| 00:23.51 | *** join/#asterisk-dev suneye (~atcmmi@119.122.152.56) |
| 00:27.16 | *** part/#asterisk-dev deviantlinux (~deviant@unaffiliated/deviantlinux) |
| 00:43.11 | *** join/#asterisk-dev zerohalo (~zerohalo@74.61.196.236) |
| 03:47.02 | *** part/#asterisk-dev w9sh (~sph@50-79-224-193-static.hfc.comcastbusiness.net) |
| 06:23.53 | *** join/#asterisk-dev bulkorok (~chatzilla@85.183.36.36) |
| 06:41.59 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@local.xorcom.com) |
| 06:42.00 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 06:48.39 | *** join/#asterisk-dev bulkorok (~chatzilla@85.183.36.36) |
| 07:02.31 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 07:02.35 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 07:10.52 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:d45b:1278:8e1e:2289) |
| 07:39.39 | *** join/#asterisk-dev snuff-work (~snuffy@210.8.167.13) |
| 07:39.39 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 07:39.39 | *** join/#asterisk-dev wedhorn (~wedhorn@123-243-209-34.static.tpgi.com.au) |
| 07:39.39 | *** mode/#asterisk-dev [+oo snuff-work snuff-home] by leguin.freenode.net |
| 11:21.23 | *** join/#asterisk-dev oej (~Adium@host-95-195-149-9.mobileonline.telia.com) |
| 11:21.23 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 11:33.15 | *** join/#asterisk-dev oej (~Adium@2001:16d8:cc57:1000::42:1004) |
| 11:33.16 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 11:41.53 | *** join/#asterisk-dev italorossi (~italoross@187.60.66.11) |
| 12:00.27 | *** join/#asterisk-dev w9sh (~sph@50-79-224-193-static.hfc.comcastbusiness.net) |
| 12:18.52 | seanbright | can someone give me access to the pimp_my_sip testsuite branch in svn? |
| 12:21.47 | *** join/#asterisk-dev leifmadsen (~Leif@CPE000024c29e3a-CM602ad0881eae.cpe.net.cable.rogers.com) |
| 12:27.13 | opticron | seanbright, is it not public? |
| 12:27.20 | seanbright | my commit failed |
| 12:27.23 | seanbright | maybe my cert died |
| 12:27.23 | opticron | hmmm |
| 12:28.01 | opticron | if you can still check things out, I wouldn't expect that to be a problem |
| 12:28.09 | seanbright | you make a good point |
| 12:28.12 | opticron | could just be a permissions issue |
| 12:28.33 | opticron | I know I've committed to it before |
| 12:28.52 | opticron | we'll have to wait for more people to get in |
| 12:39.16 | oej | afternoon! |
| 12:55.55 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-yzcfkfoenytkczjh) |
| 13:05.26 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 13:05.27 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 13:21.15 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 13:21.16 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 13:25.54 | *** join/#asterisk-dev Tim_Toady (~fuzzy@snf-33276.vm.okeanos.grnet.gr) |
| 13:26.00 | *** join/#asterisk-dev dwayne (~dwayne@c-71-207-208-112.hsd1.al.comcast.net) |
| 13:35.13 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 13:35.13 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 13:39.18 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-rkmkdcoyeykhqyfj) |
| 13:39.18 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 13:39.32 | file | ZOMBIES |
| 13:39.56 | file | #define PJ_EINVALIDOP (PJ_ERRNO_START_STATUS + 13)/* 70013 */ |
| 13:40.26 | seanbright | hmm |
| 13:40.36 | seanbright | it's probably something wrong with this machine |
| 13:40.52 | file | seanbright, returned if the message has a to tag OR if the request is not dialog establishing |
| 13:41.05 | putnopvut | seanbright: can you double-check that all res_sip_* modules loaded correctly on startup? |
| 13:41.24 | putnopvut | file: it can also happen for things like if the pjsip_endpoint can't be found. |
| 13:41.45 | putnopvut | That's why I'm wondering if res_sip.so didn't load properly |
| 13:41.53 | file | putnopvut, well if the message is getting in to pre_session_setup... |
| 13:42.02 | seanbright | there's a bunch of warnings/errors from config_options.c |
