| 00:00.34 | jmls | I think that kickstart is the ideal solution for me: I can use a standard centos dvd, install the required packages and run scripts to download / compile and install asterisk etc |
| 00:01.22 | jmls | it's currently taking me 10 minutes to install my asterisk setup on a bare-metal server (using a vm) |
| 00:01.49 | tzafrir_laptop | jmls, kickstart, or whatever |
| 00:01.54 | snuff-work | mm it is rather handy.. guess we are talking centos? |
| 00:01.57 | tzafrir_laptop | (debian has something equivalent) |
| 00:02.08 | jmls | now, in order to make it lean and mean, what are the minumum packages needed forcompiling and running asterisk / zaptel ? |
| 00:02.22 | jmls | snuff-work: yeah! it's cool! |
| 00:02.51 | snuff-work | well.. jmls i always just installed core services.. aka no boxes for anything checked.. |
| 00:02.52 | tzafrir_laptop | jmls, depends on the optional features you want from asterisk |
| 00:03.10 | snuff-work | it still installed useless stuff.. cups etc.. |
| 00:03.16 | tzafrir_laptop | but I alsoready gave you links to my scripts that do exactly that |
| 00:03.35 | jmls | weeeellll, I'm compiling asterisk from source on the target machine |
| 00:03.44 | tzafrir_laptop | (prereq.sh is already in asterisk now under a different name, and likewise a zaptel one) |
| 00:03.50 | jmls | tzafrir_laptop: yeah, thanks |
| 00:04.02 | jmls | but not 1.4 ? |
| 00:04.06 | tzafrir_laptop | if you do that, consider using packages |
| 00:04.20 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 00:04.20 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 00:04.21 | snuff-work | jmls.. i can send u a msg with all the packages i used to install after a core CentOS install |
| 00:04.39 | jmls | snuff-work: that would be most helpful, kind one ;) |
| 00:04.43 | tzafrir_laptop | wonders why people like centos so much |
| 00:04.52 | tzafrir_laptop | The kernel packaging is so messed up |
| 00:04.52 | snuff-work | because its easy.. |
| 00:04.55 | jmls | it works |
| 00:04.58 | tzafrir_laptop | bah |
| 00:05.02 | jmls | :) |
| 00:05.03 | eliel | debian too |
| 00:05.06 | tzafrir_laptop | works?? |
| 00:05.43 | *** part/#asterisk-dev wolfelectronic (n=wolfelec@91.112.227.150) |
| 00:05.57 | eliel | rpm-- |
| 00:06.10 | tzafrir_laptop | it's not rpm itself |
| 00:06.22 | tzafrir_laptop | It's the people around it. Can't make up their mind |
| 00:31.21 | *** join/#asterisk-dev Deeewayne (n=dwayne@76.29.245.9) |
| 00:31.21 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 00:58.52 | jmls | urm, bugger, dog's danglies ... snuff-work ... ping ... ? |
| 01:00.38 | snuff-work | mm? |
| 01:00.46 | jmls | I , ahhhmmm, managed to close our conversation without having logging enabled ... so I lost all that wonderful information you gave me ... really sorry ... |
| 01:00.55 | snuff-work | np i resend |
| 01:00.57 | jmls | is a complete kipper |
| 01:01.15 | Corydon76-dig | not an anchovy? |
| 01:01.24 | snuff-work | mm.. kipper.. haven't heard that word in a while |
| 01:02.14 | jmls | nah, we do kippers over here ;) |
| 01:02.34 | jmls | anchovies are in American pizza's apparently .. |
| 01:02.37 | Corydon76-dig | Or sardines? |
| 01:02.40 | jmls | eeeewwww |
| 01:02.54 | Corydon76-dig | sardines are loaded with nutrients |
| 01:03.09 | jmls | so is vitamin pills |
| 01:03.10 | Corydon76-dig | Total omega-3's |
| 01:03.12 | lmadsen | anchovies are too fishy and salty for me |
| 01:03.28 | Corydon76-dig | lmadsen: hence why you should be eating fresh |
| 01:03.37 | jmls | how the hell can a fish be "too fishy" ?? |
| 01:03.38 | *** join/#asterisk-dev schmooze_Tony (n=schmooze@rrcs-67-52-214-19.west.biz.rr.com) |
| 01:03.58 | Corydon76-dig | jmls: he means it's been too long out of the ocean |
| 01:04.05 | lmadsen | jmls: that strong fish odour smell/taste I hate... but I do enjoy sushi and other non-strong-smelling fish |
| 01:04.06 | jmls | ooohhhhhhh ;) |
| 01:04.24 | lmadsen | kinda goes for cheeses too |
| 01:04.34 | Corydon76-dig | I like my sardines packed in mustard, anyway |
| 01:04.39 | jmls | esp French cheese |
| 01:04.46 | jmls | all that mould |
| 01:05.03 | Corydon76-dig | Considering that cheese IS mold... |
| 01:05.19 | jmls | I'm sure the Frnch just tell the brits that the "blue mould is good for you" so that they can take the piss out of us later .. |
| 01:05.28 | Corydon76-dig | "How do you tell when cheese has gone bad?" |
| 01:05.38 | jmls | when an Englishman likes it |
| 01:05.39 | lmadsen | waits for the punchline |
| 01:05.46 | Corydon76-dig | jmls: you know what the blue is... |
| 01:05.57 | lmadsen | when a Dane won't eat it :) |
| 01:05.57 | jmls | covers his eyes |
| 01:06.11 | snuff-work | well... try on unparsured cheese.. can grow hair! |
| 01:06.24 | Corydon76-dig | jmls: oh, you don't know what the blue is? |
| 01:06.38 | lmadsen | doesn't either |
| 01:06.40 | jmls | whistles |
| 01:06.48 | Corydon76-dig | jmls: hint: the blue in blue cheese is the cure for the Black Plague |
| 01:07.11 | Corydon76-dig | Penicillin |
| 01:07.37 | Corydon76-dig | Seriously |
| 01:07.41 | Qwell | bleu* |
| 01:08.38 | Corydon76-dig | Bubonic Plague, Black Death, you know, the thing that killed millions of Europeans in the 1500s, and the French had the cure for it all along |
| 01:08.38 | jmls | it's the Norwegians that have a really really really smelly fish .. there was some woman who was arrested for "assault with a deadly weapon" or something like that |
| 01:09.09 | jmls | And they never shared. Pfft. Typical |
| 01:09.16 | Corydon76-dig | Heh |
| 01:10.31 | *** join/#asterisk-dev mog (n=mog@74.95.48.254) |
| 01:10.31 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 01:13.47 | jmls | Need to get some zzzz's. |
| 01:13.53 | jmls | thanks guys for the help today |
| 01:14.06 | jmls | goodnight all |
| 01:23.23 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 01:23.23 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 01:28.52 | eliel | not a question for this hour, but why it is needed to lock local_pvt in chan_local/local_call() if the passed channel is locked? |
| 01:30.48 | eliel | and if for some reason it is needed, why we are not releasing the lock before the second 'return -1'; ? |
| 01:35.06 | eliel | Corydon76-dig: I was looking and I think you miss an unlock() in chan_local: http://svn.digium.com/view/asterisk/trunk/channels/chan_local.c?r1=142929&r2=146713 |
| 01:36.23 | Corydon76-dig | eliel: um, probably |
| 01:37.05 | Corydon76-dig | eliel: open a bug and attach a patch, if you wouldn't mind |
| 01:37.14 | Corydon76-dig | My trunk is presently in a state of transition |
| 01:37.55 | eliel | Corydon76-dig: ok, no problem |
| 02:44.23 | *** join/#asterisk-dev qdk_ (n=qdk@77.241.140.253.bredband.3.dk) |
| 02:55.43 | *** join/#asterisk-dev mog (n=mog@c-68-62-219-86.hsd1.al.comcast.net) |
| 02:55.43 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 03:29.02 | *** join/#asterisk-dev aliraja (n=aliraja@202.125.156.122) |
| 03:59.43 | *** join/#asterisk-dev jameswf (n=james@ip68-3-60-66.ph.ph.cox.net) |
| 07:02.31 | *** join/#asterisk-dev jcollie` (n=jcollie@fedora/jcollie) |
| 07:20.43 | jmls | morning all ! |
| 07:34.55 | *** join/#asterisk-dev kamanashisroy (n=kamanash@119.30.34.2) |
| 07:59.15 | *** join/#asterisk-dev mvanbaak (i=vanbaak@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 07:59.15 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 08:01.34 | *** join/#asterisk-dev mbranca (n=matteo@81.208.92.210) |
| 08:04.18 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 08:04.18 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 08:10.51 | *** join/#asterisk-dev denon (i=root@synapse.subneural.net) |
