps they mean it requires the radio hardware that jim dixon makes? |
| 15:14.40 | km- | the radio hardware interfaces into the FXS channels on the zaptel cards |
| 15:14.50 | km- | I'm thinking that you could instead use a modified ATA |
| 15:16.11 | km- | it's just a passing thought I had.. I have zaptel hardware already but I just can't believe that for the module it would have a requirement of only one telephony technology |
| 15:17.57 | km- | I'm going to go hunt down a channel bank for my eventual 12 band repeater :P |
| 15:18.06 | km- | thanks for the help. |
| 15:18.07 | *** part/#asterisk-dev km- (n=pgrace@aeneas.fierymoon.com) |
| 15:19.17 | russellb | jmls: please stop crashing asterisk |
| 15:46.29 | elixer | mmmm, custom ringtones from itunes |
| 15:46.37 | *** join/#asterisk-dev jcmoore (n=jcmoore@unaffiliated/tgrman) |
| 15:46.44 | elixer | the kool-aid is delicious |
| 15:48.31 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.12.183) |
| 15:54.56 | *** join/#asterisk-dev qdk_ (n=qdk@0x50c627be.bynxx11.adsl-dhcp.tele.dk) |
| 15:55.31 | *** join/#asterisk-dev yannj_fr (n=yannj_fr@vpn.intelunix.fr) |
| 16:04.18 | *** join/#asterisk-dev VoicePulse (n=contact@unaffiliated/voicepulse) |
| 16:14.52 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 16:21.51 | *** join/#asterisk-dev deeperror (n=deeperro@69-215-202-195.ded.ameritech.net) |
| 16:36.48 | *** part/#asterisk-dev nicchap (n=nicchap@204.101.222.130) |
| 16:46.20 | deeperror | i've been having a zt_exception that gets thrown and sticks into an endless loop. Any clues? http://pastebin.ca/695949 |
| 16:46.36 | Corydon76-vcch | russellb: we appear to be at an impasse |
| 16:52.11 | file | Corydon76-vcch: I assume you mean re AST_FRAME_DIGITAL |
| 16:52.24 | Corydon76-vcch | Correct |
| 16:53.06 | Corydon76-vcch | Basically it's a case "we already made it work this way and we're unwilling to follow the model you guys have laid out" |
| 16:53.55 | *** join/#asterisk-dev jfg (n=jfg@dyn-83-152-249-84.ppp.tiscali.fr) |
| 16:54.00 | jfg | hi |
| 16:54.08 | jfg | i'd like to work on adding MGCP client feature to asterisk, i read asterisk code practices, and a part of MGCP RFC, can someone help me now to really start? |
| 16:55.15 | *** join/#asterisk-dev fakhir (n=Fakhir@unaffiliated/fakhir) |
| 17:10.23 | russellb | Corydon76-vcch: yep ... i'm spent on that thread ... i'm not really interested in continuing |
| 17:10.56 | Qwell | jsmith: ping? |
| 17:11.21 | jsmith | Qwell: It's gotta be really fast... teaching another module in 2 minutes |
| 17:11.45 | Qwell | jsmith: queue pause/unpause from CLI - I could've sworn there was either a bug for that open, or it was already possible |
| 17:12.04 | Qwell | maybe not though... |
| 17:12.06 | jsmith | Qwell: I thought so too, but I couldn't find it last week when I went looking |
| 17:15.28 | putnopvut | Qwell: Want me to add that? |
| 17:15.40 | putnopvut | I don't think it would be particularly difficult. |
| 17:15.44 | Qwell | putnopvut: it's on his wishlist |
| 17:15.54 | Qwell | there's a list on the wiki |
| 17:16.00 | putnopvut | I'll have to see that. |
| 17:16.07 | putnopvut | Anyone know why that wasn't already implemented? |
| 17:16.25 | Qwell | no idea, but it seems quite useful |
| 17:16.43 | Qwell | it kinda goes back to one of the discussions we had at astridevcon... |
| 17:17.10 | Qwell | basically, anything that is an app or a cli command or a manager action...they should be the same, and all should exist if one does |
| 17:17.13 | Qwell | ideally, of course |
| 17:17.33 | file | yup, jsmith and I talked about that at dinner two nights ago |
| 17:17.37 | Qwell | well, maybe not apps...not all of them anyhow |
| 17:17.45 | putnopvut | Off the top of my head I'm not sure that there's a manager command for pausing/unpausing queue members. |
| 17:17.48 | putnopvut | Let me check. |
| 17:17.49 | Qwell | clearly you aren't going to run app_queue from the CLI |
| 17:17.57 | Qwell | or, rather, Queue() |
| 17:18.20 | Qwell | there is a manager command for it |
| 17:18.52 | putnopvut | Yep, you're right. |
| 17:19.09 | putnopvut | Adding the CLI command would be trivial, I think. |
| 17:19.13 | Qwell | likely |
| 17:19.22 | putnopvut | I'll add it in trunk, and knock an item off the list. |
| 17:19.33 | Qwell | hell, isn't there already an add/remove queuemember CLI command? |
| 17:19.36 | blitzrage | nice! just make sure you set it as strikethrough |
| 17:19.40 | Juggie | blitzrage! |
| 17:19.43 | blitzrage | Juggie! |
| 17:19.45 | putnopvut | Qwell: yes, there is. |
| 17:19.46 | Qwell | blitzrage: really? because I just deleted one |
| 17:19.51 | Qwell | putnopvut: yeah, trivial then, heh |
| 17:20.08 | blitzrage | Qwell: oh... you should keep them on there so people know what was added from the list |
| 17:20.12 | Qwell | k |
| 17:20.32 | blitzrage | at least that's what I like to do -- then you can see that things from the list were actually added, and such |
| 17:20.39 | Qwell | how do you strike through on a wiki? heh |
| 17:20.44 | blitzrage | great question :) |
| 17:22.56 | Qwell | --(text)--, apparently |
| 17:23.07 | blitzrage | interesting |
| 17:23.20 | Qwell | except it isn't working on ours |
| 17:25.44 | blitzrage | of course it isn't |
| 17:26.31 | elixer | you guys talking about voip-info? |
| 17:26.43 | blitzrage | no, the internal Digium wiki |
| 17:26.49 | elixer | ahh |
| 17:28.33 | Juggie | russellb: ping |
| 17:28.55 | russellb | pong |
| 17:28.58 | putnopvut | Qwell: You could (at least temporarily) put a "done" message in a different color next to the item. |
| 17:29.03 | Juggie | sorry for nic posting stupid bugs :) |
| 17:29.04 | blitzrage | indeed |
| 17:29.25 | Qwell | putnopvut: I cheated |
| 17:29.49 | russellb | Juggie: oh, that was that nic? |
| 17:29.53 | russellb | Juggie: yeah, wtf? |
| 17:30.11 | Juggie | yeah i du dunno, he said he tried sangoma first and thought to try there, dunno why. |
| 17:30.19 | Juggie | he has another call/ticket into sangoma now. |
| 17:30.31 | russellb | shouldn't be using that stuff in the first place :-p |
| 17:30.45 | Juggie | he never asked me, i'd imagine sangoma's on card bridgeing is breaking it. |
| 17:30.48 | russellb | but even if you do, certainly don't come to us with problems :) |
| 17:31.13 | Juggie | i dunno, he didnt clear it w/ me :) |
| 17:31.24 | Juggie | speaking of which did you have time to look @ that fixed agi patch? |
| 17:31.25 | *** join/#asterisk-dev hermuli (n=Eladamri@xdsl-83-145-207-63.nebulazone.fi) |
| 17:31.31 | jmls | russellb: sorry, but I just can't help it ;) |
| 17:31.44 | russellb | jmls: heh, it's all good |
| 17:31.53 | russellb | Juggie: i will at some point today |
| 17:32.05 | Juggie | russellb, thanks, appreciate it. |
| 17:32.17 | russellb | and definitely before a release makes it out next week :) |
| 17:32.20 | Juggie | i was thinking even say 'dead' could be moved into the agi struct. |
| 17:32.29 | jmls | russellb: maybe it's good for you. but not for me :( |
| 17:32.32 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 17:32.36 | Juggie | but thats just cleanup and not necessairy |
| 17:32.36 | russellb | jmls: sorry :( |
| 17:32.54 | Juggie | russellb, it would be nice to know why fgets/feof etc dont work as expected though |
| 17:33.05 | Juggie | i totally expected feof to work fine on STDIN |
| 17:34.03 | russellb | me too |
| 17:34.14 | Juggie | it doesnt appear to though |
