| 00:16.04 | putnopvut | M11467 |
| 00:16.04 | MuffinMan | [new] [Asterisk] Core/Channels 0011467: h extension not executing, zap channel getting stuck reported by jmls http://bugs.digium.com/view.php?id=11467 |
| 00:16.19 | putnopvut | I am fairly certain this is due to autoservice. |
| 00:16.26 | file | I would agree |
| 00:16.26 | Qwell | that's what I said |
| 00:16.36 | putnopvut | I was able to reproduce this behavior by setting the memberdelay in a queue and hanging up during that. |
| 00:16.50 | Qwell | memberdelay? |
| 00:17.10 | putnopvut | Yeah, you can set a delay between when the call is answered and when the channels are bridged. |
| 00:17.18 | putnopvut | And no, I don't know its purpose. |
| 00:17.37 | putnopvut | But during that delay, the channel is autoserviced. |
| 00:18.00 | putnopvut | So is the cure for this to not ignore AST_FRAME_CONTROL with subclass AST_CONTROL_HANGUP? |
| 00:20.26 | putnopvut | Yeah, I think that's what'll do it... |
| 00:20.58 | file | your question slightly interests me, what do you mean? |
| 00:21.28 | putnopvut | Well, it appears as though frames received when the channel is in autoservice are just freed. |
| 00:21.37 | file | not in the latest 1.4 and trunk |
| 00:21.50 | putnopvut | Okay, I see that DTMF frames are not. |
| 00:21.56 | putnopvut | Er... |
| 00:22.29 | putnopvut | Oh, wait...Okay, control frames are not simply freed either. Nevermind. |
| 00:23.16 | file | now the thing to remember is that upon hangup ast_read returns NULL, a control frame isn't always there... |
| 00:23.50 | file | so autoservice is probably servicing that, and when it gets taken out of autoservice nothing else gets added to the channel... so the ast_waitfor on it never trips... ast_read is never called to indicate it was hangup |
| 00:23.53 | file | (if I had to guess) |
| 00:24.16 | putnopvut | Ah, interesting. |
| 00:27.20 | file | you should be able to lab that up and see |
| 00:29.54 | putnopvut | You're right, it is receiving a NULL frame. |
| 00:30.39 | putnopvut | Is it safe to assume that a NULL frame will always indicate a hangup? |
| 00:31.14 | file | if your ast_astwait_whatever returns that a frame is to be read and when you read it there is none, then yes |
| 00:31.28 | putnopvut | Sweet. |
| 00:31.38 | putnopvut | I'll come up with a patch, then. |
| 00:31.50 | putnopvut | Thanks file! |
| 00:46.17 | putnopvut | Patch written! |
| 00:48.02 | Corydon76-vcch | A null frame or a NULL? |
| 00:48.51 | putnopvut | NULL |
| 01:43.24 | *** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 01:43.24 | *** mode/#asterisk-bugs [+o russellb] by ChanServ |
| 02:54.23 | *** join/#asterisk-bugs fakhir (n=fakhir@unaffiliated/fakhir) |
| 03:06.47 | *** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 03:06.47 | *** mode/#asterisk-bugs [+o russellb] by ChanServ |
| 03:45.18 | *** join/#asterisk-bugs kpfleming (i=kpflemin@nat/digium/x-6a40a80df83a1937) [NETSPLIT VICTIM] |
| 03:45.18 | *** join/#asterisk-bugs d3wayne (n=deeewayn@76.29.245.9) [NETSPLIT VICTIM] |
| 03:45.18 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) [NETSPLIT VICTIM] |
| 03:45.18 | *** join/#asterisk-bugs Qwell (i=north@pdpc/sponsor/digium/Qwell) [NETSPLIT VICTIM] |
| 03:45.18 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-574b9a75e6d08d56) [NETSPLIT VICTIM] |
| 03:45.18 | *** mode/#asterisk-bugs [+oooo kpfleming Corydon76-vcch Qwell Deeewayne] by irc.freenode.net |
| 03:45.18 | *** join/#asterisk-bugs MuffinMan (n=muffins@neutrino.joshua-colp.com) [NETSPLIT VICTIM] |
| 04:28.58 | *** join/#asterisk-bugs Qwell (i=north@pdpc/sponsor/digium/Qwell) [NETSPLIT VICTIM] |