| 08:14.16 | *** join/#asterisk-dev denon (i=denon@synapse.subneural.net) |
| 08:15.19 | *** join/#asterisk-dev denon (i=denon@synapse.subneural.net) |
| 08:42.04 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 09:18.05 | *** join/#asterisk-dev jmls (n=jmls@host217-36-208-155.in-addr.btopenworld.com) |
| 09:50.52 | *** join/#asterisk-dev qdk (n=qdk@85.235.253.141) |
| 09:57.32 | *** join/#asterisk-dev _charly_ (i=kroseneg@sunrise.schmidham.net) |
| 10:08.24 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 10:08.24 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 10:29.44 | steliosk | tzafrir_laptop : ping |
| 10:29.55 | tzafrir_laptop | steliosk, pong |
| 11:00.19 | eliel | There is no mor anotate in svnview ? |
| 11:01.52 | mvanbaak | no, that has been disabled for now |
| 11:02.47 | mvanbaak | remember the downtime of bugs and svn two weeks ago? that was caused by a webspider indexing 'svn anotate' for every rev in the repo |
| 11:03.11 | eliel | ups |
| 11:03.20 | mvanbaak | I think Russell sent a mail about it to the -dev mailinglist |
| 11:03.31 | eliel | yeaph, i remember |
| 11:03.50 | eliel | thanks |
| 11:03.56 | mvanbaak | no problem |
| 11:08.39 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:16.39 | *** join/#asterisk-dev jeffg (i=eGXbwLCi@pdpc/supporter/active/jeffg) |
| 11:18.34 | tzafrir_laptop | next after sparse, is checking dahdi source files with checkpath.pl |
| 11:19.02 | tzafrir_laptop | I wonder if it is possible to adapt that script to report asterisk codingguidelines |
| 11:24.17 | *** join/#asterisk-dev steliosk (n=Stelios@62.169.196.50) |
| 11:36.42 | *** join/#asterisk-dev Schmooze (n=Schmooze@rrcs-67-53-189-178.west.biz.rr.com) |
| 12:34.38 | *** join/#asterisk-dev wolfelectronic (n=wolfelec@91.112.227.150) |
| 12:58.51 | *** join/#asterisk-dev ctooley (n=ctooley@72-63-226-198.area2.spcsdns.net) |
| 12:58.51 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 13:16.54 | *** part/#asterisk-dev Deeewayne (n=dwayne@76.29.245.9) |
| 13:31.18 | eliel | jsmith-away: ping |
| 13:34.26 | eliel | I found a complex situation that could handle with the current syntax XML definition and is this one: |
| 13:34.47 | eliel | Gosub([[context,]exten,]priority[(arg1[,...][,argN])]) |
| 13:36.03 | *** join/#asterisk-dev bkw_ (n=brian@freeswitch/developer/bkw) |
| 13:39.21 | eliel | s/could/couldn't/ |
| 13:45.20 | *** join/#asterisk-dev dbailey (n=doug@nat/digium/x-029e6619cbd159d3) |
| 13:52.20 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:52.20 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 13:58.27 | *** join/#asterisk-dev sruffell (n=sruffell@nat/digium/x-81f67e6c5f5c4ee9) |
| 13:58.27 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:24.58 | *** join/#asterisk-dev Deeewayne (n=Deeewayn@nat/digium/x-fb3618f5c422c162) |
| 14:24.58 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 14:28.51 | *** part/#asterisk-dev wolfelectronic (n=wolfelec@91.112.227.150) |
| 14:36.08 | *** join/#asterisk-dev mog (n=mog@nat/digium/x-4534820a0f56f3c7) |
| 14:36.08 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 14:55.18 | *** join/#asterisk-dev Madkiss (i=madkiss@freenode/staff-emeritus/madkiss) |
| 14:55.19 | Madkiss | hi all |
| 14:55.28 | Madkiss | while trying to start asterisk 1.4.21, I am getting this error message: |
| 14:55.41 | Madkiss | terminate called after throwing an instance of 'std::length_error' what(): vector::_M_fill_insert |
| 14:56.05 | tzafrir_laptop | Madkiss, h323? |
| 14:56.08 | seanbright | must be |
| 14:56.30 | Madkiss | I don't think I need that, how do I disable it? |
| 14:56.37 | seanbright | noload => chan_h323.so |
| 14:56.39 | Madkiss | (and why did it work until like an hour ago?) |
| 14:56.43 | seanbright | in /etc/asterisk/modules.conf |
| 14:56.50 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 14:58.04 | Madkiss | removing h323 did not fix the problem, but, erm. |
| 14:58.11 | Madkiss | it's not 100% reproducible |
| 14:58.18 | Madkiss | like one out of 10 times, it will work |
| 14:59.30 | seanbright | the only c++ code in asterisk is chan_h323 i think |
| 14:59.34 | seanbright | looks... |
| 14:59.58 | tzafrir_laptop | also chan_vpb |
| 15:00.03 | seanbright | ah |
| 15:00.44 | tzafrir_laptop | and if you don't know what it is, you don't need it :-) |
| 15:03.54 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 15:04.36 | Madkiss | it was vpb, i think. |
| 15:06.18 | Madkiss | at least after disabling it, asterisk starts up again |
| 15:08.36 | *** join/#asterisk-dev putnopvut (n=putnopvu@nat/digium/x-f14a376e434ac6e2) |
| 15:08.36 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:17.22 | *** join/#asterisk-dev bkw_ (n=brian@freeswitch/developer/bkw) |
| 15:18.45 | *** join/#asterisk-dev jtodd (i=gbyjude4@ns2.loligo.com) |
| 15:18.45 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 15:37.26 | kpfleming | fair warning... for some reason the SMTP daemon on bugs.digium.com has not been running for some time, and i just started it. there is a bunch of mail in the queue that was never delivered, so don't be surprised if you receive some old messages in the next few hours |
| 15:39.32 | lmadsen | putnopvut: well that maybe explains why you weren't getting emails when I assigned some issues to you |
| 15:39.34 | lmadsen | expects putnopvut to get a lot of emails |
| 15:40.00 | putnopvut | lmadsen: no, that wasn't the case. |
| 15:40.11 | kpfleming | no, this would only have been mail that couldn't be delivered on the first attempt, which most mail is |
| 15:40.11 | lmadsen | fail |
| 15:49.29 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 15:49.52 | steliosk | tzafrir_laptop : found anything ? |
| 15:50.04 | tzafrir_laptop | no |
| 15:52.40 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 15:55.30 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.8.157) |
| 16:07.32 | *** join/#asterisk-dev JerJer (n=PhatJ@pdpc/supporter/bronze/jerjer) |
| 16:26.24 | *** join/#asterisk-dev ctooley (n=ctooley@doc-24-32-196-69.concordia.ks.cebridge.net) |
| 16:26.24 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 16:30.27 | *** join/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 16:30.27 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 16:33.52 | *** join/#asterisk-dev bbryant (n=brett@68.208.65.50) |
| 16:33.52 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 16:37.23 | eliel | russellb: I have a problem with the appdocsxml branch, I can't do this kind of syntax: |
| 16:37.25 | eliel | Gosub([[context,]exten,]priority[(arg1[,...][,argN])]) |
| 16:37.48 | eliel | or this: GosubIf(condition?labeliftrue[(arg1[,...])][:labeliffalse[(arg1[,...])]]) |
| 16:38.39 | russellb | yikes |
| 16:38.41 | eliel | I have created a patch to fix the first type of syntax to allow parameter having something like: paramname[(...)] or paramname(....) |
| 16:39.01 | eliel | but the gosubif kind is very difficult to generate in a consistent manner |
| 16:40.05 | eliel | what we can do is make "labeliftrue[(arg1[,...])]" be the parameter name |
| 16:40.15 | eliel | that will be a workaround |
| 16:41.05 | eliel | if we have gosubif without [(XXXX)] there is no problem at all |
| 16:41.57 | russellb | Ok, well don't stress out over it or anything, we don't have a deadline :) |
| 16:42.44 | russellb | I'll try to take a look after hours |
| 16:42.48 | russellb | to see if I can help out |
| 16:42.58 | russellb | probably won't be able to work on this again until this weekend, though |
| 16:43.31 | Corydon76-dig | eliel: heh, didn't I warn you about that syntax? |
| 16:43.58 | Corydon76-dig | or was that only mvanbaak |
| 16:44.30 | russellb | Corydon76-dig: you always have to be the one making things difficult in this, :-p |