| 17:34.20 | *** part/#asterisk-dev jtexter3 (n=jtexter3@12.159.220.114) |
| 17:34.22 | Juggie | its what caused that bug in the last commit |
| 17:34.25 | mvanbaak | heya all |
| 17:35.49 | Simon-- | hrmm..so the member state for IAX2 dyamic queue members seems to be following any single IAX2 queue call, making all agents "in use" when only one is on a call, or all ringing when only one is ringing |
| 17:38.39 | russellb | how are the members defined? |
| 17:38.50 | russellb | are they seperate peer entries, or are they extensions at the same peer? |
| 17:40.13 | mvanbaak | Qwell: I agree on 6353 comment |
| 17:40.25 | mvanbaak | the courtesy should be left alone |
| 17:42.22 | *** part/#asterisk-dev deeperror (n=deeperro@69-215-202-195.ded.ameritech.net) |
| 17:44.21 | Simon-- | russellb: Members: IAX2/austin/112@remote-queue (dynamic) (In use) has taken 1 calls (last was 27 secs ago) IAX2/austin/108@remote-queue (dynamic) (In use) has taken 1 calls (last was 6575 secs ago) |
| 17:44.42 | russellb | ah, ok |
| 17:44.56 | jsmith-busy | One peer to rule them all, and in the darkness bind them? |
| 17:44.57 | russellb | so, the reason it is doing that is because the state is actually the state of "austin" ... |
| 17:45.19 | mvanbaak | lol jsmith-busy |
| 17:45.26 | file | jsmith-busy: bind them to 192.168.2.1 port 4569? |
| 17:45.37 | Simon-- | russellb: I don't think it used to work that way :) or at least it didn't work at all, perhaps.. |
| 17:45.47 | jsmith-busy | file: No, but that reminds me of another couple of items for my wish-list |
| 17:45.50 | russellb | Simon--: i suspect to make it work, you'll have to have IAX2/austin-112 and IAX2/austin-108 |
| 17:46.11 | jsmith-busy | Simon--: It *might* work if you instead did Local/108@some-context, which then dialed over IAX2 |
| 17:46.15 | russellb | heh, perhaps .... i'm just giving you my initial guess |
| 17:46.50 | Qwell | Is there anything that can't be done with chan_local? |
| 17:47.14 | Simon-- | jsmith-busy: there are advantages to that too, but yes.. err.. I think Local inuse detection is in trunk? |
| 17:47.24 | ManxPower | Qwell: I don't think so. It cured cancer in one small study by Sweden. |
| 17:47.26 | Simon-- | not in 1.4 though |
| 17:48.30 | mvanbaak | Qwell: it wont make coffee |
| 17:48.32 | jsmith-busy | Qwell: Don't get me started :-) |
| 17:49.00 | file | uh oh, jsmith is helping someone! |
| 17:49.49 | Simon-- | yeah, so chan_local in trunk just keeps a local list of "locals" in use |
| 17:50.19 | *** join/#asterisk-dev denon (n=denon@208.122.43.201) |
| 17:50.33 | Simon-- | 1.4 just returns unknown or invalid, so switching to Local/ dyamic queue members means it'll dial people already on calls :) |
| 17:50.47 | *** join/#asterisk-dev yannj_fr (n=yannj_fr@vpn.intelunix.fr) |
| 17:51.07 | Simon-- | is there a reason why the device state stops before the second / ? |
| 17:51.34 | yannj_fr | Any comment about my feature proposal [10689] |
| 17:55.15 | blitzrage | yannj_fr: does it contain a patch to implement it? |
| 17:55.32 | blitzrage | yay it does! :) |
| 17:56.24 | yannj_fr | app_callback |
| 17:56.27 | blitzrage | indeed |
| 17:56.57 | Qwell | I don't get it - how does it schedule the callback? or is it just done right away? |
| 17:57.20 | yannj_fr | using pbx_outgoing_exten |
| 17:57.48 | Qwell | Can you give an example usage? |
| 17:57.52 | jsmith-busy | Qwell: A call file! |
| 17:57.55 | yannj_fr | it store the callee extension and when the callee hangup, set up a callback |
| 17:58.01 | Qwell | jsmith-busy: well, that's what I'm getting at |
| 17:58.21 | yannj_fr | I used to do it by scripting with .call file |
| 17:58.29 | Qwell | but then how do you initiate the callback? |
| 17:59.01 | russellb | in other words, when and how do you execute the Callback() application? |
| 17:59.01 | yannj_fr | buy using pbx_outgoing_exten with the caller as src, and the callee as exten |
| 17:59.09 | yannj_fr | ok |
| 17:59.09 | Qwell | russellb: right |
| 17:59.18 | yannj_fr | on s-BUSY for setCallback |
| 17:59.26 | yannj_fr | if you use stdexten macro |
| 17:59.45 | yannj_fr | and on H exten for GetCallback |
| 17:59.51 | Qwell | why h? |
| 17:59.59 | Qwell | why not just retry the dial? |
| 18:00.17 | russellb | i'm still not convinced it provides enough over generating a simple call file |
| 18:00.21 | Simon-- | hrm. I think this is fundamentally broken. SIP channels are "SIP/ext-1234", so why would channel names always truncate to "IAX2/foo-1234"? they should be "IAX2/foo/123-1234" instead, no? |
| 18:00.44 | Qwell | Simon--: you won't see SIP/ext/5551212-1234 |
| 18:00.49 | file | SIP provides it's own devicestate |
| 18:01.04 | file | and ohhhh... |
| 18:01.08 | file | yeah, using - is bad |
| 18:01.12 | file | there's a bug open about it |
| 18:01.26 | Qwell | file: I think he means 1234==uniqueid |
| 18:01.36 | yannj_fr | russellb : I am ok that it looks like , because I replace my macro by this, but it is also more simpler to configure and integrate |
| 18:01.59 | Simon-- | Qwell: ok, I get it, each SIP client is registered separately, and we're registering an entire remote office behind one IAX client.. |
| 18:02.08 | Qwell | right...I think |
| 18:02.19 | russellb | i have to run to lunch ... bbl ... |
| 18:02.24 | Simon-- | :) |
| 18:02.33 | Qwell | like, you can dial SIP/bob/5551212, and you can dial IAX2/bob/5551212, both will remove the dialed number |
| 18:02.51 | file | the peer is technically in use |
| 18:02.53 | Qwell | (where "number" needn't actually be a number, of course) |
| 18:03.01 | Simon-- | I suppose it needs to be able to check "SIP/bob" for inuse, not "SIP/bob/5551212" for inuse |
| 18:03.42 | Simon-- | damn. that's a hard one :) |
| 18:05.44 | Simon-- | our problem from an agent perspective is that SIP/123 is a local agent, IAX2/austin/123 is a remote agent. we could set up an IAX2/austin123 (for every agent), but that would be unmaintainable |
| 18:05.55 | file | chan_local with group counting |
| 18:07.50 | Qwell | yeah, chan_local rocks for that stuff |
| 18:07.55 | Simon-- | this is kind of a "bug" for me right now because earlier 1.4 releases did not support IAX2 inuse, so we were doing it via a remote dialplan check. iax2 device state checking was added somewhere 1.4.10-ish, and then this broke :) |
| 18:08.08 | Qwell | I think we have an example of that in the queues-with-callback-members doc |
| 18:08.15 | Simon-- | k, will dig |
| 18:08.46 | file | and if you were really cheeky and were doing queues across multiple boxes you could use DUNDi in that chan_local to check the status of the agents across a cluster... but others could talk more about that |
| 18:08.51 | Qwell | nope, guess we don't |
| 18:09.16 | Qwell | oh wait, it does talk about it briefly |
| 18:10.02 | putnopvut | Stupid DeMorgan's Law *grumble grumble* |
| 18:11.23 | Simon-- | oh, and there's actually a bug with it too -- the hangup of the channel isn't clearing the queue state of any IAX2/austin, even though it sets them all to inuse when the call is first connected |
| 18:11.55 | file | even if they are all hungup? |
| 18:12.42 | Simon-- | is it matching calls from the channel list possibly not initiated from the queue? |
| 18:12.55 | Simon-- | if so, no, they are not all hung up yet :) |
| 18:13.49 | file | that is what it does... it looks at all the channels up |