| 13:42.06 | seanbright | i assume those are irrelevant |
| 13:42.17 | putnopvut | Yeah, that's because of missing documentation |
| 13:42.45 | seanbright | is there a pastebinit for centos? |
| 13:42.54 | seanbright | i'll just post the whole log |
| 13:43.00 | putnopvut | MKAY |
| 13:45.41 | seanbright | http://pastebin.com/raw.php?i=P278gdzi |
| 13:48.17 | putnopvut | seanbright: can you do it again with core debug enabled as well and set to, say, level 3? |
| 13:48.28 | seanbright | how do i do that with the testsuite? |
| 13:49.05 | putnopvut | Does the testsuite use your asterisk.conf file or does it provide its own? |
| 13:49.17 | seanbright | looks around |
| 13:49.17 | seanbright | me? |
| 13:49.18 | seanbright | i dunno |
| 13:49.22 | seanbright | it's your testsuite |
| 13:49.22 | seanbright | heh |
| 13:49.24 | putnopvut | heh |
| 13:49.37 | putnopvut | seanbright: The question was posed to anyone listening :) |
| 13:50.15 | seanbright | let me try updating my asterisk.conf |
| 13:50.38 | putnopvut | Well, the problem is it appears that there's a logger.conf file that the testsuite uses by default. |
| 13:50.59 | putnopvut | And it puts everything into the "full" file, including debug, so you're good. |
| 13:51.02 | seanbright | i have a feeling that the testsuite setup instructions are misleading |
| 13:51.09 | seanbright | it says not to 'make samples' |
| 13:51.12 | seanbright | which i have not done |
| 13:53.30 | *** join/#asterisk-dev kharwell (~kharwell@nat/digium/x-xxxgyrgryjkngsix) |
| 13:54.04 | seanbright | putnopvut: http://pastebin.com/raw.php?i=1e4SAV2q |
| 13:56.39 | putnopvut | looks |
| 13:57.07 | seanbright | i apologize in advance for what is surely a misconfiguration on my part |
| 13:57.20 | putnopvut | ugh, well that didn't help. Sorry to make you have to do it a second time. |
| 13:57.25 | seanbright | awesome |
| 13:57.40 | seanbright | maybe centos is locking something down that it shouldn't |
| 13:57.46 | seanbright | i turned off selinux |
| 13:58.00 | snuff-home | putnopvut: thanks for the review.. i hope to do some work in fixing ppls issues in the next 48hrs |
| 13:58.28 | putnopvut | snuff-home: yeah. Sorry if I came across as harsh or anything. With user documentation I have to be a stickler though :) |
| 13:58.59 | snuff-home | well its my random dump by looking @ code only.. not fully understanding all of SIP stuff.. |
| 13:59.48 | snuff-home | that 'invite/reinvite' stuff you mention.. i was only looking at the code options to decide what to put in the enum |
| 13:59.52 | snuff-home | doc |
| 14:00.16 | seanbright | putnopvut: are these lines relevant at all? |
| 14:00.17 | seanbright | [May 14 09:52:46] ERROR[20095] config_options.c: Could not find option suitable for category 'bob' named 'host' at line 29 of |
| 14:00.20 | seanbright | [May 14 09:52:46] DEBUG[20095] res_sorcery_config.c: Could not create an object of type 'endpoint' with id 'bob' from configuration file 'res_sip.conf' |
| 14:00.21 | putnopvut | seanbright: PJ_EINVALIDOP is an annoying because it seems like there are about 1039014 ways that might get returned. The dialog-creation-specific reasons clearly are not why it happened though. |
| 14:01.08 | putnopvut | seanbright: well that means that endpoint bob doesn't exist. However, if an endpoint doesn't exist, we should be responding with a 403 for INVITEs from that endpoint. |
| 14:01.18 | putnopvut | We shouldn't reach dialog creation at all. |
| 14:01.36 | file | the messages aren't looking from bob, either |
| 14:01.42 | file | looking from? aren't from |
| 14:01.45 | putnopvut | right |
| 14:01.48 | seanbright | there are similar ones for alice |
| 14:02.09 | file | it found alice, so she's fine |
| 14:02.11 | putnopvut | But I see in the log you posted that the endpoint identifier was able to tell the incoming message was from alice |
| 14:02.16 | snuff-home | mm... my parallella is getting close to being shipped.. yey |