| 16:44.33 | russellb | j/k ;-) |
| 16:45.09 | eliel | Corydon76-dig: heh, don't remember... |
| 16:45.18 | Corydon76-dig | I'm always difficult |
| 16:46.57 | eliel | the solution will be: |
| 16:46.57 | Corydon76-dig | Not as difficult as oej, but hey! ;-) |
| 16:47.08 | putnopvut | kpfleming: heh, I'm suddenly getting notifications of new bug notes from July :) |
| 16:47.46 | kpfleming | yeah, sorry |
| 16:48.03 | Corydon76-dig | Mem'ries... all alone in the mooooonlight |
| 16:48.04 | eliel | 1. Redo the syntax <parameter><argument> definition |
| 16:48.46 | Corydon76-dig | 2. Come up with another syntax that makes standard documentation difficult. |
| 16:48.47 | eliel | 2. Go on with hiding the internals of "labeliftrue[(arg1[,...])]" using it as a name of a parameter |
| 16:48.51 | Corydon76-dig | 3. Go to 1. |
| 16:49.32 | putnopvut | lol, and one from march, too. That's just awesome. |
| 16:49.47 | putnopvut | i'm not complaining, I just think it's funny since these are long-closed issues. |
| 16:50.08 | eliel | 4. Remove Gosubif hehe |
| 16:50.33 | Corydon76-dig | putnopvut: you were expecting the SMTP server to be smarter than that? |
| 16:50.52 | putnopvut | Corydon76-dig: no, I'm just commenting on the situation in general. |
| 16:51.21 | putnopvut | Corydon76-dig: it would be even funnier if I hadn't read Kevin's warning about the SMTP daemon being restarted due to its not functioning properly. |
| 16:51.30 | Corydon76-dig | eliel: you have to admit, though, that the syntax is actually fairly understandable |
| 16:51.56 | Corydon76-dig | Arcane it is not. |
| 16:52.48 | Corydon76-dig | What's even nicer is when the label is just a priority alias |
| 16:52.50 | eliel | Corydon76-dig: yes! I don't think the problem is in Gosub |
| 16:53.08 | Corydon76-dig | Gosub(stdexten(101,SIP/101)) |
| 16:53.40 | Corydon76-dig | It looks downright magical |
| 16:53.47 | eliel | that is not a problem, the problem is only in GosuIf |
| 16:54.12 | eliel | Gosub is a problem with the current branch but I have a patch to fix that situation. |
| 16:54.43 | eliel | I didn't commit it yet, because if we will redo all the syntax definition again it is not necesary to fix this problem |
| 16:55.33 | eliel | (I don't like the idea of thinking again the syntax) |
| 16:55.47 | eliel | so, the workaround is there to fix the problem :-) |
| 16:56.21 | Corydon76-dig | Heh, I love architecting design |
| 16:56.59 | eliel | mee too, but no when we think we have finished. :) |
| 16:57.04 | Corydon76-dig | Architecting a design is when programming comes closest to a form of art |
| 17:01.11 | kpfleming | hey... i had a thought about our recent wav/WAV/Wav debacle |
| 17:01.31 | kpfleming | what if our standard file extensions were actually .gsm.wav, .sln8.wav and .sln16.wav |
| 17:01.51 | russellb | it's certainly a lot more expressive .. |
| 17:01.59 | kpfleming | most players will still work |
| 17:02.00 | russellb | and probably the best we can do with the current model |
| 17:02.04 | russellb | right |
| 17:02.06 | kpfleming | since the last extension is .wav |
| 17:02.15 | *** join/#asterisk-dev wolfelectronic (n=wolfelec@91.112.227.150) |
| 17:02.26 | Corydon76-dig | s/debacle/experiment/ |
| 17:02.32 | kpfleming | lol |
| 17:02.37 | kpfleming | tomato. watermelon |
| 17:03.29 | Corydon76-dig | fruits of labor |
| 17:03.43 | kpfleming | sure |
| 17:03.45 | kpfleming | so, thoughts? |
| 17:03.54 | putnopvut | What problem is this addressing again? |
| 17:04.00 | putnopvut | wav/WAV/Wav? |
| 17:04.03 | kpfleming | 'wav' is just a container, not a complete format |
| 17:04.19 | Corydon76-dig | and so WAV becomes .gsm.wav ? |
| 17:04.24 | kpfleming | yep |
| 17:04.34 | putnopvut | interesting. |
| 17:05.46 | tzafrir_laptop | but how does it help to address the diffrent subformats? |
| 17:06.15 | kpfleming | it changes from using different cases of wav/WAV/etc. to explicit names |
| 17:06.26 | kpfleming | we will know that .gsm.wav is actually WAV49 GSM |
| 17:06.26 | Corydon76-dig | tzafrir_laptop: it's only really addressing the case-insensitive nature of certain filesystems |
| 17:06.39 | tzafrir_laptop | ok |
| 17:06.40 | kpfleming | and people not usually having to care what the case is |
| 17:06.40 | *** join/#asterisk-dev TonyM (n=TonyM@softins.claranet.co.uk) |
| 17:07.12 | kpfleming | so if someone records into SLIN8 WAV on a Windows PC, then copies the file over onto the Asterisk system, it breaks |
| 17:07.20 | kpfleming | because Asterisk thinks it is WAV49 GSM |
| 17:07.44 | TonyM | Did someone poke carolina.digium.com? I've just received some old emails from it dating back to Dec 2007! |
| 17:08.02 | Qwell | TonyM: yep |
| 17:08.25 | TonyM | hehe - cool |
| 17:08.25 | Corydon76-dig | TonyM: with a trident |
| 17:09.19 | Corydon76-dig | or no, an enema... and then everything started flowing again... |
| 17:09.39 | russellb | ugh |
| 17:09.48 | Corydon76-dig | giggles |
| 17:10.05 | TonyM | well I expect I got away lightly with only 4 pending emails (well, so far, at least) |
| 17:10.07 | Corydon76-dig | comes up with the BEST analogies |
| 17:21.55 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 17:23.35 | eliel | appdocsxml: ping |
| 17:26.15 | appdocsxml | pong |
| 17:26.23 | *** join/#asterisk-dev philipp64 (n=chatzill@63.224.43.239) |
| 17:26.42 | appdocsxml | redo my syntax please! |
| 17:26.55 | eliel | appdocsxml: ok, I will |
| 17:27.49 | putnopvut | lol |
| 17:31.41 | *** part/#asterisk-dev wolfelectronic (n=wolfelec@91.112.227.150) |
| 17:51.18 | atis_work | why don't just read wav header for format detection? |
| 17:51.40 | kpfleming | how do you have multiple formats available then? |
| 17:51.46 | kpfleming | if you want SLIN8 and SLIN16, you can't |
| 17:51.53 | kpfleming | because the file names have to differ in some way |
| 17:52.46 | atis_work | i still can have .sln and .gsm |
| 17:53.13 | kpfleming | sure |
| 17:53.28 | kpfleming | but people want them in WAV wrappers, so they can play them on Windows and in email clients |
| 17:53.43 | atis_work | well, that's recording part |
| 17:54.09 | kpfleming | what? |
| 17:54.18 | atis_work | btw, as i have noticed playback of 8khz file is problematic anywhere |
| 17:54.41 | kpfleming | uhh... people play back 8KHz recordings from voicemail in email all the time |
| 17:54.44 | kpfleming | including me |
| 17:54.45 | atis_work | i mean, voicemail can record .wav with wav49 or whatever else |
| 17:54.46 | kpfleming | it works fine |
| 17:54.58 | kpfleming | sure, but then how do you play it back? |
| 17:55.05 | kpfleming | what if they want multiple formats? |
| 17:55.34 | atis_work | not sure what are you asking now |
| 17:55.53 | atis_work | voicemails can be recorded in whatever .wav required |
| 17:56.09 | Corydon76-dig | atis_work: multiple formats, again |
| 17:56.23 | atis_work | there's one specified in voicemail.conf |
| 17:56.25 | kpfleming | and then when someone calls in to voicemail to play them back, you are forced to transcode, because you coudln't store multiple formats |
| 17:56.28 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 17:57.52 | atis_work | no, i didn't said that you can't keep multople formats |
| 17:57.52 | Corydon76-dig | atis_work: I think you've totally missed the point |
| 17:57.52 | atis_work | recording -> .gsm, .ulaw, .wav(SLIN16) |
| 17:57.55 | atis_work | playback <- .ulaw, .gsm, .wav (whatever there is) |
| 17:58.15 | Corydon76-dig | So you can only record in a single wav format ? |
| 17:58.44 | atis_work | Corydon76-dig: you can record in all native formats, why would you need to have more than one .wav |
| 17:59.30 | atis_work | if you record .wav, it's defineately meant for usage outside asterisk |