| 18:17.04 | Simon-- | ok, so the bug is simply that it is not initializing them to inuse if IAX2/austin*anything is inuse when it dynamically adds them |
| 18:17.09 | Simon-- | that should be easy to fix |
| 18:27.53 | Simon-- | ok, so ast_device_state("IAX2/austin/123") is returning false, but the devicestate changethread which app_queue registers is throwing away the second-slash part, causing it to then apply future changes to the members |
| 18:28.13 | *** join/#asterisk-dev anonymouz666 (n=anonymou@189.25.190.36) |
| 18:28.57 | Simon-- | either ast_device_state() or its caller should throw away the second slash part, or everything should be changed to track both levels :) |
| 18:30.36 | Simon-- | (member state is being initialized in create_queue_member(), but the call to ast_device_state() is passing the whole interface string) |
| 18:30.37 | *** join/#asterisk-dev d3wayne (n=deeewayn@c-71-228-186-75.hsd1.al.comcast.net) |
| 18:30.37 | *** mode/#asterisk-dev [+o d3wayne] by ChanServ |
| 18:33.23 | jmls | putnopvut: pig |
| 18:33.28 | jmls | lol |
| 18:33.28 | jmls | ping |
| 18:35.07 | Corydon76-vcch | Dunno, that's a compliment around here... |
| 18:35.11 | Corydon76-vcch | Oink |
| 18:35.48 | Corydon76-vcch | Later, I'm going out of range for a bit |
| 18:41.45 | *** join/#asterisk-dev bweschke (n=BJ_Wesch@c-24-0-209-5.hsd1.nj.comcast.net) |
| 18:41.45 | *** mode/#asterisk-dev [+o bweschke] by ChanServ |
| 18:43.46 | file | bweschke: !!! |
| 18:43.55 | bweschke | file: !!! |
| 18:44.00 | bweschke | what's up ? |
| 18:44.05 | file | oh nothing... |
| 18:44.09 | bweschke | Qwell: easy now... :-p |
| 18:44.09 | file | jsmith is just teaching agents/queues |
| 18:44.25 | Qwell | pfft, I was being nice |
| 18:44.29 | Qwell | I almost trolled on him hard. |
| 18:44.30 | bweschke | lol |
| 18:44.33 | bweschke | indeed you were |
| 18:44.42 | Qwell | I was mad yesterday before I saw his email |
| 18:44.45 | blitzrage | bweschke: !!! |
| 18:44.47 | bweschke | jsmith is the best |
| 18:44.51 | Qwell | just because he's an idiot who can't read notes |
| 18:44.56 | Qwell | (twice..) |
| 18:44.58 | bweschke | blitzrage: !!! |
| 18:45.09 | blitzrage | file: are you at the Montreal training class? |
| 18:45.16 | file | blitzrage: why yes, yes I am |
| 18:45.23 | blitzrage | so many things make sense now |
| 18:45.36 | bweschke | how's everyone feel about that branch for the event manager memory leak ? it seems like a valid fix to me |
| 18:46.42 | file | I think he's said the word roundrobin 5 times in the last minute... |
| 18:47.11 | file | bweschke: ooh, I just ordered one too |
| 18:48.38 | bweschke | putnopvut: ping |
| 18:49.11 | Simon-- | good lunch? ;) |
| 18:50.02 | file | russellb: eep |
| 18:50.24 | bweschke | Simon--: if you call Yuengling(s) lunch, yes ... it was great, thx |
| 18:51.20 | davevg-btwtech | bj codes better with a few in him |
| 18:51.52 | bweschke | davevg-btwtech: u should know! lol |
| 18:52.40 | *** join/#asterisk-dev MaXpEr (n=maxper@host32-218-dynamic.17-87-r.retail.telecomitalia.it) |
| 18:53.00 | bweschke | russellb: has the discussion of astobj2 performance vs. hashtabs gone out of thread ? I hadn't seen anything new on that |
| 18:53.17 | MaXpEr | Hi! |
| 18:53.25 | MaXpEr | I have some timeout problems and would like to understand better the ast_safe_sleep function... |
| 18:53.34 | MaXpEr | What does this function exactly do? |
| 18:53.38 | bweschke | it sleeps |
| 18:53.50 | MaXpEr | chan_mgcp seems not to sleep properly... |
| 18:53.59 | bweschke | and exits if something happens to the channel you're asking it to watch |
| 18:54.02 | MaXpEr | That should sleep (by default) 16 seconds, but it doesn't... |
| 18:54.15 | russellb | bweschke: no, it hasn't |
| 18:54.16 | MaXpEr | It sleeps, on my PC, just 5 seconds... |
| 18:54.30 | russellb | bweschke: i made a small change that increased the performace by 75% in the test app ... |
| 18:54.54 | bweschke | russellb: that's pretty huge - given the numbers being discussed before |
| 18:55.04 | russellb | yeah ... |
| 18:55.19 | russellb | i still think you can't exactly compare the two |
| 18:56.12 | bweschke | no.. it's kinda apples to oranges .. I agree |
| 18:56.30 | bweschke | but it was such a huge diff .. I was quite surprised |
| 18:57.31 | russellb | me too |
| 18:57.34 | bweschke | I mean ... with the latest improvement - would we be hurting or helping performance by implementing channel_refcounts - and is the hurt worth it to protect us from refcount troubles |
| 18:57.35 | bweschke | ? |
| 18:57.54 | russellb | i've tried some other tings to improve it, but haven't seen tangible results |
| 18:58.07 | russellb | the channel refcount thing will only help performance |
| 18:58.29 | russellb | well ... in the cases where the performance was most affected by the choice of container ... it will help a lot |
| 18:59.04 | russellb | some simple operations will be a little more expensive, but not by much |
| 18:59.16 | russellb | i'll document all of it when i get to a good point |
| 18:59.26 | russellb | explaining what is different, what is improved, etc |
| 19:00.07 | bweschke | really? I haven't looked too close at the changes to be honest .. ya - I guess if you had to traverse a whole linked list, for sure, that'd be faster |
| 19:00.18 | russellb | yeah, so let me provide an example ... |
| 19:00.22 | MaXpEr | is there anyone that may help me with chan_mgcp? |
| 19:00.43 | russellb | there is a family of functions for traversing the list of active channels (channel_walk_locked()) |
| 19:00.53 | bweschke | right |
| 19:00.54 | russellb | the core of that is _very_ expensive |
| 19:01.00 | bweschke | I bet those would be ALOT faster |
| 19:01.03 | bweschke | good pt |
| 19:01.11 | russellb | because for every time you call it to get the next channel, you have to traverse the *entire* list to make sure the last channel is still in the list |
| 19:01.17 | bweschke | yep |
| 19:01.31 | bweschke | got a little familiar with those when I was looking at the older spy stuff |
| 19:01.43 | russellb | heh |
| 19:02.05 | russellb | but astobj2 provides a more efficient way of handling it |
| 19:02.26 | russellb | where in the worst case, you are only having to traverse the current hash bucket |
| 19:02.30 | russellb | not the entire container |
| 19:02.35 | russellb | for every time you want the next channel |
| 19:02.39 | bweschke | gotcha |
| 19:02.40 | bweschke | ya |
| 19:02.49 | bweschke | I can definitely see how that would improve performance |
| 19:02.56 | russellb | there are also the ast_get_channel() functions |
| 19:03.05 | russellb | which instead of traversing the list, is a hash table lookup |
| 19:03.25 | russellb | so, O(n) to O(1) in general .. |
| 19:03.32 | bweschke | I dunno why but I was thinking you were going to start messing with the frame queue stuff or something like that |
| 19:03.55 | russellb | not at the moment, no |
| 19:04.08 | *** part/#asterisk-dev MaXpEr (n=maxper@host32-218-dynamic.17-87-r.retail.telecomitalia.it) |
| 19:04.16 | russellb | i'm just trying to fix places where it's possible for channels to disappear ... |
| 19:04.16 | bweschke | that's where I was thinking there might have been implications from the performance discussion |
| 19:04.21 | bweschke | gotcha |
| 19:04.29 | russellb | and improve performance while i'm at it :) |
| 19:04.46 | russellb | currently, the channel lock is (ab)used to try to make sure channels dont' disappear |
