| 03:55.56 | *** join/#asterisk-dev xpot-mobile (~xpot@70-91-210-233-BusName-Utah.hfc.comcastbusiness.net) |
| 07:16.36 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 07:16.36 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 07:31.37 | *** join/#asterisk-dev alecdavis (~alec@202-78-149-14.cable.telstraclear.net) |
| 07:32.04 | alecdavis | corydon76-dig: ping |
| 07:33.55 | Corydon76-dig | alecdavis: yawn |
| 07:40.21 | alecdavis | i thought so. |
| 07:40.29 | alecdavis | I just emailed you. |
| 07:41.33 | alecdavis | in essence, the voicemail change, is that trying to check INBOX and URGENT |
| 07:42.58 | Corydon76-dig | It's not a change, per se |
| 07:43.32 | Corydon76-dig | Rather, the point is to make it behave sanely |
| 07:45.28 | alecdavis | ok. but if the call to has_voicemail has a null 'folder' pointer, will it now not check the urgent folder? |
| 07:46.50 | Corydon76-dig | It will |
| 07:48.52 | alecdavis | you missed trunk, caught the 1.6.1 and 1.6.2 branches |
| 07:49.22 | alecdavis | it's only nearly 02:00 isn;t it? |
| 07:55.53 | Corydon76-dig | No, I did not miss trunk |
| 07:56.28 | Corydon76-dig | And yes, it's late, and I'm tired, so I'm going to bed. |
| 07:57.09 | alecdavis | opps, I had just checked out trunk svn and was looking in wrong area. |
| 07:58.53 | Corydon76-dig | Rather than assuming anything about how the code works, please test it. There's even a unit test for it |
| 08:01.23 | alecdavis | still segfaults trunk :( |
| 08:18.17 | alecdavis | M16927 |
| 08:18.19 | MuffinMan | [new] [Asterisk] Applications/app_voicemail 0016927: Segfault branches, and trunk, when DAHDI FXS port goes off hook reported by alecdavis https://issues.asterisk.org/view.php?id=16927 |
| 08:55.08 | *** join/#asterisk-dev codefreeze-lap (~murf@mail.parsetree.com) |
| 08:55.08 | *** mode/#asterisk-dev [+o codefreeze-lap] by ChanServ |
| 09:43.58 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-dgyuhelbeqwxqcsb) |
| 09:43.58 | *** join/#asterisk-dev angler (~angler@pdpc/sponsor/digium/angler) |
| 10:21.19 | *** part/#asterisk-dev alecdavis (~alec@202-78-149-14.cable.telstraclear.net) |
| 11:03.03 | *** join/#asterisk-dev Dovid (~annon@213.8.121.90) |
| 12:24.20 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:24.20 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:24.43 | *** join/#asterisk-dev codefreeze-lap (~murf@mail.parsetree.com) |
| 15:24.43 | *** mode/#asterisk-dev [+o codefreeze-lap] by ChanServ |
| 16:03.26 | *** join/#asterisk-dev alecdavis (~sivad@202-78-149-14.cable.telstraclear.net) |
| 16:06.31 | alecdavis | corydon76-dig: ping |
| 16:17.20 | *** join/#asterisk-dev atis_work (~atis_work@193.238.212.171) |
| 16:31.28 | *** join/#asterisk-dev korihor (~korihor@190.205.251.97) |
| 17:02.45 | alecdavis | how do you get a unit test to trap a segfault? |
| 17:09.44 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 17:09.44 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 17:17.06 | *** part/#asterisk-dev alecdavis (~sivad@202-78-149-14.cable.telstraclear.net) |
| 17:19.06 | *** join/#asterisk-dev ArtemMakhutov (~ArtemMakh@ip-95-223-6-41.unitymediagroup.de) |
| 17:19.57 | ArtemMakhutov | Hello |
| 17:23.11 | ArtemMakhutov | Is it sufficient to signal AST_CAUSE_USER_BUSY or must I also signal AST_CONTROL_BUSY to indicate a busy state of the remote end? |
| 18:06.24 | Corydon76-dig | You have to send a control frame. The cause is only valid within a HANGUP control frame |
| 18:17.25 | ArtemMakhutov | What is the difference between AST_CAUSE_NO_USER_RESPONSE and AST_CAUSE_NO_ANSWER? |
| 18:23.43 | Corydon76-dig | First one means that we did not get progress back from the remote peer |
| 18:24.45 | Corydon76-dig | Second one means that the ANSWER event never happened |
| 18:27.30 | ArtemMakhutov | ah, ok, thx |
| 18:33.43 | ArtemMakhutov | I am getting hangup-causes from the GSM network, which are pretty the same like the Q931 causes defined in include/asterisk/causes.h |
| 18:35.31 | ArtemMakhutov | Is it safe just to set the asterisk hangupcause to the GSM hangupcause, because there are some hangupcauses, that are not defined in include/asterisk/causes.h? |
| 19:16.49 | *** join/#asterisk-dev atis_work (~atis_work@193.238.212.171) |
| 19:25.39 | Kobaz | i gots some bugs |