| 17:59.30 | Corydon76-dig | atis_work: different sizes for different uses. Think uncompressed for mixing, compressed for mobile devices |
| 18:00.03 | atis_work | Corydon76-dig: and you would send them all out in one email? |
| 18:00.06 | kpfleming | yes |
| 18:00.11 | kpfleming | or store them in IMAP |
| 18:00.36 | Corydon76-dig | atis_work: I might send different formats to different recipients, depending upon the target device |
| 18:01.31 | atis_work | kpfleming: ok, that sounds reasonable |
| 18:01.46 | atis_work | Corydon76-dig: would you know what is target device when sending email notification? |
| 18:02.01 | Corydon76-dig | atis_work: I know, based upon the user |
| 18:02.16 | Corydon76-dig | atis_work: the email format is specifiable per user |
| 18:02.31 | atis_work | sure, but so there is recording format |
| 18:03.02 | Corydon76-dig | Right, and the names would otherwise conflict |
| 18:04.18 | atis_work | well, you can always write format=wav(slin16) in .conf file |
| 18:04.32 | atis_work | i think it's more problem when you can't drop in any .wav for asterisk |
| 18:04.46 | atis_work | altough it's not that good for transcoding |
| 18:05.47 | atis_work | kpfleming: you mentioned that you can play 8khz files anywhere? have you tried windows media player (which users usually choose :) |
| 18:06.58 | Corydon76-dig | atis_work: you DO realize that most of the developers here don't use Windows at all, right? |
| 18:07.11 | atis_work | Corydon76-dig: sure, me neither |
| 18:07.24 | Corydon76-dig | Mandating specific Windows programs to use is NOT what we're about |
| 18:07.25 | atis_work | Corydon76-dig: but it's hard to explain users, that they have to install another player :p |
| 18:07.30 | *** join/#asterisk-dev rene- (n=renemend@200.34.66.137) |
| 18:07.31 | *** join/#asterisk-dev festr_ (n=festr@ns.hiro.cz) |
| 18:08.36 | atis_work | Corydon76-dig: what i was trying to explain is, that you have to convert recordings anyway, before sending them to users |
| 18:08.46 | atis_work | so far, best what i have found is 8khz mp3 |
| 18:09.36 | festr_ | asterisk 1.4: if using realtime for sip and voicemail(odbc), why WMI subscription is not updated immediatly? I've to make new sybscriotion to reflect new state of voicemail. MWI checking interval is 10secs |
| 18:10.29 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:22.25 | eliel | I need some help with this definition: http://pastebin.com/d4ab4fc23 |
| 18:22.59 | eliel | this is the way to describe the GotoIf syntax: GotoIf(condition?[labeliftrue][:labeliffalse]) |
| 18:23.06 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@local.xorcom.com) |
| 18:23.06 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 18:23.20 | eliel | the big problem is the separator |
| 18:24.14 | eliel | as you can see we have two levels: paramater and argument, I think we need N levels |
| 18:24.33 | eliel | because an argument could be made of multiple parameter inside |
| 18:25.11 | eliel | so, I think the best way to avoid this situation would be using only arguments in a multiple level schema |
| 18:26.29 | codefreeze-lap | eliel: I don't think the args are really recursive in this respect. |
| 18:28.23 | codefreeze-lap | eliel: How do you picture the multiple parameters? |
| 18:29.58 | eliel | mutiple parameters are handled by an attribute multiple="true" |
| 18:30.47 | tzafrir_laptop | kpfleming, there's one small userspace-interface-related issue I'd like to sort out before upcoming dahdi release: |
| 18:31.14 | codefreeze-lap | eliel: I guess I'm trying to imagine a GotoIf with multiple params... how would it appear? |
| 18:31.33 | eliel | Gotoif with multiple params? |
| 18:31.37 | tzafrir_laptop | rename dahdi device names from 'dahdictl' to 'dahdi!ctl' etc. |
| 18:31.55 | tzafrir_laptop | This would make the udev rules unnecessary (except the permissions ones) |
| 18:32.12 | codefreeze-lap | eliel: yes, do you have an example? |
| 18:33.35 | tzafrir_laptop | http://svn.digium.com/view/dahdi?view=rev&revision=4769 |
| 18:33.37 | eliel | codefreeze-lap: nop, do you? |
| 18:33.56 | tzafrir_laptop | but I guess that something else might be required at that spot |
| 18:34.51 | codefreeze-lap | eliel: no, neither can I... |
| 18:35.11 | eliel | but I am generating the syntax here, not all the possible values |
| 18:35.30 | eliel | so I need with a suitable xml generate something like: GotoIf(condition?[labeliftrue][:labeliffalse]) |
| 18:37.46 | atis_work | eliel: btw, why do you make labeliftrue optional? is it intended so? |
| 18:38.22 | codefreeze-lap | eliel: if we are talking xml syntax here, then your allowing an argsep in parameter would cover it... |
| 18:39.13 | eliel | the pastebin: http://pastebin.com/d4ab4fc23 show how we are handling this syntax |
| 18:39.23 | eliel | s/show/shows |
| 18:40.02 | eliel | but we need a better way to manage this complex syntax, instead of doing what I am doing, because we have MORE complex syntax than this one. |
| 18:40.46 | kpfleming | tzafrir_laptop: didn't we already try that and it failed for some users? |
| 18:40.52 | codefreeze-lap | eliel: atis_work makes a good point, I'd imagine that if you don't have a labeliftrue, you'd have to have a labeliffalse,.... |
| 18:41.06 | tzafrir_laptop | kpfleming, no, I don't recall this failing anywhere |
| 18:41.29 | tzafrir_laptop | sadly, I don't recall anyone besides me trying this |
| 18:42.17 | tzafrir_laptop | but this is what the udev maintainers suggested, instead of keeping most of the zaptel rules |
| 18:42.57 | tzafrir_laptop | (suggested for zaptel, that is) |
| 18:43.12 | codefreeze-lap | eliel: you have a construct to assert that among a group of optional items, at least 1 must be included? |
| 18:43.51 | eliel | codefreeze-lap: don't think so, it will make too hard to read a syntax like that |
| 18:44.10 | atis_work | codefreeze-lap: well, i didn't mean exactly that, but looking at xml it looks like that <destination .. required> does that |
| 18:44.51 | eliel | atis_work: yeaph |
| 18:44.54 | eliel | maybe |
| 18:44.56 | eliel | good point! |
| 18:44.58 | kpfleming | tzafrir_laptop: so how recent of a version of udev is required for these rules to work? |
| 18:45.10 | atis_work | codefreeze-lap: i'm not very familiar with the if?a:b syntax, but having if?:b seems confusing, as you can normally inverse condition |
| 18:45.16 | tzafrir_laptop | "recent" == newer of 0.17 |
| 18:45.31 | kpfleming | that old? go ahead and merge it then |
| 18:45.32 | codefreeze-lap | atis_work: good point, it would imply that there must be something there... good |
| 18:46.15 | codefreeze-lap | atis_work: yes, but ?:whatever gives you more.... power! ;) |
| 18:46.20 | *** join/#asterisk-dev _ys (n=yuri@80.70.236.69) |
| 18:46.22 | atis_work | codefreeze-lap: is it allowed in C? |
| 18:46.52 | atis_work | if so, then let it be :) |
| 18:47.06 | kpfleming | no, both sides of the : in C must have expressions |
| 18:47.46 | codefreeze-lap | I knew there must a good reason to avoid that. |
| 18:47.55 | eliel | ok, we are talking about diferent things |
| 18:48.03 | eliel | I am not discussing the Gotoif syntax |
| 18:48.15 | eliel | or the gosubif |
| 18:48.32 | eliel | I am trying to figure out a way to represent them in XML :) |
| 18:49.15 | codefreeze-lap | eliel: what's the diff between a parameter and an argument, then? |
| 18:50.14 | eliel | that is what I am trying to figure out, don't know if this <parameter><argument> distinction makes any sense with this free kind of syntaxs |
| 18:52.06 | eliel | I think of the parameter as the first level container (param1,param2,param3) |
| 18:52.37 | eliel | then we can have inside each param a single value, or a group of values (arguments). |