| 19:04.58 | bweschke | ya |
| 19:05.08 | russellb | but you can't use a lock inside the object to protect the object from disappearing ... |
| 19:05.14 | bweschke | would rw vs. r locks only help that? |
| 19:05.16 | russellb | it just doesn't work :) |
| 19:05.39 | russellb | not really, no |
| 19:05.53 | russellb | you can't have any kind of lock inside the object protect the object from disappearing |
| 19:05.59 | bweschke | k |
| 19:06.07 | putnopvut | bweschke: pong |
| 19:06.09 | Qwell | so, denub me.. when would astobj2 be useful over current linked lists? |
| 19:06.20 | russellb | Qwell: that's a loaded question, kind sir :) |
| 19:06.24 | Qwell | ;) |
| 19:06.30 | Qwell | well, what types of lists, I guess? |
| 19:06.32 | bweschke | Qwell: from what I understand, when your code has to traverse the whole list often, you're definitely going to get a boost |
| 19:06.46 | Qwell | or, perhaps the better question would be when wouldn't it be useful? |
| 19:07.04 | bweschke | additionally - if you're having trouble keeping track of refcounts to objects in your current list, astobj2 makes your life ALOT easier |
| 19:07.28 | bweschke | because it kinda just handles all that provided you give it refs and unrefs where appropriate before use |
| 19:07.34 | russellb | astobj2 as it stands today can not be used where the order of objects in a container matters |
| 19:07.34 | bweschke | and after use |
| 19:07.47 | bweschke | ya - that's a definite 'con' in some cases |
| 19:07.55 | russellb | that is a general statement for hash tables in general ... |
| 19:08.02 | Qwell | right |
| 19:08.03 | bweschke | though .. the behavior isn't something 'new' for us .. realtime exhibits the same behavior for the most part, no ? |
| 19:08.23 | russellb | Qwell: if you'd like to see an example of where reference tracking is important, i can show you what i did in chan_iax2 |
| 19:08.35 | russellb | bweschke: true |
| 19:08.46 | Qwell | russellb: well, I see where it's important |
| 19:08.53 | russellb | bweschke: but it couldn't replace anything where the list is used as a FIFO |
| 19:09.05 | bweschke | punopvut: how good are you feeling about the refcount app_queue implementation in 1.4 - good for prod yet, or still some hesitation ? |
| 19:09.22 | putnopvut | hesitation. |
| 19:09.35 | putnopvut | It seems like every day or so, there's some sort of tweak that needs to be done. |
| 19:10.06 | bweschke | putnopvut: thank u sir.. I was getting ready to start beating it up - but the client I was going to put on it just quieted down from other 'issues' and I didn't want to start hearing from them again |
| 19:10.36 | putnopvut | We've cleared up some pretty big memory issues since the initial commit, so you wouldn't most likely see those. |
| 19:10.41 | russellb | Qwell: if you have any questions, feel free to stop by... |
| 19:10.53 | putnopvut | But it's still possible there are some edge cases not taken care of and maybe a small memory leak here or there. |
| 19:12.16 | russellb | i gave a pretty thorough look over |
| 19:12.17 | bweschke | is there memory debugging options with astobj2 similar to the malloc debugging that already exists? |
| 19:12.37 | russellb | the malloc debug stuff will show you astobj2 objects, too |
| 19:12.55 | russellb | also, you can define AO2_DEBUG and get the astobj2 specific stats on how many objects have been allocated and so forth |
| 19:14.08 | putnopvut | ...is that something new, or am I just dumb? |
| 19:14.35 | russellb | the stats were already there |
| 19:14.42 | russellb | i made it so they are _only_ there with that defined just now |
| 19:15.46 | bweschke | I'm wondering if there was a somewhat easy way to automate unit test scripts through app_queue and then use CLI output from memory allocation to dbl check ourselves against memleaks |
| 19:20.46 | jmls | putnopvut: my realtime agents are still showing as "no calls yet" even when they've had 100+ calls |
| 19:21.01 | jmls | 1.4 r82286 |
| 19:21.27 | jmls | but I suspect I know why - and blitzrage may be interested as well |
| 19:21.31 | putnopvut | jmls: thanks for letting me know. |
| 19:21.41 | putnopvut | jmls: what do you think? |
| 19:21.52 | jmls | I use a view on my agents |
| 19:22.09 | jmls | the view is only for agents with paused = 0 |
| 19:22.12 | blitzrage | jmls: at this point I'm interested, but won't be able to dedicate any time to testing anything :D |
| 19:22.28 | putnopvut | jmls: I see what you're saying. |
| 19:22.41 | jmls | so, if an agent goes paused (on a call etc) then they are no longer in the "table" |
| 19:22.52 | jmls | and therefore would probably be removed from the list |
| 19:23.05 | blitzrage | jmls: right -- that makes sense |
| 19:23.08 | blitzrage | that's how I do it |
| 19:23.13 | *** join/#asterisk-dev JimVanM (n=jimvanm@bas1-toronto63-1096579178.dsl.bell.ca) |
| 19:23.19 | blitzrage | JimVanM: on snap! |
| 19:23.24 | jmls | they go ready again, paused = 0, they appear in the table again and get "added" back into the list, but as a new member |
| 19:23.24 | blitzrage | s/on/oh/ |
| 19:23.27 | putnopvut | Well, new support for a "paused" field for realtime members has been added. |
| 19:23.37 | jmls | mmmm. |
| 19:23.49 | putnopvut | If you use a column called "uniqueid" as a primary key for your queue members. |
| 19:24.07 | putnopvut | Then when a member is paused, it will change the "paused" field for that member in realtime to 1. |
| 19:24.12 | jmls | is that a uniqueid within the queue, or as a whole ? |
| 19:24.19 | putnopvut | as a whole |
| 19:24.53 | putnopvut | I went with the uniqueid method of identifying a member since for realtime writes, Asterisk has no facilities for handling multiple column primary keys. |
| 19:25.00 | jmls | surely if a member is a member of more than one queu that would cause a problem |
| 19:25.19 | jmls | because member 1 of queue 1 has id of x |
| 19:25.27 | jmls | member 1 of queue 2 has id of y |
| 19:25.34 | putnopvut | That's exactly the point. |
| 19:25.43 | jmls | member 1 goes busy on queue 1, he need to go busy on queue 2 |
| 19:25.57 | putnopvut | Busy status isn't reflected in realtime. |
| 19:26.07 | putnopvut | There's a global interface list in app_queue that handles that. |
| 19:26.08 | jmls | I meant paused, sorry |
| 19:26.15 | putnopvut | jmls, not necessarily. |
| 19:26.56 | jmls | for us, if an agent is on a call, he *must* not receive any other queue calls |
| 19:27.03 | putnopvut | This is actually worth discussing though, because one person might assume to pause a member means to pause across all queues. |
| 19:27.14 | jmls | it does for us ;) |
| 19:27.27 | putnopvut | Hmmm... |
| 19:27.33 | jmls | that's why we have to use a view - the app sets the paused flag across all queues |
| 19:27.45 | jmls | and therefore automatically "removes" the member from all queues |
| 19:28.01 | putnopvut | Okay, well, I can modify the pausing function to make it so it CAN pause on all queues if configured as such. |
| 19:28.33 | jmls | I honestly expected that to be default behaviour. No wonder I was having problems ... |
| 19:28.36 | putnopvut | I could make a new general option for queues.conf to pause across all queues yes or no. |
| 19:28.56 | jmls | would get my vote. |
| 19:29.02 | putnopvut | There's a similar bug on the tracker right now regarding wrapup times as well. |
| 19:29.27 | putnopvut | that'll be difficult. |