| 04:28.58 | *** mode/#asterisk-bugs [+o Qwell] by ChanServ |
| 04:36.36 | *** join/#asterisk-bugs kpfleming (i=kpflemin@nat/digium/x-6a40a80df83a1937) [NETSPLIT VICTIM] |
| 04:36.36 | *** join/#asterisk-bugs d3wayne (n=deeewayn@76.29.245.9) [NETSPLIT VICTIM] |
| 04:36.36 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) |
| 04:36.36 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-574b9a75e6d08d56) |
| 04:36.36 | *** join/#asterisk-bugs MuffinMan (n=muffins@neutrino.joshua-colp.com) [NETSPLIT VICTIM] |
| 04:36.36 | *** mode/#asterisk-bugs [+ooo kpfleming Corydon76-vcch Deeewayne] by irc.freenode.net |
| 05:05.51 | *** join/#asterisk-bugs kpfleming (i=kpflemin@nat/digium/x-6a40a80df83a1937) |
| 05:05.51 | *** join/#asterisk-bugs d3wayne (n=deeewayn@76.29.245.9) |
| 05:05.51 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) |
| 05:05.51 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-574b9a75e6d08d56) |
| 05:05.51 | *** join/#asterisk-bugs MuffinMan (n=muffins@neutrino.joshua-colp.com) |
| 05:05.51 | *** mode/#asterisk-bugs [+ooo kpfleming Corydon76-vcch Deeewayne] by irc.freenode.net |
| 05:08.34 | *** join/#asterisk-bugs implicit (n=implicit@ip68-105-92-210.sd.sd.cox.net) |
| 05:08.34 | *** mode/#asterisk-bugs [+o implicit] by ChanServ |
| 05:26.07 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-8471c2111ff7dd7b) [NETSPLIT VICTIM] |
| 05:26.07 | *** mode/#asterisk-bugs [+o Deeewayne] by ChanServ |
| 05:26.35 | *** join/#asterisk-bugs kpfleming (i=kpflemin@nat/digium/x-6a40a80df83a1937) [NETSPLIT VICTIM] |
| 05:26.35 | *** join/#asterisk-bugs d3wayne (n=deeewayn@76.29.245.9) [NETSPLIT VICTIM] |
| 05:26.35 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) [NETSPLIT VICTIM] |
| 05:26.35 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-574b9a75e6d08d56) |
| 05:26.35 | *** join/#asterisk-bugs MuffinMan (n=muffins@neutrino.joshua-colp.com) [NETSPLIT VICTIM] |
| 05:26.35 | *** mode/#asterisk-bugs [+ooo kpfleming Corydon76-vcch Deeewayne] by irc.freenode.net |
| 05:29.18 | *** join/#asterisk-bugs d3wayne (n=deeewayn@76.29.245.9) |
| 05:31.48 | *** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 05:31.48 | *** mode/#asterisk-bugs [+o russellb] by ChanServ |
| 06:00.22 | *** join/#asterisk-bugs MuffinMan (n=muffins@neutrino.joshua-colp.com) |
| 06:27.16 | *** join/#asterisk-bugs implicit (n=implicit@ip68-105-92-210.sd.sd.cox.net) |
| 06:27.16 | *** mode/#asterisk-bugs [+o implicit] by ChanServ |
| 07:41.43 | *** join/#asterisk-bugs wyoming (n=steve_mu@216.166.159.235) |
| 09:58.25 | *** join/#asterisk-bugs oej (n=olle@193.126.30.214) |
| 10:23.51 | *** part/#asterisk-bugs jmls (n=jmls@62.49.235.130) |
| 11:08.37 | *** join/#asterisk-bugs caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:09.51 | *** join/#asterisk-bugs oej (n=olle@193.126.30.214) |
| 11:48.42 | *** join/#asterisk-bugs jmls (n=jmls@62.49.235.130) |
| 12:45.15 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) |
| 12:45.15 | *** mode/#asterisk-bugs [+o Corydon76-vcch] by ChanServ |
| 12:50.06 | *** join/#asterisk-bugs zerohalo (n=zeroHalo@pool-71-162-97-18.bstnma.east.verizon.net) |
| 13:45.19 | *** join/#asterisk-bugs jmls (n=jmls@62.49.235.130) |
| 13:46.20 | *** join/#asterisk-bugs Corydon76-vcch (n=tilghman@pdpc/supporter/bronze/Corydon76-home) |
| 13:46.20 | *** join/#asterisk-bugs Qwell (i=north@pdpc/sponsor/digium/Qwell) |
| 13:46.20 | *** join/#asterisk-bugs dan42 (n=lung@24-148-96-186.ip.mhcable.com) |
| 13:46.20 | *** join/#asterisk-bugs Juggie (n=Juggie@CPE001601df17fb-CM001a6682d7b6.cpe.net.cable.rogers.com) |