| 18:52.46 | codefreeze-lap | Well, if parameter allowed parameter within itself, then you wouldn't need argument. And, your syntax with both is inconsistent; you would have to to put argument inside condition... |
| 18:52.50 | eliel | param1=val1&val2 |
| 18:53.04 | eliel | so: (val1&val2,param2,param3) |
| 18:54.01 | eliel | right now, parameter is not allowed with itself |
| 18:54.36 | codefreeze-lap | is parameter allowed in argument? |
| 18:54.43 | eliel | codefreeze-lap: nop |
| 18:55.18 | eliel | that is why we have a two level schema to describe the syntax (right now) |
| 18:55.58 | seanbright | kpfleming: gcc has an extension for that, too. |
| 18:56.18 | seanbright | char *x = y ? : "foo"; |
| 18:56.31 | kpfleming | uhh... what is the result if not specified? |
| 18:56.39 | seanbright | y |
| 18:56.43 | russellb | that seems bizarre |
| 18:56.48 | seanbright | i wouldn't lie to you |
| 18:56.55 | russellb | y? seriously? |
| 18:57.09 | russellb | ugh, why wouldn't you just say y ? y : "foo" |
| 18:57.20 | russellb | that's a lame extension :-p |
| 18:57.25 | seanbright | well, if you had this: |
| 18:57.42 | seanbright | something = really->long->chain->of->stuff ? : default; |
| 18:58.05 | russellb | so it's an extension for laziness |
| 18:58.11 | seanbright | oh don't even go there |
| 18:58.19 | russellb | haha |
| 18:58.26 | codefreeze-lap | eliel: well, if you have go deeper, I'd make parameter recursive, and get rid of argument |
| 18:58.26 | seanbright | mr. tsk_proc |
| 18:58.35 | russellb | seanbright: touche |
| 18:58.46 | kpfleming | what if the really->long->chain->of->stuff has side effects? |
| 18:58.59 | kpfleming | presumably it doesn't get executed twice, but it would if you repeated it |
| 18:59.16 | seanbright | all i'm saying... is that GCC has an extension that allows you to not specify the true portion of the expression |
| 18:59.23 | seanbright | not saying it should be used. |
| 18:59.36 | seanbright | let me look it up so i know i'm not lying... |
| 19:02.13 | codefreeze-lap | he's right. You can do it. |
| 19:02.36 | Qwell | kpfleming: (really ? (really->long ? really->long->chain : NULL) : NULL) ... |
| 19:02.51 | seanbright | http://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/Conditionals.html#Conditionals |
| 19:02.58 | codefreeze-lap | Just compiled as an experiment: char *x = argv[0] ? : "nothing"; |
| 19:02.59 | seanbright | "Conditionals with Omitted Operands" |
| 19:03.02 | kpfleming | Qwell: lol |
| 19:03.22 | seanbright | a-ha |
| 19:03.25 | seanbright | kpfleming: "When it becomes useful is when the first operand does, or may (if it is a macro argument), contain a side effect." |
| 19:03.33 | kpfleming | senses a batch of commits in the next few days to replace existing statements with this extension |
| 19:03.41 | kpfleming | is very smrt |
| 19:03.52 | seanbright | C# has a similar construct |
| 19:04.01 | kpfleming | they stole it |
| 19:04.09 | seanbright | string s = x ?? y; |
| 19:04.13 | Qwell | similar operand to what? |
| 19:04.35 | Qwell | oh, jeez... |
| 19:04.42 | seanbright | useful for nullable types |
| 19:04.56 | kpfleming | all types are nullable, if you have memset() |
| 19:05.16 | seanbright | Nullable<int> i = null; |
| 19:05.27 | seanbright | or "int? i = null" as a shorthand |
| 19:05.43 | seanbright | is MSmrt |
| 19:07.44 | seanbright | wow |
| 19:07.45 | seanbright | http://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/Case-Ranges.html#Case-Ranges |
| 19:08.08 | Corydon76-dig | nullifies seanbright |
| 19:08.08 | seanbright | gcc <3 |
| 19:08.47 | Qwell | 'A' + 26 == 'a'? |
| 19:08.55 | Qwell | I could've sworn there was a gap |
| 19:08.55 | seanbright | 32 i thought |
| 19:09.12 | Qwell | yeah, there's a gap. that example is bad bad bad |
| 19:09.12 | seanbright | flip the 6th bit |
| 19:09.32 | Qwell | wait no, my eyes just syck |
| 19:09.34 | Qwell | suck too |
| 19:10.18 | putnopvut | case ranges? Wow |
| 19:10.21 | Corydon76-dig | Qwell: might want to make an appointment with the ophthamologist then |
| 19:10.43 | seanbright | putnopvut: that's the 4.3.2 docs, not sure when that was added |
| 19:10.47 | Qwell | Corydon76-dig: it looks like a 'z' here |
| 19:10.53 | Qwell | odd font |
| 19:11.06 | putnopvut | Qwell: yeah, it looked like a lowercase z to me too. |
| 19:11.10 | putnopvut | It is a bad font. |
| 19:11.18 | Corydon76-dig | Use Times Roman, not Zapf Dingbats |
| 19:11.25 | seanbright | actually, according to the docs it has been there since 2.95.3 at least |
| 19:11.37 | putnopvut | seanbright: really? wow. |
| 19:11.44 | seanbright | so we can use it all over the place! |
| 19:11.45 | seanbright | heh |
| 19:11.45 | putnopvut | That's a neat feature. |
| 19:11.51 | putnopvut | seriously we could |
| 19:12.04 | seanbright | well, until mvanbaak complains that it won't compile under openbsd |
| 19:12.09 | seanbright | :) |
| 19:12.10 | putnopvut | lol |
| 19:12.11 | Corydon76-dig | Heh |
| 19:12.12 | Qwell | shouldn't take long |
| 19:12.16 | seanbright | 3... |
| 19:12.17 | seanbright | 2... |
| 19:12.20 | seanbright | :) |
| 19:12.30 | *** join/#asterisk-dev philippel (n=p_lindhe@pool-98-111-70-106.sttlwa.fios.verizon.net) |
| 19:13.43 | seanbright | well, we use GCC extensions elsewhere. |
| 19:14.32 | lmadsen | hi |
| 19:18.45 | seanbright | labels as values... man... these guys are crazy |
| 19:19.19 | jtodd | Hey, it doesn't compile under OpenBSD! |
| 19:19.26 | jtodd | I thought I'd get it in there edgewise... |
| 19:19.31 | seanbright | hehe |
| 19:20.58 | tzafrir_laptop | <seanbright> actually, according to the docs it has been there since 2.95.3 at least |
| 19:21.05 | tzafrir_laptop | in other words: |
| 19:21.13 | tzafrir_laptop | <seanbright> gcc <3 |
| 19:23.20 | Qwell | <3 or < 3? |
| 19:23.38 | russellb | zing! |
| 19:23.42 | seanbright | i was saying <3 |
| 19:23.47 | seanbright | but yes, < 3 as well |
| 19:23.56 | Qwell | I think that would be >= 3 |
| 19:24.15 | seanbright | and if you copy and paste by nick, please include the @ so people know how important i am |
| 19:24.21 | putnopvut | seanbright: does gcc support labeled breaks? (please say yes) |
| 19:24.30 | *** mode/#asterisk-dev [-o seanbright] by russellb |
| 19:24.32 | seanbright | putnopvut: http://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/Labels-as-Values.html#Labels-as-Values |
| 19:24.36 | russellb | not such a big deal now, are you |
| 19:24.38 | seanbright | russellb: saw that coming a MILE away |
| 19:24.59 | russellb | i saw ... you coming ... a mile away ............ |
| 19:25.10 | *** mode/#asterisk-dev [+o seanbright] by ChanServ |
| 19:25.22 | seanbright | pets ChanServ |
| 19:25.42 | russellb | get a room |
| 19:25.48 | seanbright | hehe |
| 19:26.00 | seanbright | putnopvut: i think the answer to your query is 'no' |
| 19:26.07 | putnopvut | dang |
| 19:26.20 | russellb | goto! |
| 19:26.31 | seanbright | there's always that |
| 19:27.13 | seanbright | with local lables |
| 19:27.17 | seanbright | labels* |
| 19:28.22 | seanbright | ... |
| 19:28.31 | seanbright | oh, nm. |
| 19:56.35 | tzafrir_laptop | commit dahdi 5180 |
| 19:56.42 | MuffinMan | [5180] Committed by tzafrir on 2008-10-30T19:16:08.804867Z: Use correct length for the, well, dahdi transcoder device name. |
| 19:56.42 | MuffinMan | .. |
| 19:56.42 | MuffinMan | .. http://svn.digium.com/view/dahdi?view=rev&revision=5180 |
| 19:57.10 | tzafrir_laptop | didn't this issue prevent /dev/dahdi/transcode from being generated earlier on? |
| 19:58.10 | tzafrir_laptop | I just don't recall anybody reporting it, but I guess most such reports don't go through bugs.digium.com |