| 19:29.44 | putnopvut | It's hard for me to see this in any way other than new feature. |
| 19:30.07 | jmls | urm, when a member is paused, it is not paused across all queues ? |
| 19:30.33 | jmls | as I mentioned, I simply expected that to be the case, and thought that it was a bug when it wasn't .. |
| 19:30.39 | jmls | :) |
| 19:30.45 | putnopvut | Nope, it's specific to whatever queue is specified. There are ways with the Manager (and I believe dialplan) to specify all queues. |
| 19:31.32 | jmls | hmm. So I would have to use local realtime channels to pause the queuemember directly from the dialplan ? |
| 19:31.41 | putnopvut | No, actually. |
| 19:31.54 | putnopvut | If you pause from the dialplan but don't specify a queue, then it will pause on all queues. |
| 19:33.05 | jmls | yeah, that's what I meant - but you have to be in a local channel in order to specify the pause on the member that has just got the call |
| 19:33.30 | jmls | otherwise it just rings the phone ... |
| 19:33.48 | putnopvut | Why is that? |
| 19:34.05 | jmls | how else would you know who's got the call, so that you can pause the member ? |
| 19:34.15 | putnopvut | you can use setinterfacevar in queues.conf |
| 19:34.23 | putnopvut | Which will set the variable MEMBERINTERFACE |
| 19:34.28 | jmls | yeah ... |
| 19:34.31 | putnopvut | Which is the interface of the member who took the call. |
| 19:34.34 | jmls | yeah ... |
| 19:35.17 | jmls | but if the interface is not local, then the dialplan is finished before you get the chance to run PauseQueueMember |
| 19:35.44 | jmls | that variable is only available after the call is ended, is it not ? |
| 19:36.02 | putnopvut | It gets set when the member answers the phone. |
| 19:36.06 | jmls | yes. |
| 19:36.17 | putnopvut | And if you want to pause on the basis that a member does not answer the phone, use autopause. |
| 19:36.31 | jmls | by then it's too late - how would you pause the member on all queues ? |
| 19:36.53 | putnopvut | You've got me there. |
| 19:37.15 | jmls | if you use local, then you can do it in the dialplan _before_ you call the phone |
| 19:37.31 | jmls | if the dial fails, unpause them |
| 19:38.00 | putnopvut | True. |
| 19:38.10 | jmls | we already use chan_local for the members, as we can then use Jabber to send a popup message to the agent before the call comes through |
| 19:39.15 | jmls | I'll give the pause dialplan option a try. Is uniqueID a number or char field ? |
| 19:39.15 | putnopvut | So it sounds like I need to add an option to queues.conf to pause across all queues when a member is paused. |
| 19:39.22 | putnopvut | number |
| 19:39.38 | putnopvut | the suggestiion is a monotonically increasing integer. |
| 19:39.54 | jmls | that would be the best solution going forward, as it means that you don't have to use the dialplan to pause |
| 19:40.10 | putnopvut | Right. Unfortunately, it'll only go in trunk. |
| 19:40.29 | jmls | i thought that I was being so elequoent as well ;) |
| 19:40.39 | putnopvut | Unless someone higher up approves of it going into 1.4 |
| 19:41.00 | jmls | there's higher people than you ? |
| 19:42.36 | jmls | ooooo, I've found a place where is says +101 works .. |
| 19:42.52 | jmls | I thought that jumping was removed from 1.4 ? |
| 19:49.25 | russellb | deprecated |
| 19:49.28 | russellb | and disabled by default |
| 19:49.39 | russellb | enabled in extensions.conf if you want it |
| 19:50.17 | jmls | oh, so we should probably remove all reference to in the the application documentation ? |
| 19:50.36 | jmls | PauseQueuemember and UnPauseQueueMember mention it |
| 19:50.54 | jmls | Show Application PauseQueueMember |
| 20:03.34 | bweschke | sorry I'm coming late to this discussion, but I can think of a reason why you wouldn't want this behavior |
| 20:03.50 | putnopvut | Which? Pausing across all queues? |
| 20:03.55 | bweschke | yes |
| 20:04.30 | bweschke | times when you're using the queue in conjunction with a dialer application and you're using pause to hold more calls from that queue only from going to the queue member while we're waiting for them to disposition their present call |
| 20:05.13 | putnopvut | This is why I think it should be configurable whether they're paused across all queues or not. |
| 20:05.23 | bweschke | it's going to have to be |
| 20:05.32 | bweschke | removing functionality is definitely not a wise thing |
| 20:17.16 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 20:25.18 | *** join/#asterisk-dev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 20:34.15 | Deeewayne | MuffinMan: dance |
| 20:34.54 | *** join/#asterisk-dev jfg (n=jfg@dyn-83-152-249-84.ppp.tiscali.fr) |
| 20:35.06 | russellb | <("<) (>")> |
| 20:35.47 | Deeewayne | oh, It's the baker that dances... |
| 20:36.59 | bweschke | russellb: u guys fix any memleaks recently in channel.c - oh say... maybe since r64426 ? |
| 20:37.06 | bweschke | branch 1.4 of course .. |
| 20:37.10 | russellb | in the past 20k revisions? probably |
| 20:37.14 | Qwell | HMM |
| 20:37.20 | Qwell | I may add ^W support to the asterisk CLI |
| 20:37.26 | Qwell | proper ^W support, that is |
| 20:37.28 | putnopvut | NO DON'T!! |
| 20:37.30 | russellb | Qwell: control-D, EOF? |
| 20:38.12 | putnopvut | Or do if you want. |
| 20:38.12 | putnopvut | It's no biggie. |
| 20:38.12 | Qwell | huh? |
| 20:38.12 | russellb | Qwell: if you can figure it out, that would rock ... |
| 20:38.12 | Qwell | already did :P |
| 20:38.12 | *** join/#asterisk-dev anonymouz666 (n=anonymou@189.25.132.243) |
| 20:38.13 | russellb | ORLY |
| 20:38.13 | Qwell | mmhmm, simple |
| 20:38.13 | bweschke | tons and tons and tons of 1 and 5 byte allocs on channel.c from line 803 and line 804 |
| 20:38.13 | Qwell | el_set(el, EL_BIND, "^W", "ed-delete-prev-word", NULL); |
| 20:38.21 | Qwell | there are a crapton of options we can use... main/editline/help.c |
| 20:38.36 | bweschke | <PROTECTED> |
| 20:38.56 | Qwell | I don't know who chose ^D for redraw screen |
| 20:39.03 | Qwell | I think there is another shellism for that |
| 20:41.58 | Qwell | w00t, it works.. |
| 20:44.32 | blitzrage | Qwell: w00t! |
| 20:45.21 | elixer | ^L is redraw screen |
| 20:45.30 | Qwell | elixer: should be |
| 20:45.36 | Qwell | it's ^D in asterisk though... |
| 20:45.44 | elixer | ^D should "quit" |
| 20:45.45 | Qwell | erm, no, ^L is clear |
| 20:45.46 | elixer | imho |
| 20:46.07 | elixer | ah yes |
| 20:46.32 | Qwell | redisplay should be ^R |
| 20:46.46 | Qwell | erm, no, heh, that's a viism |
| 20:47.04 | blitzrage | ummm... latest 1.4 for some reason kicks me out of a remote console screen when it gets a bunch of CLI output... |
| 20:47.15 | blitzrage | I think Asterisk is restarting or something.... |
| 20:47.34 | elixer | can you make ^D do EOF and exit the CLI? |
| 20:47.36 | snuff-away | bweschke.. bug russ about coverity may help find many more memory issues ;) |
| 20:47.38 | elixer | purty please? |
| 20:47.58 | bweschke | coverity ? |
| 20:48.20 | snuff-away | bweschke.. scan.coverity.com |
| 20:48.40 | snuff-away | OSS can get checks done for free :) |
| 20:49.00 | bweschke | oh.. |
| 20:49.22 | bweschke | Digium mentioned at the dev conf and had a fellow come from IBM ISS who had scanned the codebase quickly |
| 20:49.26 | Qwell | elixer: it does EOF, but it doesn't exit the CLI |
| 20:49.47 | elixer | Qwell: can you make ^D exit the CLI? :-) |
| 20:49.57 | elixer | Qwell: you're a big strong programmer guy |