| 14:02.19 | seanbright | actually, no there aren't |
| 14:02.26 | seanbright | ok |
| 14:02.53 | seanbright | i can add logging to _create_uas i guess |
| 14:03.15 | file | seanbright, did you try it outside the testsuite? |
| 14:03.35 | seanbright | file: not yet, no. |
| 14:03.39 | file | ack |
| 14:03.49 | seanbright | but not to be a dick... shouldn't it work in the testsuite? |
| 14:04.08 | seanbright | who am i kidding... that was totally to be a dick |
| 14:04.15 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 14:04.15 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 14:04.18 | file | it should |
| 14:04.32 | file | and it does, for others |
| 14:04.39 | seanbright | i will set up outside the testsuite |
| 14:04.55 | file | so something specific is hitting you and isolating where exactly is useful |
| 14:05.18 | file | if it's testsuite specific (be it config or something) fine, if it's Asterisk side also fine |
| 14:10.58 | seanbright | file: crashes outside of the testsuite |
| 14:11.25 | file | well that's annoying |
| 14:11.37 | file | what version of CentOS? |
| 14:11.41 | seanbright | 6.4 |
| 14:11.45 | seanbright | 64bit |
| 14:12.33 | putnopvut | ffffffuuuuuuuuuu |
| 14:13.06 | file | stares at putnopvut |
| 14:13.49 | *** join/#asterisk-dev leedm777 (~leedm777@nat/digium/x-xmlqaevmnvarwybx) |
| 14:14.00 | seanbright | will PJ_LOG propogate to res_sip_logger? |
| 14:14.13 | file | no, it only logs SIP messages |
| 14:14.25 | seanbright | alrighty |
| 14:14.27 | file | the pj logger hasn't been redirected, yet |
| 14:15.09 | seanbright | i guess i could run in gdb and break in pjsip_dlg_create_uas |
| 14:19.05 | file | following the white rabbit... deeper and deeper it goes |
| 14:21.06 | *** join/#asterisk-dev hehol (~hehol@217.9.101.222) |
| 14:31.08 | *** join/#asterisk-dev oej (~Adium@2001:16d8:cc57:1000::42:1004) |
| 14:31.09 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 14:35.38 | seanbright | if (!endpt) |
| 14:35.46 | putnopvut | Wow |
| 14:35.49 | file | ...wow |
| 14:35.55 | putnopvut | in create_dialog() ? |
| 14:35.58 | seanbright | aye |
| 14:36.15 | *** join/#asterisk-dev jrose_atDigium (~jrose_atD@nat/digium/x-zugcctinqjazlfbr) |
| 14:36.26 | putnopvut | Hm, I wonder if the pjsip ua layer didn't register properly or something... |
| 14:36.35 | putnopvut | rereads logs |
| 14:37.07 | file | silly us, we don't check the return value |
| 14:37.18 | seanbright | i'm using pjproject trunk, if that is relevant |
| 14:38.41 | *** join/#asterisk-dev leedm777 (~leedm777@nat/digium/x-kafaadvbfowokmdq) |
| 14:39.12 | seanbright | pjsip_ua_get_endpt: mod_ua.endpt is NULL |
| 14:39.24 | seanbright | (obviously) |
| 14:40.09 | *** join/#asterisk-dev saghul (~saghul@ip3e830637.speed.planet.nl) |
| 14:40.46 | putnopvut | pjproject trunk? As in from pjproject.org? |
| 14:40.51 | seanbright | yes |
| 14:40.53 | putnopvut | or pjsip.org |
| 14:40.56 | seanbright | um |
| 14:40.59 | putnopvut | Yeah, that could be a problem. |
| 14:41.06 | putnopvut | sorry, I just got the name wrong earlier |
| 14:41.13 | seanbright | http://svn.pjsip.org/repos/pjproject/trunk |
| 14:41.41 | putnopvut | So, Qwell made some modifications to pjsip so that their libraries could be made into shared objects and be dynamically linked. |
| 14:42.03 | putnopvut | That version of pjproject is in a git repo at github.org/asterisk |
| 14:42.06 | Qwell | https://github.com/asterisk/pjproject |
| 14:42.12 | putnopvut | That's the version we've all been using. |
| 14:42.13 | snuff-home | u can't compile asterisk without adding -fPIC when compiling pjsip |
| 14:42.21 | seanbright | i did add -fPIC |
| 14:42.26 | seanbright | i'll try with the github version |
| 14:42.26 | putnopvut | I haven't had that problem before. |
| 14:42.28 | snuff-home | mm that's all i did |
| 14:42.45 | snuff-home | it worked last time i tried.. long enough about |
| 14:42.45 | snuff-home | *ago |