| 20:01.32 | philipp64 | howdy. anyone know if there's a bug in how comments are parsed in the .conf files? I couldn't find any open bugs. I'm running 1.4.21.2. |
| 20:01.59 | putnopvut | philipp64: I have not heard of any such bugs. |
| 20:02.14 | philipp64 | If I grep /etc/asterisk/*.conf for '^TRUNK', I see one and only one value. it's the right one. |
| 20:02.52 | philipp64 | if I grep for 'TRUNK=', I get: ';TRUNK=Zap/g2 ...' |
| 20:03.54 | philipp64 | if i stop and restart asterisk, then do "show globals" on a console, I get: TRUNK=Zap/g2 |
| 20:04.13 | philipp64 | if I do "dialplan reload" and then "show globals", I get the right value. |
| 20:04.19 | philipp64 | it's very weird. |
| 20:04.49 | philipp64 | what gives? |
| 20:04.53 | putnopvut | That is odd |
| 20:05.09 | seanbright | grep -r /etc/asterisk/*.conf "^\s*TRUNK" |
| 20:05.13 | seanbright | err |
| 20:05.20 | seanbright | the -r is extraneous |
| 20:06.23 | philipp64 | <PROTECTED> |
| 20:06.25 | philipp64 | <PROTECTED> |
| 20:06.32 | seanbright | weird |
| 20:06.54 | seanbright | tzafrir_laptop: before your commit, there was no room for \0, after your commit there is |
| 20:07.00 | seanbright | not sure if there are implications to that |
| 20:07.14 | seanbright | unless sizeof("literal") doesn't include the trailing null |
| 20:08.41 | philipp64 | sizeof() should always measure the entire string, null included. |
| 20:08.52 | seanbright | i never remember |
| 20:09.24 | philipp64 | seanbright: got a minute to look at something out-of-band? |
| 20:09.41 | seanbright | not really, mostly just lurking at the moment |
| 20:09.44 | kpfleming | no, sizeof is totally broken there |
| 20:09.46 | kpfleming | completely |
| 20:09.49 | philipp64 | it's a SIP signalling issue. |
| 20:09.51 | kpfleming | sizeof is being passed a char * |
| 20:09.58 | kpfleming | it will return 4, or 8, depending on pointer size |
| 20:10.03 | kpfleming | totally completely broken |
| 20:10.14 | seanbright | really? |
| 20:10.21 | kpfleming | yes |
| 20:10.26 | seanbright | hrm. |
| 20:10.32 | kpfleming | any constant string in double quotes is a string constant |
| 20:10.36 | philipp64 | ah... you're passing it a pointer to a string? not the string? |
| 20:10.40 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:10.40 | kpfleming | i would think so |
| 20:10.41 | philipp64 | then yes, what they said. |
| 20:10.53 | kpfleming | i don't know of any case where a string in double quotes is actually an array |
| 20:11.04 | kpfleming | maybe... but i wouldn't rely on it in this case |
| 20:11.11 | kpfleming | someone should write a test.c that tests it :-) |
| 20:11.28 | philipp64 | well, "literal" is the same as {'l', 'i', 't', 'e', 'r', 'a' |
| 20:11.35 | seanbright | sizeof("foo bar yay wheeee!") = 20 |
| 20:11.37 | philipp64 | 'l', '\0} |
| 20:11.40 | philipp64 | damn. |
| 20:11.47 | philipp64 | can't type for beans today. |
| 20:11.53 | kpfleming | ok then |
| 20:12.05 | kpfleming | so the question is does it leave space for the trailing null character |
| 20:12.08 | seanbright | http://pastebin.ca/1240917 |
| 20:12.52 | seanbright | sizeof("foo bar yay wheeee!") = 20 |
| 20:12.53 | seanbright | strlen("foo bar yay wheeee!") = 19 |
| 20:13.07 | kpfleming | xlnt |
| 20:13.15 | seanbright | ? |
| 20:13.27 | putnopvut | seanbright: it's how cool people say "excellent" |
| 20:13.31 | seanbright | ohhh |
| 20:13.34 | seanbright | i'm confused |
| 20:13.35 | tzafrir_laptop | siezof("dahdi!") = 7 |
| 20:13.36 | putnopvut | and how kpfleming says it too ;) |
| 20:13.36 | kpfleming | excellent (with fingers steppled ala montgomery burns) |
| 20:13.42 | kpfleming | putnopvut: snap |
| 20:14.01 | seanbright | so what were you saying is broken? |
| 20:14.30 | russellb | C is such teh weird |
| 20:14.42 | kpfleming | nothing |
| 20:14.42 | philipp64 | strlen() is always one less than sizeof(), yeah. |
| 20:14.45 | kpfleming | never mind what i said |
| 20:14.52 | kpfleming | the code tzafrir_laptop committed is correct |
| 20:14.52 | russellb | always is a strong word! |
| 20:15.02 | putnopvut | philipp64: not if the array is sized larger than the string by more than 1. |
| 20:15.07 | putnopvut | But for a literal string, yea. |
| 20:15.12 | philipp64 | well, except of course if you have: strlen("this\0string\0has\0nulls\in\0it!") |
| 20:15.20 | seanbright | ok, so the pre-commit code should have been "+ 4" |
| 20:15.22 | kpfleming | cheater |
| 20:17.06 | philipp64 | anyone on this list good at debugging SIP/SBC issues? |
| 20:17.13 | russellb | nope |
| 20:17.14 | seanbright | putnopvut is |
| 20:17.15 | philipp64 | I'm seeing something I can't figure out. |
| 20:17.17 | seanbright | and russellb |
| 20:17.19 | seanbright | and file |
| 20:17.22 | seanbright | and oeej |
| 20:17.23 | russellb | seanbright: damn you |
| 20:17.24 | seanbright | oej |
| 20:17.30 | file | wasn't me |
| 20:17.37 | philipp64 | ok... and who is feeling brave? |
| 20:17.40 | russellb | seanbright is the best person to ask, though |
| 20:17.42 | seanbright | throws everyone under the bus and goes back to work |
| 20:17.57 | philipp64 | cowards. :-) |
| 20:18.24 | putnopvut | welp, I'll be brave. I've never actually diagnosed an SBC issue, though... |
| 20:19.05 | philipp64 | I've got a trace. it's ftp://ftp.redfish-solutions.com/pub/20081012-135851.pcap |
| 20:19.52 | putnopvut | philipp64: When I paste that in my browser, I get a "no such file or directory" message box. |
| 20:19.56 | philipp64 | call comes in on a SIP trunk. voice menu... caller dials extension... ring to SIP phone (SPA942) behind Asterisk/firewall device. |
| 20:20.03 | philipp64 | (try it without the "pub") |
| 20:20.17 | putnopvut | That worked. Thanks :) |
| 20:20.34 | philipp64 | proftpd is a little weird about the "autoswitch to /pub" on anonymous login... doesn't always work...) |
| 20:21.05 | philipp64 | media stream then gets redirected directly to phone, and asterisk is out of media path. |
| 20:21.26 | philipp64 | problem is, caller never hears called-party's voice. |
| 20:22.04 | philipp64 | it's asterisk 1.4.21. the phone is 192.168.1.7. the PBX is 192.168.1.1/66.232.79.143. The SBC is 60.232.80.9. |
| 20:22.44 | philipp64 | Oh, and asterisk is running on a linux 2.6.20 kernel with the the nf_sip_nat and nf_conntrack_sip plugin enabled. |
| 20:23.01 | putnopvut | What do those plugins do? |
| 20:23.26 | philipp64 | they do the VIA/port-number/ip address munging on the stream. |
| 20:24.05 | putnopvut | Which phone is 192.168.1.7? The callee? |
| 20:24.18 | philipp64 | it's an ALG that looks at the media info from the invite, sets up the RTP stream holes in the firewall, and rewrites the "invite" to use the appropriate NATted port #'s. |
| 20:24.29 | philipp64 | 192.168.1.7 is the called-party. |
| 20:24.41 | philipp64 | the PSTN is behind 66.232.80.9. |
| 20:24.43 | putnopvut | Okay, that's what I thought, given the 192.168.1 address. |
| 20:25.15 | philipp64 | (that's the SBC... the actual soft switch is a coppercom at 172.16.x.x behind 66.232.80.9) |
| 20:27.11 | putnopvut | I'm checking out the pcap right now... |
| 20:27.17 | philipp64 | ok. |
| 20:27.41 | philipp64 | I used wireshark to play back the RTP streams, and the audio is definitely present in it... |
| 20:28.17 | philipp64 | at least on the inside leg of the call. |
| 20:28.43 | philipp64 | I'm trying to get someone at the carrier to instrument the call on either side of the SBC to see if it's being dropped. |
| 20:29.22 | philipp64 | but... it's a mom and pop ILEC... they aren't IP-heads... and even if there's a bug in the Coppercom switch, Coppercom is out of business... so they almost don't want to know. :-( |