| 19:26.34 | Kobaz | it looks like in late versions of asterisk 1.6.0, if you softhangup a local channel, that spawns other channels via dial()... it doesn't properly hang up those children channels |
| 19:27.37 | Corydon76-dig | You're going to have to be more specific and file an issue |
| 19:28.33 | Kobaz | yeah |
| 19:28.36 | Kobaz | i'm working on it |
| 19:37.22 | Kobaz | i think i'll just keep writing unit tests, and stick with 1.6.0.19 for production systems |
| 19:37.29 | Kobaz | lots of stuff is broken in later versions |
| 19:41.57 | *** join/#asterisk-dev alecdavis (~alec@202.36.76.21) |
| 19:49.31 | Corydon76-dig | Morning, Alec |
| 20:03.26 | *** join/#asterisk-dev pyite (~dschreibe@unaffiliated/pyite) |
| 20:16.07 | alecdavis | morning. |
| 20:16.32 | alecdavis | Have you seen the problem? |
| 20:17.07 | Corydon76-dig | The issue tracker is full of problems |
| 20:17.20 | *** join/#asterisk-dev atis_work (~atis_work@193.238.212.171) |
| 20:17.29 | alecdavis | M16927 |
| 20:17.31 | MuffinMan | [new] [Asterisk] Applications/app_voicemail 0016927: [patch] Segfault branches, and trunk, when DAHDI FXS port goes off hook reported by alecdavis https://issues.asterisk.org/view.php?id=16927 |
| 20:19.27 | alecdavis | SVN branches 1.6.1 and 1.6.2 and trunk all segfault, due to the app_voicemail sanity changes. |
| 20:21.10 | Kobaz | alecdavis: take a number and get in line please |
| 20:22.38 | Kobaz | alecdavis: i've seen voicemail crashes with 1.6.0.22 and up as well... they look related to the reference counting stuff that was recently added |
| 20:24.15 | Kobaz | er, no... wait |
| 20:24.23 | Kobaz | this is musiconhold |
| 20:24.27 | alecdavis | kobaz: umm.. |
| 20:24.45 | Kobaz | never mind :P |
| 20:27.33 | Corydon76-dig | alecdavis: when you break the unit tests, it becomes your responsibility to fix it |
| 20:29.30 | alecdavis | corydon76-dig: I changed and committed the app_voicemail unit test, to check for a NULL folder (instead of INBOX), which is how the code calls has_voicemail. |
| 20:30.15 | Corydon76-dig | alecdavis: Did the unit test succeed before you changed it? |
| 20:30.25 | alecdavis | yes it did. |
| 20:30.32 | Corydon76-dig | Did it fail after you changed it? |
| 20:30.40 | alecdavis | no. |
| 20:30.50 | Corydon76-dig | Really? |
| 20:30.52 | alecdavis | wrong, yes it does. |
| 20:31.02 | Corydon76-dig | Well, there you go. |
| 20:31.46 | alecdavis | #2 0x08068475 in ast_app_has_voicemail (mailbox=0xb684b98c "", folder=0x0) |
| 20:32.00 | alecdavis | notice folder is called with NULL |
| 20:32.32 | alecdavis | that is from the GDB traces reported in 16927. |
| 20:32.52 | Corydon76-dig | You still broke the unit test |
| 20:35.34 | alecdavis | how do we trap a segfault, in unit tests. |
| 20:39.22 | Corydon76-dig | You trap SIGSEGV with sigaction |
| 20:41.13 | alecdavis | A segfault, with the TEST_FRAMEWORK as a developer is a bit terse, sure I'd rather report the problem, but at least we know there is an underlying problem. |
| 20:47.10 | alecdavis | corydon76-dig: fixing the unit test will not fix the currently broken branches and trunk from a users point of view. |
| 20:47.32 | Corydon76-dig | That depends upon how you fix the unit test |
| 20:47.57 | Corydon76-dig | You don't have to change the unit test to fix it... in fact, it's probably better if you fix the code that caused the unit test to fail |
| 20:48.13 | *** join/#asterisk-dev dseeb_ (~Adium@CPE-58-169-81-65.lns1.bat.bigpond.net.au) |
| 21:14.15 | *** join/#asterisk-dev obnauticus (~obnauticu@about/windows/regular/obnauticus) |
| 21:16.39 | alecdavis | corydon76-dig: unfair blame, the unit test highlighted the problem. |
| 22:02.38 | Kobaz | alecdavis: i think the blame is in the right spot... who ever broke the code should have realized it by running the unit tests and seeing it fail |
| 22:21.05 | *** part/#asterisk-dev ArtemMakhutov (~ArtemMakh@ip-95-223-6-41.unitymediagroup.de) |
| 22:42.08 | alecdavis | kobaz: the original unit test passed the broken code which was submitted by Tilghman, causing trunk and branches to segfault when ever you go off hook. Only later did I modify the unit test to highlight the flaw. |
| 22:43.16 | Kobaz | oh okay. that's good then |
| 22:43.18 | alecdavis | it's all good now, though, branches and trunk fixed, the same way as I'd suggested in 16927. |