| 20:49.59 | Qwell | no, I mean...the eof command doesn't make it exit :p |
| 20:50.01 | elixer | Qwell: the bestest |
| 20:50.07 | elixer | heh |
| 20:50.34 | elixer | Qwell: would it be possible to make EOF exit? (jesus this is a painful conversation) |
| 20:50.43 | elixer | work with me here! |
| 20:50.43 | Qwell | probably |
| 20:51.30 | jfg | can someone give me light on something about asterisk and mgcp ? |
| 20:52.09 | jfg | in fact, actually, asterisk can only be a gateway ? |
| 20:52.43 | jfg | and cannot register as an end point on a call agent ? |
| 20:53.31 | Qwell | elixer: yep, trivial |
| 20:53.47 | elixer | Qwell: god bless you |
| 20:54.13 | elixer | make sure you mention me in the commit comment :P |
| 20:54.28 | russellb | Qwell: that's what i was trying to ask about when you first brought this up :) |
| 20:54.34 | Qwell | does anybody ever use ^D? |
| 20:54.39 | Qwell | russellb: oh :p |
| 20:54.54 | russellb | i think a lot of people try to, becase a lot of other apps support that |
| 20:55.02 | russellb | like mysql, for example |
| 20:55.09 | Qwell | think I should go ahead and commit it? I added ^L, ^W, and changed ^D |
| 20:55.13 | Qwell | (to trunk) |
| 20:55.20 | russellb | yes! |
| 20:55.33 | elixer | bash maps ^D to complete-word by default as well |
| 20:55.41 | Qwell | complete-word? |
| 20:55.45 | elixer | err |
| 20:55.48 | elixer | tab completion |
| 20:55.48 | Qwell | bash maps ^D to logout :p |
| 20:55.51 | Qwell | ahh, yeah |
| 20:55.56 | Qwell | ^I |
| 20:55.59 | elixer | ahh |
| 20:56.00 | Qwell | (tab) |
| 20:56.12 | Qwell | asterisk maps both ^I and ? |
| 20:56.26 | elixer | i'm missing the pinky on my left hand, so tab is no good to me |
| 20:56.30 | Qwell | any other requests? |
| 20:56.43 | Qwell | elixer: well, CTRL-I is always there...same thing as tab |
| 20:56.48 | elixer | (i was kidding) |
| 20:56.49 | elixer | heh |
| 20:56.53 | Qwell | well, still :p |
| 20:57.23 | Qwell | I don't know why ^W was doing line deletions before... |
| 20:57.31 | Qwell | that's not a default mapping |
| 20:57.36 | elixer | as long as ^A, ^E, and ^K still work when you are done, i am happy |
| 20:57.46 | Qwell | what're those? |
| 20:57.54 | elixer | beginning of line, end of line, kill |
| 20:57.58 | Qwell | kill? |
| 20:58.00 | elixer | kill to end of line |
| 20:58.05 | Qwell | ahh |
| 20:58.11 | Qwell | what's kill to start of line? |
| 20:58.19 | elixer | good question |
| 20:58.28 | Qwell | ^U |
| 20:58.35 | elixer | that kills everything for me |
| 20:58.39 | elixer | in asterisk CLI, anyway |
| 20:58.44 | Qwell | oh |
| 20:59.25 | Qwell | it should be kill line prev |
| 20:59.43 | elixer | well then yes, i have one more request ;-) |
| 21:00.10 | elixer | then we need to add some cli commands that let you bind keys on the fly |
| 21:00.10 | Qwell | ahh, except in emacs mode, because emacs is skilly |
| 21:00.12 | Qwell | silly* |
| 21:00.16 | elixer | and a lisp interpreter |
| 21:01.26 | Corydon76-vcch | Does that translate "Hey big boyth" to "Hey big boys"? |
| 21:01.41 | elixer | ba dum du *ching* |
| 21:01.45 | elixer | try the veal! |
| 21:01.52 | jcmoore | lisp... eeish |
| 21:01.57 | Qwell | odd, ^L already clears the screen for me - even without that patch |
| 21:02.03 | elixer | yeah, same here. |
| 21:02.23 | jcmoore | naming a programming language "lisp" is like naming a baby "ugly" |
| 21:02.34 | Corydon76-vcch | That's the terminal clearing the screen |
| 21:03.07 | Corydon76-vcch | jcmoore: or "Not My Children" and "Unloved" |
| 21:03.14 | jcmoore | :-) |
| 21:03.19 | Qwell | Corydon76-vcch: I see |
| 21:03.27 | elixer | Qwell: ohhh... transpose-words... think of the limited possibilities |
| 21:03.42 | Corydon76-vcch | jcmoore: that would be a biblical reference, if you didn't catch it |
| 21:06.09 | jfg | are my questions stupid ? |
| 21:06.20 | elixer | jfg: only that one |
| 21:06.28 | elixer | :P |
| 21:06.28 | jcmoore | ah, the more literal acronym is "Lots of Irritating Silly Parentheses" |
| 21:06.51 | Qwell | ha...^D isn't possible... and that's why it redisplays |
| 21:06.55 | elixer | jfg: i'd help out, but i know nothing about mgcp and only slightly more about asterisk |
| 21:06.57 | Qwell | it works fine from -c, but not -r |
| 21:07.13 | elixer | Qwell: why is it not possible? |
| 21:07.13 | jfg | elixer: ok, thanks anyway |
| 21:07.17 | Corydon76-vcch | jfg: I think a great many people would like to forget that MGCP exists |
| 21:07.19 | Qwell | you get an EOF from logs, or something |
| 21:07.34 | Qwell | so any time there is output on the screen, you get booted, because it's calling the EOF handler I had to add |
| 21:07.34 | jfg | Corydon76-vcch: but in france all providers use MGCP |
| 21:07.53 | Corydon76-vcch | <insert France joke here> |
| 21:08.04 | elixer | They're always there when they need you... |
| 21:08.07 | Qwell | or maybe not impossible, just not trivial |
| 21:08.10 | jfg | Corydon76-vcch: so may people would like asterisk to better supports mgcp, and i'd like to implement it |
| 21:08.19 | Corydon76-vcch | jfg: go for it, then |
| 21:08.22 | Corydon76-vcch | ;-) |
| 21:08.24 | elixer | Qwell: well you've got some time to kill |
| 21:08.36 | jmls | hmm, does anyone use Amazon s3 for storage ? |
| 21:08.49 | jfg | Corydon76-vcch: yes but i need light on some points |
| 21:08.52 | elixer | Qwell: why does output affect the input to the CLI? |
| 21:08.59 | Corydon76-vcch | jfg: which are? |
| 21:09.01 | Qwell | no idea |
| 21:09.27 | Qwell | I'll pastebin the patch - you're more than willing to try it |
| 21:09.32 | elixer | k |
| 21:09.41 | Corydon76-vcch | s/willing/welcome/ ? |
| 21:09.42 | Juggie | elixer, did you find a project? |
| 21:09.46 | Qwell | yes, welcome |
| 21:09.51 | Qwell | willing too |
| 21:09.53 | *** join/#asterisk-dev bkruse (i=bkruse@nat/digium/x-79861d6cd699fa1a) |
| 21:09.57 | elixer | Juggie: not really, no. |
| 21:10.09 | elixer | Juggie: my project is to lurk here and work on my one-liners. |
| 21:10.13 | elixer | :-) |
| 21:10.17 | Juggie | i have a half written CDR project i've effectively run out of time for. |
| 21:10.19 | bkruse | Corydon76-vcch: ugh |
| 21:10.20 | bkruse | whats up |
| 21:10.24 | Juggie | well more then half, the API is done. |
| 21:10.31 | Juggie | i just need to modify the CDR backends. |
| 21:10.34 | Corydon76-vcch | bkruse: you figure out that vm bug for the gui? |
| 21:10.50 | Qwell | http://uuoc.com/1903 |
| 21:10.52 | jfg | Corydon76-vcch: i'd like to know what asterisk does with mgcp |
| 21:10.57 | Juggie | elxir, if you'd like to take a look you'd be more then welcome to |
| 21:11.11 | Juggie | the goal is to eliminate CDR loss by creating a queue for each backend. |
| 21:11.13 | Corydon76-vcch | bkruse: I know next to nothing about ajax, which is why I'm shaky on touching anything other than the bare html |
| 21:11.16 | jfg | Corydon76-vcch: i think asterisk can just be a gateway (or maybe a call agent) ? |
| 21:11.24 | jfg | and cannot register on a call agent |
| 21:11.28 | jfg | true ? |
| 21:11.31 | Qwell | elixer: try it out - maybe you can figure out why it's not working |
| 21:11.33 | Corydon76-vcch | jfg: I believe you're correct |
| 21:11.36 | Qwell | it works great if you use -c though |
| 21:11.43 | elixer | Juggie: i could take a look, sure. |
| 21:11.53 | bkruse | Corydon76-vcch: thats ok? |
| 21:12.02 | elixer | Juggie: or do you need a firm commitment? |