| 20:31.36 | putnopvut | Okay, I'm not seeing from the pcap you've given where Asterisk exits from the media stream. All the RTP seems to be flowing through Asterisk in the pcap |
| 20:34.28 | philipp64 | let me go back and look... |
| 20:34.29 | putnopvut | I also think console output may be helpful. Packet number 1089 is a CANCEL that Asterisk sends out, but I'm not sure why it has chosen to do that. |
| 20:35.02 | philipp64 | yeah, I was looking at that too... |
| 20:35.29 | philipp64 | would that put the call into half-duplex? |
| 20:35.33 | philipp64 | (simplex?) |
| 20:35.59 | putnopvut | A cancel would actually end the call. |
| 20:36.20 | putnopvut | it's what's used if you need to end a call before the initial INVITE has received a 200 OK. |
| 20:36.49 | putnopvut | actually |
| 20:37.06 | putnopvut | It does look like that may be the problem here. |
| 20:37.47 | putnopvut | There's a 9 second gap between when the CANCEL is OK'd and when the phone sends a BYE to Asterisk. |
| 20:38.05 | philipp64 | so what should be happening when asterisk rings out the SIP handset after the voice menu? |
| 20:38.22 | philipp64 | how should the media stream get redirected? |
| 20:38.28 | seanbright | Corydon76-dig: did committing that thread change hurt as much as i think it did? |
| 20:39.24 | putnopvut | philipp64: I may have misunderstood you originally, but I would expect that through reINVITES, Asterisk would send the media from your phone to the SBC and from the SBC directly to the phone. |
| 20:39.36 | putnopvut | What's throwing me off is the CANCEL. |
| 20:40.23 | putnopvut | philipp64: I have an idea. |
| 20:40.30 | philipp64 | and the media stream isn't going from 192.168.1.7 => 66.232.80.9... it's still flowing through asterisk. |
| 20:40.38 | putnopvut | correct. |
| 20:41.15 | putnopvut | Actually, nevermind about me having an idea, I was reading part of the capture incorrectly. |
| 20:42.25 | philipp64 | hmmmm... nat=always is on for the phone. |
| 20:42.36 | philipp64 | wouldn't that stop it from talking directly? |
| 20:42.59 | philipp64 | or is the "canreinvite=no" doing that? |
| 20:43.16 | putnopvut | canreinvite=no would prevent Asterisk from trying to redirect the media away. |
| 20:43.28 | Corydon76-dig | seanbright: eh? |
| 20:43.43 | seanbright | the cannot join detached threads commit |
| 20:43.56 | Corydon76-dig | Wasn't his patch |
| 20:44.03 | *** join/#asterisk-dev write_erase (n=Olivier@royale.aixmarseille.com) |
| 20:44.06 | philipp64 | putnopvut: ok, I thought that only affected the signalling stream. |
| 20:44.16 | seanbright | true |
| 20:44.20 | Corydon76-dig | Even a blind squirrel occasionally finds a nut |
| 20:44.23 | seanbright | hehe |
| 20:44.25 | putnopvut | philipp64: no, canreinvite=no means that Asterisk will not try to offload the media. |
| 20:44.45 | philipp64 | well, then things get even stranger. |
| 20:44.59 | Corydon76-dig | Or, in his case, even a crazy nut occasionally make s a good point |
| 20:45.13 | *** join/#asterisk-dev JunK-Y (n=junky@modemcable156.137-20-96.mc.videotron.ca) |
| 20:45.17 | philipp64 | because if I create an extension with Monitor() turned on, and dial that (ringing to the same phone), then I get 2-way voice. |
| 20:45.47 | seanbright | Corydon76-dig: well i look forward to the next installment :) |
| 20:46.41 | Corydon76-dig | seanbright: of a loony? |
| 20:47.07 | seanbright | something like that. |
| 20:47.15 | putnopvut | philipp64: that is indeed odd. Having a monitor on a channel is a condition for not allowing native bridging (i.e. having the media flow directly between the phones). |
| 20:47.34 | putnopvut | But so is canreinvite=no. |
| 20:47.37 | Corydon76-dig | seanbright: the more successful we are, the more loons we will attract |
| 20:48.12 | seanbright | very true |
| 20:48.17 | seanbright | that's how i ended up here |
| 20:48.47 | Corydon76-dig | seanbright: compared to that guy, we're all perfectly normal |
| 20:48.55 | seanbright | that's saying a lot |
| 20:49.10 | write_erase | Do SIP support configuration reload without disturbing in progress communications, in Asterisk ? |
| 20:49.18 | putnopvut | write_erase: yes |
| 20:49.53 | seanbright | hmmm |
| 20:50.04 | seanbright | ./bootstrap is failing for me all of a sudden |
| 20:50.16 | putnopvut | ./bootstrap.sh |
| 20:50.20 | seanbright | that too |
| 20:50.26 | putnopvut | oh, I thought that was the problem :) |
| 20:50.32 | seanbright | heh, if only |
| 20:50.35 | write_erase | I plan to add this feature to chan-sccp ... Hope this is not too hard to do :-) |
| 20:50.53 | seanbright | aclocal: unrecognized option -- `-Iautoconf' |
| 20:51.54 | putnopvut | philipp64: I don't think I'm going to be able to solve what's going on there with just the pcap. I don't have any idea why it is that Asterisk is sending that CANCEL to the SBC. I think that a console trace with the debug and verbosity turned up, as well as SIP debugging turned on would be helpful. |
| 20:53.06 | seanbright | ah, it's kpfleming's fault |
| 20:53.08 | seanbright | heh |
| 20:54.18 | seanbright | 'aclocal -I autoconf' works |
| 20:54.23 | seanbright | 'aclocal -Iautoconf' does not |
| 20:55.10 | kpfleming | oops |
| 20:55.19 | seanbright | is it specific to my version? |
| 20:55.21 | philipp64 | sip debug, verbosity 15... what other debugs? |
| 20:55.39 | seanbright | that change was made a while ago, someone else should have run into it |
| 20:55.43 | philipp64 | and should I enable canreinvite? |
| 20:55.46 | seanbright | heh |
| 20:56.06 | seanbright | ok, i'll commit |
| 20:58.25 | philipp64 | putnopvut: what about "canreinvite=update", or "directrtpsetup=yes" ? |
| 20:58.44 | putnopvut | philipp64: don't set either of those. |
| 20:58.52 | putnopvut | philipp64: directrtpsetup is very screwy |
| 20:59.00 | philipp64 | good to know. |
| 20:59.36 | putnopvut | Oh, and with regards to what other debugs, I'd suggest that you set the debug level up for chan_sip.c. "core set debug 15 chan_sip.c" |
| 20:59.45 | philipp64 | and on the PBX peering with the SBC, I can turn "canreinvite=yes" on, because the PBX has a public address? |
| 21:00.19 | putnopvut | I |
| 21:00.22 | putnopvut | oops |
| 21:01.35 | philipp64 | so where do I want "canreinvite=no" set? |
| 21:01.44 | putnopvut | philipp64: if you have canreinvite=no on any call legs, then reinvites to offload the media will not take place. |
| 21:02.09 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 21:02.10 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 21:02.49 | putnopvut | The problem is, I'm a bit fuzzy on how NAT figures into the reINVITE picture. |
| 21:05.32 | seanbright | commit commit commit |
| 21:05.33 | MuffinMan | Error calling said() for svn: RuntimeError unknown opt_revision_t type |
| 21:06.18 | philipp64 | putnopvut: sent output in a /query |
| 21:06.54 | philipp64 | ok, so no one has canreinvite set. |
| 21:08.17 | ctooley | So, one has to wonder what the "oops" for taking out 436800 ports of capacity with a backhoe is like. |
| 21:09.24 | putnopvut | ctooley: what brought that up? |
| 21:09.47 | ctooley | "Please be aware that we have now confirmed that there is a major fiber cut in the NE. MCI is saying it is near Versailles Virginia and they estimate that it has impacted over 650 DS3s." |
| 21:10.17 | ctooley | "Ahhhh shit!" |
| 21:14.07 | putnopvut | Whenever I see a city in the American South called "Versailles" I always am amused at how the city's name is pronounced. |
| 21:15.08 | ctooley | putnopvut, It's pronounced in perfect a understandable Southeastern United States accent. |
| 21:18.36 | *** join/#asterisk-dev wacky__ (n=abourget@IP-208-88-110-46.mtl.fibrenoire.ca) |
| 21:19.01 | putnopvut | ctooley: that's true; however it isn't pronounced at all the same as the French city of the same name. |