| 22:46.25 | Kobaz | hal: yes i know, but like rtoad said... it's general use. of course there is collisions |
| 22:46.31 | Kobaz | er |
| 22:46.32 | Kobaz | wrong chan |
| 23:01.21 | *** join/#asterisk-dev jdoe (jdoe@falseprophet.ca) |
| 23:01.41 | Kobaz | and explain what you're doing... don't just say 'something broke' |
| 23:02.33 | jdoe | ;) |
| 23:02.39 | Kobaz | do de do de do |
| 23:02.43 | jdoe | the patch to fix issue 15609 broke followme for me, hangs up immediately when I answer the call. If I revert that patch I get the "you have an incoming call, press 1 to accept" message. |
| 23:03.06 | Kobaz | M15609 |
| 23:03.15 | MuffinMan | [closed] [Asterisk] Core/Channels 0015609: [patch] WARNING[23025]: channel.c:952 __ast_queue_frame: Exceptionally long voice queue length queuing to Local reported by aragon https://issues.asterisk.org/view.php?id=15609 |
| 23:03.48 | jdoe | patch is https://issues.asterisk.org/file_download.php?file_id=24912&type=bug |
| 23:03.56 | Kobaz | pastebin your console and configs |
| 23:04.21 | jdoe | sure, give me one sec, I need to sanitize them... |
| 23:05.22 | jdoe | http://pastebin.ca/1816162 |
| 23:05.35 | jdoe | there's what I see on the console... do you need all configs, or is extensions.conf good enough? |
| 23:05.44 | jdoe | pastebinning a bunch of stuff is going to suck a bit. |
| 23:06.41 | Kobaz | it looks like you have different dialplan? |
| 23:07.00 | jdoe | same configs on both. |
| 23:08.21 | Kobaz | i'm not familiar with followme, but... reopen the issue and paste your console output, and relavent configs |
| 23:10.59 | jdoe | will do, thanks. |
| 23:11.29 | snuff-work | its interesting.. out of the 500 odd bugs 100 are 'ready for testing' yet no test love shown :( |
| 23:11.46 | Kobaz | poor bugs |
| 23:12.14 | Kobaz | ready for testing should require a unit test to be submitted before issue closure |
| 23:16.30 | jdoe | this is probably a dumb question, but ... how do I reopen it? I see no options to change status or even reply to it. |
| 23:25.52 | jdoe | ah, I don't have access. I'll just report a new one. |
| 23:31.26 | snuff-work | what bug do u need re-opened? |
| 23:31.29 | snuff-work | i can do it |
| 23:32.32 | jdoe | 15609 |
| 23:32.33 | Kobaz | if you're the original submitter you can reopen it |
| 23:32.51 | jdoe | I'm not. I was just upgrading a 1.6.1.6 install and noticed the problem now. |
| 23:33.57 | snuff-work | jdoe: since that's such a HUGE long issue.. |
| 23:34.12 | snuff-work | probably best to open new one and then we can LINK it to the old |
| 23:34.47 | jdoe | works for me. Uh, any guidelines on severity? I mean it's completely broken for me, but if I'm the only one who's found it... |
| 23:35.45 | snuff-work | put major.. if a bug marshal deems otherwise they'll change it |
| 23:42.30 | jdoe | whoops... so much for sanitizing it, left a real number in there > < |
| 23:43.45 | jdoe | snuff-work: issue is 16929, dunno if I can link it. |
| 23:44.43 | snuff-work | mm there is another post about followme and 1 way rtp |
| 23:44.53 | snuff-work | is that what ur experiencing? |
| 23:45.59 | snuff-work | M16923 |
| 23:46.02 | MuffinMan | [new] [Asterisk] Applications/app_followme 0016923: RTP traffic only seen in one direction when using FollowMe() reported by uxbod https://issues.asterisk.org/view.php?id=16923 |
| 23:46.42 | snuff-work | jdoe: can u self edit your issue or whould u like me to fix it (phone number in bug desc) ? |
| 23:47.33 | jdoe | might be the same, symptoms are similar. I didn't tcpdump or anything, but in my case asterisk dials out but when I answer I get dead air, on the console I see the channel exit. |
| 23:47.47 | snuff-work | bb in 15.. need to restart pc |
| 23:47.51 | jdoe | it *is* that patch though, 1.6.1.13 with the patch reverted works fine. |
| 23:48.04 | jdoe | and yeah, if you could fix the ticket for me, that would be awesome. I can't edit it myself. |
| 23:52.44 | snuff-work | which bit needs changing in tick? |
| 23:52.46 | snuff-work | *ticket |
| 23:55.11 | jdoe | the 1009 number should be 7775551212 |
| 23:56.08 | snuff-work | paste which line.. |
| 23:57.20 | snuff-work | number=>17789181009&200,30 <== remove the 1009 ? |
| 23:58.18 | jdoe | should be number=>17775551212&200,30 |