| 13:46.20 | *** mode/#asterisk-bugs [+ooo Corydon76-vcch Qwell Juggie] by irc.freenode.net |
| 14:07.16 | *** join/#asterisk-bugs Pagautas (n=bigman@83.171.14.250) |
| 14:21.03 | *** join/#asterisk-bugs eliel (n=eliel@200.59.172.38) |
| 14:29.21 | *** join/#asterisk-bugs putnopvut (i=putnopvu@nat/digium/x-b9820a83e0eac07d) |
| 14:29.21 | *** mode/#asterisk-bugs [+o putnopvut] by ChanServ |
| 14:53.45 | *** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 14:53.45 | *** mode/#asterisk-bugs [+o russellb] by ChanServ |
| 15:34.18 | *** join/#asterisk-bugs jsmith (n=jsmith@68-246-38-7.area3.spcsdns.net) |
| 15:34.18 | *** mode/#asterisk-bugs [+o jsmith] by ChanServ |
| 15:45.37 | russellb | putnopvut: file ... what was the problem you were trying to fix that inspired the autoservice change? |
| 15:45.57 | file | M11467 |
| 15:45.58 | MuffinMan | [closed] [Asterisk] Core/Channels 0011467: [patch] h extension not executing, zap channel getting stuck reported by jmls http://bugs.digium.com/view.php?id=11467 |
| 15:46.15 | *** join/#asterisk-bugs kpfleming (i=kpflemin@nat/digium/x-9188e237b21ab828) |
| 15:46.16 | *** mode/#asterisk-bugs [+o kpfleming] by ChanServ |
| 15:47.31 | russellb | well i don't see how it helps |
| 15:48.14 | russellb | it's not going to take the channel out of autoservice any sooner |
| 15:48.21 | *** join/#asterisk-bugs blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 15:48.21 | *** mode/#asterisk-bugs [+o blitzrage] by ChanServ |
| 15:49.03 | russellb | and there is nowhere else in Asterisk that reacts to a NULL from ast_read that way (that i can think of, or find from a quick look) |
| 15:49.25 | Qwell | it isn't that it sticks around in autoservice, it's that when it's being serviced, the hangup is basically ignored...or something |
| 15:50.08 | russellb | eh? |
| 15:51.01 | russellb | perhaps the zap channel never again indicates that anything is available? so ast_waitfor() blocks forever after it returns? |
| 15:51.12 | russellb | instead of just returning NULL again? |
| 15:51.31 | blitzrage | russellb: hey, I got thinking last night that 8925 is probably a good candidate for a team branch |
| 15:51.33 | Qwell | the way putnopvut explained it to me last night.. when a channel is being autoserviced, it drops frames on the floor, except for certain types - like dtmf. it queues those up after it stops being serviced. |
| 15:52.59 | russellb | i know :) I wrote the code to make it work that way just last week :-p |
| 15:53.07 | russellb | previously it dropped _everything_ |
| 15:53.21 | russellb | but the change still doesn't make sense to me |
| 15:53.47 | putnopvut | file indicated to me that a NULL indicates that we received a hangup. |
| 15:54.01 | file | channel goes into autoservice, autoservice handles hangup (calls ast_read gets NULL yada), channel gets taken out of autoservice, ast_waitfor_yada is called on it, that never trips because nobody else writes to the channel (as it was hungup) |
| 15:54.25 | russellb | file: that's what i was looking for. the fact that the zap channel never returns again |
| 15:54.37 | file | I wouldn't say it's a zap channel issue specifically |
| 15:54.46 | file | I could certainly see it with SIP as well |
| 15:54.54 | putnopvut | Yeah, it happened with SIP, too. |
| 15:55.12 | putnopvut | I reproduced the problem myself with a SIP phone and the patch I wrote fixed it. |
| 15:55.20 | russellb | alrighty, cool |
| 15:55.36 | russellb | i just wanted to understand why it was needed, and now i get it |
| 15:55.37 | russellb | thanks. |
| 15:55.43 | putnopvut | All righty. |
| 15:55.46 | file | get it? got it? good. |