| 21:12.03 | jfg | ok, so what i need is to make asterisk able to connect to my provider call agent |
| 21:12.08 | bkruse | Corydon76-vcch: OH, didnt read your first comment, yes, its fixed |
| 21:12.11 | jfg | but i dunno where to add the code |
| 21:12.18 | jfg | and what code to use |
| 21:12.18 | bkruse | Its an option for each individual user now :D |
| 21:12.19 | Juggie | elixer, no, my work on this is at a standstill at the moment, too much going on for me. |
| 21:12.23 | Corydon76-vcch | bkruse: ah, good |
| 21:12.24 | jfg | chan_mgcp.c i suppose ? |
| 21:12.29 | Juggie | you can play with it and let me know if you make any progress |
| 21:12.30 | Corydon76-vcch | jfg: yes |
| 21:12.36 | Juggie | there is very little work left. |
| 21:12.41 | elixer | Juggie: ok, cool. is it in a branch? |
| 21:12.46 | bkruse | Corydon76-vcch: thanks for the initial patch that gave me the information, appreciate it :] |
| 21:12.53 | Juggie | elixer, yeah, http://svn.digium.com/view/asterisk/team/juggie/ |
| 21:13.05 | Corydon76-vcch | jfg: you may want to make it match up with the language in other parts of Asterisk, type=peer and type=user |
| 21:13.24 | Corydon76-vcch | although whether type=friend makes any sense is up to you |
| 21:13.33 | Juggie | elixer, the best backend to rewrite would be cdr_adaptive_odbc corydon's baby. |
| 21:13.42 | jfg | Corydon76-vcch: could other chan_*protocol*.c help me ? |
| 21:13.44 | Juggie | well, not rewrite, retrofit, update, whatever. |
| 21:13.47 | jfg | Corydon76-vcch: yes friend make any sense |
| 21:13.54 | elixer | Juggie: right, gotcha. |
| 21:14.00 | Corydon76-vcch | jfg: I don't know enough about the protocol to know for sure |
| 21:14.11 | Juggie | elixer, essentially, adaptive_odbc does multiple databases within the driver |
| 21:14.12 | Corydon76-vcch | bkruse: yw |
| 21:14.22 | Juggie | and i've moved that logic to the cdr core, so you just now would do multiple registrations. |
| 21:14.32 | Juggie | and the core handles the rest. |
| 21:14.47 | Juggie | it just passes you a void * to tell you which backend you have to write to |
| 21:14.59 | Juggie | (of course you pass a void * in when registering so you can make the link) |
| 21:15.52 | jfg | Corydon76-vcch: ok thanks, so the file that i'll modify is chan_mgcp.c, any other ? |
| 21:16.08 | elixer | Juggie: i've never even looked at the CDR code before. i'll take a look at the revisions to see that API changes. |
| 21:16.14 | Corydon76-vcch | jfg: dunno, does MGCP use RTP? |
| 21:16.23 | Juggie | its also intended to be non blocking so if there is a problem you just return a failure (you'll see them defined in cdr.h) and the cdr core will send the cdr to the backend again |
| 21:16.51 | file | Corydon76-vcch: it does |
| 21:16.58 | jfg | Corydon76-vcch: i dunno, it depends if asterisk is used as a gateway or as a call agent |
| 21:17.14 | Juggie | elixer, essentially each registered backend instance gets its own worker thread. multiple threads per backend should the backend register more then once. |
| 21:17.19 | jfg | but i think it's already implemented, no ? |
| 21:17.25 | Corydon76-vcch | jfg: you probably won't need to alter main/rtp.c, but if you do, you'll want to ensure that it still works with sip |
| 21:17.37 | jfg | Corydon76-vcch: ok |
| 21:17.52 | Corydon76-vcch | jfg: as far as coordination with rtp goes, you may want to look at chan_sip.c |
| 21:18.01 | jfg | ok |
| 21:18.07 | jfg | but it's a very big file |
| 21:18.13 | Corydon76-vcch | Yes, it is |
| 21:18.33 | elixer | Juggie: ok. i'll take a look and ask questions when i have them. |
| 21:18.36 | file | you shouldn't need to muck with rtp.c |
| 21:18.38 | Juggie | elixer, if you look @ http://svn.digium.com/view/asterisk/team/juggie/NoLossCDR/cdr/cdr_csv.c?revision=78334&view=markup you can see iv'e already modified cdr_csv to follow the new model |
| 21:18.47 | jfg | file: ok |
| 21:19.04 | Juggie | adaptive_odbc is a little tricker because it supports multiple databases |
| 21:19.05 | file | just chan_mgcp |
| 21:19.15 | jfg | and hm, must i recompile every time asterisk and re install it ? |
| 21:19.39 | Juggie | elixer, essentially, where it reads its config and defines multiple databases you need to add multiple registrations to asterisk rather then one. |
| 21:19.42 | Corydon76-vcch | jfg: generally, yes |
| 21:19.53 | jfg | or can i just move chan_mgcp.so to my already running asterisk ? |
| 21:19.53 | elixer | Juggie: i understand |
| 21:20.01 | Juggie | k, let me know if you have any questions then |
| 21:20.10 | Juggie | it would be awesome to get it finished up but i've totally ran out of time. |
| 21:20.23 | Corydon76-vcch | jfg: might, but you'll still need to do either an unload/load or a restart |
| 21:20.27 | elixer | Juggie: understood. thanks for throwing it over the fence. |
| 21:20.28 | Juggie | new female in my life is quite time demanding. |
| 21:20.30 | Juggie | np. |
| 21:20.30 | jfg | Corydon76-vcch: ok |
| 21:20.46 | elixer | Juggie: well there are worse problems :-) |
| 21:20.52 | Juggie | indeed. |
| 21:21.21 | putnopvut | Hmmm, so regarding the new CLI commands to pause/unpause queue members: should I update CHANGES in the CLI section or Queues section? |
| 21:21.33 | Corydon76-vcch | CLI |
| 21:21.45 | putnopvut | kthx |
| 21:26.24 | russellb | i vote the Queues section :) |
| 21:26.30 | russellb | but it's not a big deal ... |
| 21:26.31 | Qwell | I vote the SIP section |
| 21:26.36 | russellb | as long as it is in there |
| 21:26.47 | russellb | we can reorganize stuff later if we feel like it |
| 21:28.34 | putnopvut | ...hmmm already added to the CLI section. |
| 21:34.43 | *** join/#asterisk-dev angler (i=angler@pdpc/sponsor/digium/angler) |
| 21:54.26 | dioedu | hello, few days ago, i did a question about the problem with DEVSTATE of sip channels. I was using 1.2.17 and did some tests with 1.2.18. Both had the same problem with DEVSTATE where sip client show IN USE when i had my sip client FREE. I did the downgrade to 1.2.14 and this problem doesn't happen again. |
| 21:54.56 | dioedu | Just to inform... |
| 21:57.24 | russellb | what about 1.2.25 or wahtever the latest release is? |
| 21:59.19 | bkruse | what option in sip.conf will see the contact field, particulary the name before the domain eg variablehere@iphere |
| 22:02.50 | tzafrir_laptop | 1.2.24 |
| 22:06.01 | tzafrir_laptop | I wonder what about plans for a new Zaptel release |
| 22:19.27 | jsmith-busy | tzafrir_laptop: I'd be happy for one |
| 22:19.52 | russellb | how about next week when i was planning to do asterisk, anyway? |
| 22:19.56 | russellb | i'll ask cresl1n too |
| 22:20.08 | tzafrir_laptop | sure |
| 22:20.14 | jsmith | Sounds good to me... |
| 22:20.43 | file | TE120P |
| 22:20.53 | jsmith | Yeah, that's what I meant. |
| 22:20.56 | russellb | ORLY |
| 22:21.05 | file | it wasn't in the init script |
| 22:21.14 | file | which made everyone here unhappy |
| 22:21.25 | file | (since their card did not work) |
| 22:21.28 | tzafrir_laptop | it actually was in the sysconfig file :-) |
| 22:21.40 | tzafrir_laptop | in 1.4 only, though |
| 22:22.04 | russellb | oops |
| 22:22.07 | russellb | yay file and jsmith |
| 22:24.32 | tzafrir_laptop | I also windered about updating drivers in zaptel 1.2 |
| 22:25.00 | tzafrir_laptop | but I'll wait for cresl1n with that |
| 22:25.09 | Qwell | tzafrir_laptop: as far as what? |