| 14:42.49 | snuff-home | and now sleep time |
| 14:43.08 | putnopvut | seanbright: you'll need to be sure to get rid of ALL traces of pjproject from your system before installing the git version. |
| 14:43.08 | putnopvut | Also, be sure to pass --enable-shared and --prefix=/usr to the pjproject configure script |
| 14:43.14 | *** join/#asterisk-dev opticron (~opticron@pianoben.ch) |
| 14:43.15 | *** mode/#asterisk-dev [+o opticron] by ChanServ |
| 14:43.21 | putnopvut | snuff-home: were you able to make calls successfully? |
| 14:43.22 | mjordan | methinks we should probably write these instructions down |
| 14:43.24 | Qwell | --prefix=/usr isn't needed, if your distro isn't dumb |
| 14:43.28 | putnopvut | heh |
| 14:43.29 | Qwell | mjordan: It's on the list |
| 14:43.36 | seanbright | Qwell: centos 6.4 "dumb?" |
| 14:43.38 | putnopvut | Qwell: is centos 6.4 dub? |
| 14:43.40 | Qwell | so is pushing that stuff back... |
| 14:43.55 | putnopvut | dub? dumb. |
| 14:43.56 | Qwell | putnopvut: It drops much bass. |
| 14:44.15 | Qwell | but, yes, CentOS is dumb, because /lib64 |
| 14:44.33 | file | we should bump it up the list |
| 14:44.35 | seanbright | with the github branch, is -fPIC required? i assume no |
| 14:44.39 | Qwell | seanbright: no |
| 14:44.43 | seanbright | so: ./configure --enable-shared --prefix=/usr |
| 14:45.05 | Qwell | seanbright: I also disable a bunch of extra crap, like resample and codecs |
| 14:45.17 | Qwell | It should mostly work without those though. |
| 14:45.28 | seanbright | do you have your configure arguments laying around? |
| 14:45.41 | seanbright | maybe on the right hand side of your desk? under some paperwork |
| 14:46.04 | file | Qwell, don't you have packages? |
| 14:46.05 | Qwell | --with-external-speex --with-external-gsm --with-external-srtp --disable-sound --disable-resample |
| 14:46.09 | Qwell | file: I DO! |
| 14:46.11 | Qwell | actually |
| 14:46.31 | seanbright | that'd be even better |
| 14:46.44 | Qwell | 64-bit? |
| 14:46.47 | seanbright | aye |
| 14:47.03 | seanbright | so this is my fault |
| 14:47.11 | file | our fault for not having it documented |
| 14:47.17 | Qwell | http://packages.asterisk.org/centos/6/current/i386/RPMS/pjproject-2.1-0.digium1.2_centos6.i686.rpm http://packages.asterisk.org/centos/6/current/i386/RPMS/libsrtp-1.4.4-4.20101004cvs.el6.i686.rpm http://packages.asterisk.org/centos/6/current/i386/RPMS/libsrtp-devel-1.4.4-4.20101004cvs.el6.i686.rpm |
| 14:47.20 | Qwell | grab those 3 |
| 14:47.34 | seanbright | i now realize that the testsuite docs on the wiki don't reflect any of the pjsip stuff |
| 14:49.06 | Qwell | actually, this too http://packages.asterisk.org/centos/6/current/i386/RPMS/pjproject-devel-2.1-0.digium1.2_centos6.i686.rpm |
| 14:49.07 | Qwell | umm |
| 14:49.07 | Qwell | except, s/i386/x86_64/ all those |
| 14:49.07 | seanbright | and those docs actually have you (partially) install pjsip from pjsip.org |
| 14:49.08 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 14:49.09 | *** mode/#asterisk-dev [+o malcolmd] by leguin.freenode.net |
| 14:49.39 | file | some of the tests use pjsua |
| 14:49.46 | seanbright | yeah |
| 14:50.05 | seanbright | so i'm to install pjsip from github for asterisk, and pjsua for testsuite from pjsip.org? |
| 14:50.31 | seanbright | the wiki just needs updating. nbd. |
| 14:55.00 | seanbright | rebuilding asterisk now... stay tuned for the exciting conclusion |
| 14:55.35 | seanbright | annnd it works |
| 14:55.50 | seanbright | file: putnopvut: sorry for wasting your time guys |
| 14:55.58 | file | it's fine, it's our fault for the documentation |
| 14:56.10 | file | and that's actually useful because we now know what would happen under that scenario |
| 14:56.25 | putnopvut | Yeah, though I can't accurately explain *why* |
| 14:56.32 | file | I can |
| 14:56.59 | file | each module had its own complete independent copy of pjproject loaded into memory |
| 14:57.09 | Qwell | none of which had ever been inited |