| 21:19.52 | ctooley | putnopvut, It's generally not exactly an exact replica of the French city either, now is it? :) |
| 21:20.44 | kpfleming | yes. exact. |
| 21:21.18 | philipp64 | putnopvut: here in the pacific northwest, we have "coquilles". Want to guess how it's pronounced? |
| 21:21.45 | putnopvut | How? |
| 21:21.53 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 21:21.54 | philipp64 | coke-ills. |
| 21:22.00 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 21:22.00 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 21:22.47 | philipp64 | hurts my ears sometimes. |
| 21:23.19 | philipp64 | ca fait mal aux oreilles. |
| 21:23.42 | seanbright | freedom fries! |
| 21:25.16 | philipp64 | team america, f***ing yeah! |
| 21:25.20 | seanbright | heh |
| 21:26.21 | wacky__ | hey guys |
| 21:27.20 | wacky__ | I've installed 2 * 1.6 and I get garbled sound on both of them.. |
| 21:27.38 | wacky__ | is the reception of * 1.6 good in general ? |
| 21:28.15 | jameswf | wacky__: did you turn off optimization |
| 21:28.56 | wacky__ | jameswf: I tried recompiling with gcc-4.1 .. |
| 21:29.03 | jameswf | wacky__: did you turn off optimization |
| 21:29.08 | wacky__ | no |
| 21:29.12 | jameswf | try it |
| 21:29.14 | wacky__ | I didn't |
| 21:29.17 | wacky__ | did it solve much things for others ? |
| 21:29.22 | seanbright | ~gsmbug |
| 21:29.23 | jbot | [~gsmbug] there is a bug compiling Asterisk with GCC 4.2 where optimization errors cause GSM transcoding to distort heavily. Compile with GCC 4.1 or lower, or follow this patch : http://bugs.digium.com/view.php?id=11243 Also please read : http://forums.digium.com/viewtopic.php?p=116731&sid=3c65e49ec840380ed20f1c8426382b39 |
| 21:29.28 | wacky__ | ok, I'll try it out ! :) thanks for the tip |
| 21:29.36 | wacky__ | seanbright: yeah, I've been pointed to this already |
| 21:29.40 | kpfleming | hasn't that been fixed already? |
| 21:29.47 | seanbright | shrugs |
| 21:30.00 | wacky__ | ok I'll have a look at compiling it without optimization |
| 21:30.06 | wacky__ | thanks |
| 21:30.06 | *** part/#asterisk-dev wacky__ (n=abourget@IP-208-88-110-46.mtl.fibrenoire.ca) |
| 21:30.46 | jameswf | kpfleming: from my understanding in 1.4.22+ but the symptoms are close enough that a 2 minute rebuild isnt going to hurt |
| 21:31.27 | kpfleming | true |
| 21:31.34 | kpfleming | unless older gcc is hard to get your hands on |
| 21:38.06 | snuff-work | kpfleming: i will get around to re-testing that alignment branch soon i hope.. |
| 21:38.12 | ctooley | How do you tell if RTP has successfully been reinvited away? |
| 21:38.37 | kpfleming | rtp debig |
| 21:38.41 | kpfleming | or debug |
| 21:39.17 | russellb | or put your ear to the cable and see if you can hear it coming into the server or not |
| 21:39.22 | ctooley | right, so if I don't see anything, it's not there. |
| 21:39.49 | seanbright | russellb: does that really work? |
| 21:40.19 | russellb | of course, would i lie to you? |
| 21:40.38 | seanbright | i suppose not |
| 21:40.45 | seanbright | runs off to trie |
| 21:40.48 | seanbright | try* |
| 21:42.12 | seanbright | hmmm. rtp must be broken on my setup then. |
| 21:42.27 | seanbright | anywayz. |
| 21:43.05 | kpfleming | seanbright: you probably don't have certified SIP-5e cables |
| 21:43.48 | seanbright | can i buy them by typing /server SIP-5e cables? |
| 21:43.55 | seanbright | :) |
| 21:44.07 | Qwell | seanbright: You can only get them through certified resellers. |
| 21:44.13 | Qwell | Qwell Communications happens to be one |
| 21:44.44 | seanbright | i'll have my people get in touch |
| 21:47.17 | russellb | someone should call random voip equipment resellers about trying to buy SIP cables and see how they react |
| 21:47.19 | russellb | and record it |
| 21:47.25 | russellb | so that I can be amused |
| 21:48.24 | seanbright | or better yet, start selling SIP cables and make bank |
| 21:49.09 | russellb | sounds like a great idea for your next business venture |
| 21:49.11 | russellb | let me know how that goes |
| 21:49.41 | seanbright | gold plated rj45s |
| 21:49.43 | file | SIP Certified! |
| 21:50.11 | putnopvut | So, phillip64 and I have been pm'ing, and it seems his issue is that Asterisk is putting a private IP address in the c= line of the SDP it sends, when it should be public. |
| 21:50.13 | russellb | certified to reliably transmit SIP traffic at speeds up to 1000 Mbps |
| 21:50.29 | putnopvut | I'm not sure about NAT settings in sip.conf, and so I'm wondering if that is something that is controllable there. |
| 21:50.44 | russellb | putnopvut: externip=<externalip> set? |
| 21:51.00 | file | along with localnet |
| 21:51.11 | seanbright | and he should be using sip certified cables |
| 21:51.16 | russellb | clearly |
| 21:51.19 | seanbright | from BrightTech |
| 21:51.43 | putnopvut | russellb: I can check. He's got a meeting right now. |
| 21:54.03 | snuff-work | mmm.. |
| 21:54.07 | snuff-work | M13645 |
| 21:54.11 | MuffinMan | [acknowledged] [Asterisk] Channels/chan_iax2 0013645: chan_iax2 isn't using HANGUP anymore? reported by dzajro (Karma: neutral) http://bugs.digium.com/view.php?id=13645 |
| 21:54.17 | snuff-work | any use me responding to it.. |
| 21:54.34 | snuff-work | i dont see how its that hard for a guy to click like 3 buttons to send in an agreement |
| 21:55.27 | snuff-work | i could understand if u still had to fax something etc.. |
| 21:59.19 | putnopvut | snuff-work: the reporter seems like he's bike shedding right now. |
| 21:59.33 | putnopvut | I'll put a note up. |
| 22:00.09 | snuff-work | k thanks |
| 22:01.09 | russellb | snuff-work: you did the right thing, thanks |
| 22:07.44 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 22:07.44 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 22:09.13 | putnopvut | snuff-work: I left a note. No telling if it will actually help. |
| 22:10.07 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 22:13.43 | *** join/#asterisk-dev schmooze (n=schmooze@rrcs-67-52-214-19.west.biz.rr.com) |
| 22:15.10 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 22:29.36 | putnopvut | kpfleming: I know that you made a change to datastores recently such that they aren't restricted to being on just channels. Did this change make it into 1.4, or was that a trunk change? |
| 22:29.55 | kpfleming | i don't believe it was in 1.4 |
| 22:30.03 | kpfleming | it was for manager sessions having datastores |
| 22:30.10 | putnopvut | kpfleming: thanks |
| 22:37.28 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-179-75-202.static.bezeqint.net) |
| 22:37.28 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 22:39.03 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 22:48.26 | *** part/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 22:49.44 | *** join/#asterisk-dev DarylVOIP (n=daryl@75.147.121.177) |
| 22:55.57 | *** join/#asterisk-dev anthm][ (n=anthm@99-205-234-209.area4.spcsdns.net) |
| 22:56.38 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 22:59.27 | snuff-work | heh.. the guy now posted the patch to the dev mailing list |
| 22:59.51 | Qwell | ugh |
| 23:05.32 | putnopvut | beats his head on his desk |
| 23:28.12 | jameswf | off the top of your headsdid the 1.6 rx's or betas have any anger issues or bugs involving em_w |
| 23:28.24 | jameswf | *rc |
| 23:38.30 | putnopvut | anger issues? |
| 23:45.34 | seanbright | has anger issues |
| 23:46.36 | putnopvut | seanbright: you should be a pro wrestler |
| 23:50.58 | putnopvut | seanbright: lol, I just saw your reply to Anton on the -dev list |
| 23:53.54 | *** join/#asterisk-dev telecos (n=sergio@153.166.219.87.dynamic.jazztel.es) |