| 22:26.27 | tzafrir_laptop | well, m current habbit is to maintain basically the same xpp/ subdirectory in 1.2 and 1.4 |
| 22:26.37 | russellb | that's fine |
| 22:26.38 | tzafrir_laptop | Which is convinient for me |
| 22:26.51 | russellb | you can still fix things back in the 1.2 branch of zaptel |
| 22:27.13 | Qwell | we still add new drivers to 1.2, so... |
| 22:27.29 | tzafrir_laptop | ok |
| 22:28.06 | russellb | i fought it a little bit, but alas, i lost |
| 22:28.52 | russellb | figured i'd save it for things i thought were more important :) |
| 22:28.59 | tzafrir_laptop | your next chance will be when things start to actualy move in zaptel. Which may or may not break automatic svnmerge |
| 22:29.02 | russellb | "choose your battles" as they say ... |
| 22:29.10 | Qwell | russellb: like, free redbull when we move? |
| 22:29.18 | Qwell | I'll fight with you on that :) |
| 22:29.19 | russellb | ha, good call! |
| 22:29.27 | putnopvut | I don't like redbull |
| 22:29.32 | Qwell | pfft |
| 22:29.34 | putnopvut | Not by itself anyway. |
| 22:29.38 | *** join/#asterisk-dev mitcheloc (n=mitchel@adsl-68-120-230-95.dsl.irvnca.pacbell.net) |
| 22:29.41 | Qwell | fine, free vodka too |
| 22:29.42 | Qwell | :P |
| 22:29.43 | russellb | ok, free beverage of choice |
| 22:38.19 | putnopvut | I'm not sure if anyone here was paying attention to my conversation with jmls earlier, but I've decided to add a configuration option to queues.conf to specify that if a queue member is paused, do you automatically pause that member in all queues in which he's a member. |
| 22:38.44 | putnopvut | What' I'm not sure of is if this should be a global (general) setting or if it would be better to specify such an option for specific members. |
| 22:39.01 | putnopvut | Also, if anyone has an objection to such an idea, speak up. |
| 22:39.24 | *** join/#asterisk-dev tsurko (n=tsurko@78.90.100.214) |
| 22:39.50 | Qwell | well, "member" is difficult |
| 22:39.58 | blitzrage | putnopvut: if you can make it so you set it globally, but can override for a member.. that'd be cool |
| 22:39.59 | putnopvut | difficult, but doable. |
| 22:40.20 | blitzrage | if you're going to do it... might as well do it right, eh? :) |
| 22:41.23 | putnopvut | Right. I've made a plan of how to do it either way, I just want to hear feedback before I actually get coding. |
| 22:43.45 | russellb | so dude is about to commit a new channel driver, heh |
| 22:43.49 | russellb | chan_usbradio i think |
| 22:43.54 | Qwell | hmm |
| 22:44.27 | russellb | we shall see! |
| 22:44.28 | *** join/#asterisk-dev Math` (n=math@bas4-montreal19-1242357874.dsl.bell.ca) |
| 22:44.35 | Qwell | trunk only? :p |
| 22:44.59 | russellb | i assume so |
| 22:45.02 | russellb | O.O |
| 22:45.18 | *** join/#asterisk-dev yannj_fr (n=yannj_fr@vpn.intelunix.fr) |
| 22:45.20 | russellb | i was helping him with some build questions he had |
| 22:45.22 | Qwell | 1.0, 1.2, 1.4, trunk :D |
| 22:45.30 | russellb | i told him to just commit it, and i would fix the configure script part of it |
| 22:45.46 | Qwell | how come he never gets on irc? |
| 22:45.47 | russellb | so don't be surprised when it goes in and builds by default with some dependency on a library that isn't checked for :) |
| 22:45.53 | russellb | that's just not his style |
| 22:45.56 | russellb | email isn't his style either |
| 22:46.00 | russellb | he likes to just call, heh |
| 22:46.05 | Qwell | damn phone |
| 22:46.06 | Qwell | s |
| 22:46.12 | russellb | apparently he called mark s. with these questions first, heh |
| 22:46.25 | file | and then marko redirected to you? |
| 22:46.32 | russellb | indeed |
| 22:46.55 | file | oic |
| 22:47.57 | jmls | putnopvut: I agree with blitzrage: general but overrideable on an member |
| 22:48.45 | *** join/#asterisk-dev tsurko (n=tsurko@78.90.100.214) |
| 22:50.04 | putnopvut | All right. I'm going to go with the general setting which can be overridden by a member, then. |
| 22:54.30 | blitzrage | putnopvut: you should have just done it and not asked :) |
| 22:57.07 | putnopvut | Well, I hadn't thought of the idea of having a member override the general setting. I kind of had the mindset of one or the other. |
| 22:57.14 | putnopvut | I'm glad I asked. |
| 22:57.38 | blitzrage | yay! |
| 22:57.41 | blitzrage | glad I was useful :) |
| 23:13.49 | snuff-work | mm the bug about ulaw not being compliant to spec.. that was fixed he mentioned something about alaw as well.. anything happen on that? (cant remember bug no |
| 23:14.37 | file | russellb: *gasp* rude! |
| 23:14.48 | russellb | <3 |
| 23:14.52 | file | </3 |
| 23:15.10 | Qwell | snuff-away: it would've been fixed together, no doubt |
| 23:15.22 | file | I don't know where that fruit has been |
| 23:15.27 | russellb | ! |
| 23:15.42 | Qwell | sounds like innuendo |
| 23:15.56 | file | Qwell: the kind I get email about? |
| 23:16.02 | Qwell | what? |
| 23:16.12 | file | nothing... nothing.... |
| 23:16.15 | Qwell | spam? |
| 23:16.26 | Qwell | something like that |
| 23:16.29 | file | yes, something |
| 23:17.47 | Qwell | ANYWAYS |
| 23:18.07 | russellb | HI |
| 23:18.50 | russellb | .. and then to an application which decodes it, creates a local channel, and passes it back into asterisk .... |
| 23:19.06 | Qwell | need to reencode it too |
| 23:19.25 | codefreeze | snuff-work: bugs in alaw and ulaw, yes, I 'oversaw' that... |
| 23:19.45 | file | me reencodes Qwell in H263 |
| 23:19.47 | file | eep |
| 23:19.50 | file | I lost my / |
| 23:20.05 | Qwell | not even H263+? How rude |
| 23:20.05 | russellb | you lost your ...root? |
| 23:20.20 | russellb | and also ... where'd his audio go? |
| 23:20.23 | russellb | is he now muted? |
| 23:20.29 | file | I made it disappear |
| 23:20.31 | *** mode/#asterisk-dev [+b %Qwell!*@*] by Qwell |
| 23:20.34 | putnopvut | audio is overrated. |
| 23:20.39 | *** mode/#asterisk-dev [-b %Qwell!*@*] by Qwell |
| 23:20.40 | Qwell | :( |
| 23:20.40 | russellb | ha |
| 23:20.47 | *** mode/#asterisk-dev [+b %Qwell!*@*] by russellb |
| 23:20.50 | Qwell | pwnt |
| 23:20.52 | Qwell | :p |
| 23:20.54 | *** mode/#asterisk-dev [-o Qwell] by russellb |
| 23:20.56 | file | nubly |
| 23:21.00 | putnopvut | oi! |
| 23:21.04 | russellb | pwnt! |
| 23:21.05 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 23:21.07 | Qwell | orly? |
| 23:21.09 | russellb | :) |
| 23:21.14 | *** mode/#asterisk-dev [-b %Qwell!*@*] by Qwell |
| 23:21.15 | Qwell | :D |
| 23:21.25 | russellb | it's a good thing i don't have a massop CLI command in this client ... |
| 23:21.25 | Qwell | it's so 6:20 |
| 23:21.29 | russellb | i'd probably randomly do it |
| 23:21.32 | Qwell | massop? |
| 23:21.36 | russellb | mop? |
| 23:21.39 | Qwell | massdeop? |
| 23:21.40 | russellb | op everyone in the channel |
| 23:21.42 | Qwell | oh |
| 23:21.52 | putnopvut | Do it in #asterisk |
| 23:21.55 | russellb | lol |
| 23:21.56 | putnopvut | It'll be funny. |
| 23:21.58 | russellb | that would be chaos |
| 23:22.21 | putnopvut | Only 282 people in there right now... |
| 23:22.23 | Qwell | nobody would have access to chanserv or anything |
| 23:22.51 | file | everyone is always opped in my heart <3 |
| 23:22.55 | russellb | awwww |
| 23:22.56 | file | Qwell: except you |
| 23:23.06 | Qwell | am I +a? |
| 23:23.20 | Qwell | +a > +o |
| 23:23.26 | file | I'm afraid in my heart you are nothing... |
| 23:23.29 | file | and my heart is +m |