| 14:57.10 | Qwell | yay! |
| 14:57.15 | file | well, one had |
| 14:57.20 | file | the res_sip copy |
| 14:57.20 | Qwell | well, 2 |
| 14:57.23 | Qwell | BUT, yeah |
| 14:57.29 | seanbright | annnnd the testsuite now passes |
| 14:57.33 | seanbright | huzzah |
| 14:57.41 | file | but when res_sip_session tried to access stuff it was accessing its own uninitialized copy |
| 14:57.43 | Qwell | file: res_rtp_asterisk :D |
| 14:58.05 | file | Qwell, well at pjlib level SURE! but not pjsip :P |
| 14:58.40 | Qwell | Don't poke holes in my hole poking. |
| 14:58.43 | Qwell | glares |
| 15:16.26 | *** join/#asterisk-dev malcolmd_ (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 15:16.26 | *** mode/#asterisk-dev [+o malcolmd_] by ChanServ |
| 15:27.38 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-wkcakjxgcrinxylh) |
| 15:41.33 | *** part/#asterisk-dev leedm777 (~leedm777@nat/digium/x-kafaadvbfowokmdq) |
| 15:47.20 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@bzq-218-155-170.cablep.bezeqint.net) |
| 15:47.20 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 15:53.50 | *** part/#asterisk-dev oej (~Adium@2001:16d8:cc57:1000::42:1004) |
| 16:07.16 | *** join/#asterisk-dev serafie1 (~erin@nat/digium/x-ooifivbpcmbgfrvh) |
| 16:15.50 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 16:15.50 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 17:15.58 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 17:15.58 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 17:42.15 | seanbright | mjordan: can you give me write access to the pimp_my_sip testsuite? |
| 17:42.21 | seanbright | commit access, i should say |
| 17:45.24 | mjordan | Qwell: ^^^ |
| 17:45.32 | seanbright | Qwell: can you give me write access to the pimp_my_sip testsuite? |
| 17:45.35 | seanbright | commit access, i should say |
| 17:45.41 | mjordan | seanbright: Qwell is the keeper of the keys. I merely route information |
| 17:45.47 | seanbright | roger that |
| 17:45.55 | mjordan | is a glorified Cisco router |
| 17:46.16 | seanbright | over valued? |
| 17:46.16 | file | pfft |
| 17:46.33 | seanbright | (i kid) |
| 17:46.33 | Qwell | What is the path? |
| 17:46.40 | seanbright | /svn/testsuite/asterisk/team/group/pimp_my_sip |
| 17:46.55 | mjordan | seanbright: and liable to brick when you least expect it |
| 17:46.56 | Qwell | done probably |
| 17:47.02 | seanbright | asdasd |
| 17:47.06 | seanbright | gah |
| 17:47.10 | seanbright | sorry about that |
| 17:47.29 | seanbright | Qwell: thanks, works. |
| 17:57.41 | *** join/#asterisk-dev WIMPy (~wimpy@e183095026.adsl.alicedsl.de) |
| 17:59.52 | *** join/#asterisk-dev oej (~Adium@2001:16d8:cc57:1000::42:1005) |
| 17:59.52 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:30.13 | seanbright | file: any objection to my adding ast_sip_is_content_type? |
| 18:30.20 | seanbright | to res_sip.h |
| 18:30.26 | file | nope |
| 18:30.37 | file | just make sure you put it in the res_sip.exports.in file |
| 18:30.44 | seanbright | yeah, it's there |
| 18:32.33 | *** join/#asterisk-dev philipp64 (~chatzilla@66-232-81-197.ctcweb.net) |
| 18:33.42 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 18:33.42 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:35.15 | Kobaz | does it make sense to see multiple INVITEs during a dialog when you don't have canreinvite enabled? |
| 18:35.24 | file | seanbright, very nice |
| 18:36.20 | file | Kobaz, it's possible to receive multiple INVITEs, yes |
| 18:36.36 | seanbright | session expiration sends an INVITE i believe |
| 18:36.44 | file | yeah it can |
| 18:36.47 | file | session timers |
| 19:04.47 | *** join/#asterisk-dev maxhbp204 (~chatzilla@115.248.240.58) |
| 19:05.41 | maxhbp204 | Hi, how can i pass argument to pass on mailcmd command in voicemail.conf, actually i have added mp3 conversion script in mailcmd and i need to put mailbox number in script, but i am not getting how can i put there, I have tried to put ${VM_MAILBOX}, but it is coming empty on script, can anybody help on this, i have pass like this mailcmd=/path/script ${VM_MAILBOX}, but it is not working and... |