| 15:57.25 | *** join/#asterisk-bugs fakhir (n=fakhir@unaffiliated/fakhir) |
| 16:00.19 | *** join/#asterisk-bugs ibisgroup (n=chris@ns1.ipstorm.net) |
| 16:01.06 | ibisgroup | I noticed a bug when trying to change the callerid in the extension.conf when the callerid comes in as unknown |
| 16:02.19 | ibisgroup | this only occurs if the digium card is isdn |
| 16:03.06 | jsmith | ibisgroup: Have you posted the details to the bug tracker? |
| 16:03.16 | ibisgroup | no I will do that |
| 16:03.46 | ibisgroup | do you have a link to the bug tracker? is it the bugs.digium.com? |
| 16:03.50 | Qwell | yes |
| 16:04.01 | ibisgroup | ok thanks will document in detail |
| 16:04.03 | ibisgroup | thanks |
| 16:09.03 | *** join/#asterisk-bugs juanjoc (n=juanjoc@190.2.0.145) |
| 16:25.37 | *** join/#asterisk-bugs Deeewayne (i=dwayne@nat/digium/x-e3de0baa6758bebb) |
| 16:25.37 | *** mode/#asterisk-bugs [+o Deeewayne] by ChanServ |
| 17:34.02 | *** join/#asterisk-bugs eliel (n=eliel@200.59.172.38) |
| 17:49.01 | *** join/#asterisk-bugs jcmoore (n=jcmoore@unaffiliated/tgrman) |
| 18:23.13 | *** join/#asterisk-bugs bkruse (i=bkruse@nat/digium/x-0ee9b92726190abb) |
| 19:00.56 | *** part/#asterisk-bugs jmls (n=jmls@62.49.235.130) |
| 19:33.12 | *** join/#asterisk-bugs pjezek (n=pj@193.85.164.154) |
| 19:43.53 | pjezek | if anyone using latest trunk rev. and observing permanent cpu load caused by asterisk, even if no calls, report your findings in bug #0011449 |
| 19:45.31 | Qwell | pjezek: are you using gcc-4.2? |
| 19:50.20 | pjezek | I tried gcc 4.2.2 or 3.3 no change, if with or without optimalization, compiling with gcc 4.2 has even another bug #0011243 |
| 19:50.30 | Qwell | yeah, I know about that one |
| 19:50.50 | Qwell | I was just curious, because you mentioned gsm transcoding taking a lot of CPU time - I was curious if that was a similar reason |
| 19:51.41 | pjezek | gsm-alaw was only example, also other transcodings have poor performance g729-alaw |
| 19:51.57 | Qwell | well, that doesn't surprise me much on a 300mhz machine.. |
| 19:53.41 | pjezek | yes, but until r89551 everithinh was working fine, one transcoding alaw-gsm about 6% cpu load, now it's 50% load, and 15% permanant load if no calls processing. |
| 19:56.31 | pjezek | I can try to compile some other builds between r89551 and current rev., but this itaretion for bug finding will be very time consuming :( |
| 20:07.36 | *** join/#asterisk-bugs implicit (n=implicit@ip68-105-92-210.sd.sd.cox.net) |
| 20:07.36 | *** mode/#asterisk-bugs [+o implicit] by ChanServ |
| 20:07.47 | *** join/#asterisk-bugs jmls (n=jmls@62.49.235.130) |
| 21:29.33 | *** join/#asterisk-bugs JunK-Y (n=junky@modemcable183.17-83-70.mc.videotron.ca) |
| 21:29.33 | *** mode/#asterisk-bugs [+o JunK-Y] by ChanServ |
| 21:35.22 | *** join/#asterisk-bugs implicit (n=implicit@ip68-105-92-210.sd.sd.cox.net) |
| 21:35.22 | *** mode/#asterisk-bugs [+o implicit] by ChanServ |
| 21:35.30 | *** join/#asterisk-bugs blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) [NETSPLIT VICTIM] |
| 21:35.30 | *** mode/#asterisk-bugs [+o blitzrage] by ChanServ |
| 21:43.38 | *** join/#asterisk-bugs angler (i=angler@pdpc/sponsor/digium/angler) [NETSPLIT VICTIM] |
| 21:43.48 | *** join/#asterisk-bugs Corydon76-home (i=eight@pdpc/supporter/bronze/Corydon76-home) |
| 21:43.48 | *** mode/#asterisk-bugs [+o Corydon76-home] by ChanServ |
| 21:46.38 | *** join/#asterisk-bugs wyoming (n=steve_mu@216.166.159.235) |
| 22:32.12 | *** join/#asterisk-bugs citats (n=james@mrplow.gnuinternet.com) |
| 22:32.12 | *** mode/#asterisk-bugs [+o citats] by ChanServ |
| 23:01.01 | *** mode/#asterisk-bugs [+o codefreeze] by ChanServ |