| 19:05.42 | maxhbp204 | ...it is coming as blank, so can anybody help me on this please |
| 19:10.53 | *** part/#asterisk-dev maxhbp204 (~chatzilla@115.248.240.58) |
| 19:16.03 | seanbright | my god. app_voicemail |
| 19:16.31 | seanbright | mkdir res/res_vm |
| 19:16.35 | seanbright | sorry, wrong window |
| 19:30.26 | Kobaz | ah |
| 19:30.59 | Kobaz | file: i have this call with just like invite crazyness |
| 19:32.19 | seanbright | Kobaz: pastebin the trace |
| 19:32.26 | seanbright | assuming it doesn't have private info in it |
| 19:33.07 | Kobaz | www.kobaz.net/misc/tmp/holio |
| 19:33.17 | Kobaz | it's a pdml file for wireshark |
| 19:35.52 | seanbright | my wireshark won't open it |
| 19:35.58 | seanbright | can you export as a normal pcap? |
| 19:43.29 | Kobaz | nope |
| 19:43.32 | seanbright | k |
| 19:43.33 | Kobaz | that's not an option it gives |
| 19:43.41 | seanbright | what is "it?" |
| 19:44.07 | Kobaz | i did |
| 19:44.11 | Kobaz | export as PDML XML |
| 19:44.18 | seanbright | out of what tool? |
| 19:44.22 | Kobaz | wireshark |
| 19:44.52 | seanbright | you can do a File -> Save As |
| 19:45.01 | Kobaz | oh |
| 19:45.07 | Kobaz | you can do selected packets with save as |
| 19:45.09 | Kobaz | didnt know that |
| 19:45.10 | seanbright | yeah |
| 19:45.42 | Kobaz | okay |
| 19:45.45 | Kobaz | uploaded to the same file |
| 19:50.59 | seanbright | weird |
| 19:51.23 | seanbright | the only thing changing is the version number in the sdp owner attribute |
| 19:51.35 | Kobaz | and wireshark says it's not a retransmit |
| 19:51.45 | seanbright | yeah, it's not |
| 19:53.34 | seanbright | in short: i can't explain it, but it's not wrong |
| 19:53.43 | Kobaz | heh |
| 19:53.49 | seanbright | the media src/dest never changes is |
| 19:53.57 | seanbright | s/is// |
| 19:54.45 | seanbright | the only thing changing is the SDP's owner version and the branch tag in the via header |
| 19:55.19 | seanbright | what is your concern? |
| 19:55.53 | Kobaz | i've been getting complaints of people calling in and getting a prerecorded track 'this number has been disconnected' |
| 19:55.59 | Kobaz | so i found that call in my research |
| 19:56.14 | Kobaz | the guy apparently called in at 1:22 and couldnt get through and then i see the 1:23 call which has all that |
| 19:56.15 | seanbright | do you have the RTP captured as well or just the SIP? |
| 19:56.20 | Kobaz | but from the asterisk log it looks fine |
| 19:56.29 | Kobaz | like he's hitting buttons in the ivr and dialing extensions |
| 19:56.36 | Kobaz | but i was just wondering maybe something funny was going on |
| 19:56.59 | seanbright | i'm certainly no sip expert, but it looks fine to me |
| 19:57.03 | Kobaz | k |
| 19:57.16 | Kobaz | i think the issue is that the call his the adtran and the adtran returns busy for some reason |
| 19:57.18 | Kobaz | or something like that |
| 19:57.29 | Kobaz | i'm dumping the pri debugs all day so i'll wait for the next report |
| 19:58.02 | seanbright | what are you capturing the packets with? |
| 19:58.08 | Kobaz | wireshark for sip |
| 19:58.16 | Kobaz | and doing isdn l2-formatted for debug on the adtran |
| 19:58.39 | seanbright | there is a utility i use for debugging called pcapsipdump |
| 19:58.48 | seanbright | it splits each call out into it's own pcap file |
| 19:59.23 | seanbright | and you can specify the type of rtp loggins as well (rtp, rtp+rtcp, or none) |
| 19:59.24 | Kobaz | ah cool |
| 20:04.26 | Kobaz | hmm |
| 20:04.45 | Kobaz | i have a call recording here that's like partially mixed |
| 20:06.16 | seanbright | i believe the plan is to drop call recording in asterisk |
| 20:06.20 | seanbright | heh |
| 20:07.33 | Kobaz | really? |
| 20:11.58 | Kobaz | i should probably start recording to split files |
| 20:23.03 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 20:23.03 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 20:26.47 | Kobaz | this recording is really weird |
| 20:27.28 | Kobaz | like person A got transferred to B... at first the recording contains A talking and then B talking at a very very low volume, sounds like he's across a tunnel |
| 20:27.46 | Kobaz | and then after that all you hear is A.. and when B talks the whole stream is silent |
| 20:28.02 | Kobaz | when B stops talking, then A comes back |
| 20:30.57 | Kobaz | any idea what would cause that? |
| 20:31.04 | Kobaz | and this is MixMonitor |
| 20:35.36 | seanbright | yes, recordings suck |
| 20:35.40 | seanbright | that is the problem |
| 20:35.55 | *** join/#asterisk-dev jsmith (~jsmith@fedora/jsmith) |
| 20:35.55 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 20:36.40 | seanbright | rtp packets are not received at precise 20ms intervals |
| 20:36.51 | seanbright | jitterbuffers on endpoints take care of this |
| 20:37.29 | seanbright | and i don't believe audiohooks does any JBing |
| 20:38.07 | seanbright | we use mixmonitor, and haven't had any issues with it for a while |
| 20:38.11 | seanbright | we're still on 1.8 though |
| 20:38.26 | Kobaz | yeah |
| 20:38.28 | Kobaz | this is 11 |
| 20:43.20 | mjordan | 11 actually makes it even easier then 1.8, as you have func_jitterbuffer |
| 20:44.02 | Kobaz | so should i bump up the jitter buffer? |
| 20:44.19 | mjordan | well, first, what channel has the jitter buffer? And how was it applied? |
| 20:44.56 | Kobaz | i'm not doing anything with jitter buffers |
| 20:45.00 | mjordan | ah, okay. |
| 20:45.16 | Kobaz | i was wondering if ussing the jitterbuffer func would fix my issue |
| 20:45.17 | mjordan | So, I'd try putting a jitter buffer on the SIP channel using the JITTERBUFFER function |
| 20:45.20 | mjordan | it may |
| 20:45.32 | mjordan | as seanbright pointed out, if your RTP has drops then your recordings will get out of sync |
| 20:45.39 | mjordan | or at least, can get out of sync |
| 20:46.18 | Kobaz | it wasn't out of sync |
| 20:46.21 | Kobaz | it was like. gone |
| 20:46.38 | Kobaz | and when both sides were available the B side was extremely quiet |
| 20:47.02 | mjordan | doesn't sound like Asterisk is your problem |
| 20:47.23 | Kobaz | oh? |
| 20:47.25 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 20:47.35 | mjordan | it's not going to much the volume of a channel up (unless you told it to) |
| 20:47.39 | mjordan | muck |
| 20:47.42 | Kobaz | yeah i know that |
| 20:47.53 | Kobaz | but why wouldn't it be an asterisk problem? |
| 20:48.02 | mjordan | if you have drops in audio, now that could be something that a jitter buffer would help with, but a pcap of the rtp would tell you if that was actually happening |
| 20:48.03 | Kobaz | there's like, zero audio after the first few seconds from the B side |
| 20:48.16 | mjordan | k, if there's no audio, what is Asterisk supposed to do? |
| 20:48.17 | Kobaz | but apparently the A person could hear B just fine |
| 20:48.24 | Kobaz | since they were talking back and fourth |
| 20:48.36 | mjordan | did you have directmedia set up? |
| 20:48.38 | Kobaz | nope |
| 20:48.52 | mjordan | k, well, I'm out of guesses. That doesn't sound like anything I've heard before. |
| 20:49.02 | putnopvut | Even if you set directmedia up, the presence of a monitor or audiohooks would prevent direct media from happening. |
| 20:49.03 | Kobaz | like it was a normal paced conversation, but one side was silent |
| 20:49.05 | mjordan | audiohooks aren't going to do much other than get an audio frame and write it to a file |
| 20:49.42 | mjordan | putnopvut: correct. I was thinking that the presence of the audiohook occurred after the initial bridge was set, but I'm not sure how you'd get it to that point unless yo uused some interception macro (and I'm thinking that it would still happen before the bridge) |
| 20:50.52 | wedhorn | mjordan: happy for you close ASTERISK-21313 |
| 20:51.15 | mjordan | wedhorn: woot |
| 20:51.40 | mjordan | wedhorn: thanks for looking into that |
| 20:52.27 | wedhorn | no worries, also noticed your bug about skinny transfer. Assume that's just to make sure it gets done when the attended transfer stuff gets into bridging? |
| 20:53.09 | putnopvut | wedhorn: yeah, I'll be sending you an e-mail about that in the next couple of days |
| 20:53.10 | wedhorn | damn that was quick |
| 20:53.41 | wedhorn | putnopvut: okay, should be relatively easy, just would be nice to be able to test |
| 20:53.51 | putnopvut | Basically, we want chan_skinny to use the new transfer APIs in bridging, and I'm planning to get it converted. I'll be sending you an e-mail with my plans of what I'll be doing and such. |
| 20:54.19 | putnopvut | Right now, I'm just looking at chan_skinny to try to follow how transfers are performed so I don't screw anything up :) |
| 20:54.58 | wedhorn | okay, the guts should be similar to the mgcp patch (rb 2536) |
| 20:55.26 | putnopvut | All righty. |
| 20:55.50 | wedhorn | although we may want to separate blind and attended (which are currently dealt with in skinny_transfer) |
| 20:56.40 | putnopvut | Right. I'm still planning my attack for now. |
| 20:56.52 | wedhorn | mjordan: have you had a chance to look at the discussion in review 2393 |
| 20:57.37 | wedhorn | putnopvut: I'm happy to put something up (given that I wrote the original skinny transfer) |
| 20:57.39 | seanbright | has anyone compiled asterisk with gcc 4.7ish? |
| 20:57.40 | putnopvut | wedhorn: mjordan actually wanted me to have a look at that. Been super busy though. I'll set aside time on thursday for it. |
| 20:57.52 | wedhorn | cool thanks |
| 20:57.53 | seanbright | wondering if any of the auto-vectorization has made improvements |
| 20:57.58 | putnopvut | seanbright: I know we have people here who use 4.8 |
| 20:58.57 | mjordan | wedhorn: between me and putnopvut, one of will look at r2393 :-P |
| 20:59.11 | wedhorn | thanks |
| 20:59.18 | seanbright | putnopvut: i'd be interested in seeing the resulting binaries |
| 20:59.24 | seanbright | just throwing that out there |
| 20:59.54 | putnopvut | wedhorn: okay, if you wanted to put something up, that would be pretty amazing. I can give you tips on what needs to be done if you're curious. I had been planning to do it just because we were going to try to get all the remaining channel driver transfer conversions done in the next two weeks. |
| 21:00.13 | putnopvut | The thing is, most of the channel driver conversions consist of removing code more than anything. |
| 21:00.32 | putnopvut | The core will handle a lot for you that you don't need to worry about. |
| 21:01.04 | putnopvut | I think the bulk of the change in chan_skinny will be in the skinny_transfer() function since that's where the actual transfer appears to complete. |
| 21:01.09 | wedhorn | putnopvut: yep noticed that. Assuming that the chan_mgcp review is close to what's wanted, should be fairly easy to convert skinny |
| 21:02.16 | snuff-home | wedhorn: managed to crash skinny again.. via hitting cfwd on off.. haven't looked @ backtrace |
| 21:02.18 | putnopvut | Yep, chan_mgcp should be a fairly accurate model of what needs to be done. |
| 21:02.23 | wedhorn | putnopvut: the important bit in skinny is actually the function is called, especially given that it currently does it's own dialing and monitoring for a blind xfer |
| 21:02.29 | wedhorn | cool |
| 21:02.46 | wedhorn | I'll try to get something up in the next couple of days |
| 21:02.55 | putnopvut | The big difference is that we actually know of people that use chan_skinny, as opposed to chan_mgcp. |
| 21:02.56 | putnopvut | :) |
| 22:15.51 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-rkmkdcoyeykhqyfj) |
| 22:27.22 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 22:27.22 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |