| 00:24.16 | *** join/#asterisk-dev moy (~moy@64.231.54.132) |
| 01:01.36 | *** join/#asterisk-dev lot49 (~jnoh@c-24-63-27-170.hsd1.ma.comcast.net) |
| 02:00.07 | *** join/#asterisk-dev infobot (~infobot@rikers.org) |
| 02:00.07 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- http://xkcd.com/303/ |
| 05:13.48 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 05:13.48 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 06:46.36 | *** join/#asterisk-dev philipp64|laptop (~chatzilla@50-39-153-236.bvtn.or.frontiernet.net) |
| 06:48.42 | *** join/#asterisk-dev schmidts (~schmidts@Lo20-rt01.lm33.sil.at) |
| 06:48.44 | schmidts | good morning |
| 06:54.42 | *** join/#asterisk-dev hehol (~hehol@buero-gw.dortmund.loca.net) |
| 07:06.16 | *** join/#asterisk-dev sgimeno (~sgimeno@163.117.206.10) |
| 07:11.52 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 07:13.36 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 07:13.36 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 08:37.34 | *** join/#asterisk-dev irroot (~irroot@41.125.41.15) |
| 08:40.44 | *** join/#asterisk-dev irroot (~irroot@41.125.41.15) |
| 08:43.36 | *** join/#asterisk-dev CoderForLife (~Miranda@unaffiliated/coderforlife) |
| 08:47.41 | *** join/#asterisk-dev jkroon (~jkroon@dsl-241-252-52.telkomadsl.co.za) |
| 08:47.54 | jkroon | http://pastebin.com/esVSTLcT <-- another suspected ast_str misuse. |
| 09:47.49 | jkroon | russellb, regarding the patch last night - would a global patch that updates all calls that pass a ast_str** instead of a ast_str* be accepted? |
| 09:49.01 | *** join/#asterisk-dev Gugge (~gugge@91.208.16.1) |
| 09:55.51 | *** join/#asterisk-dev hehol (~hehol@buero-gw.dortmund.loca.net) |
| 09:58.36 | *** join/#asterisk-dev ccesario (~ccesario@187.17.166.162) |
| 10:04.43 | *** join/#asterisk-dev Mw3 (mw3@mw3.hu) |
| 10:39.12 | *** join/#asterisk-dev bbryant (~brett@c-76-26-221-79.hsd1.sc.comcast.net) |
| 10:39.12 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 11:27.23 | *** join/#asterisk-dev irroot (~irroot@41.16.24.26) |
| 11:32.38 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 11:32.38 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:06.48 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 12:06.48 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 12:34.23 | *** join/#asterisk-dev t_dot_zilla (~chatzilla@firewall-a.buf.ny.i-evolve.net) |
| 12:43.28 | *** join/#asterisk-dev Ulrar (~quassel@2a01:e0b:1:136:62eb:69ff:fe8f:18a0) |
| 12:55.00 | *** part/#asterisk-dev irroot (~irroot@41.16.24.26) |
| 13:35.48 | *** join/#asterisk-dev Rico29 (~rico@sar-b123.olm.fr) |
| 13:36.00 | Rico29 | hi all |
| 13:36.40 | Rico29 | Just a little question. Is there a way to do a kind of backtrace when only sip stack crashes, but thes asterisk process doesn't ? |
| 13:37.07 | Rico29 | Somebody told me to do a "core show locks" |
| 13:37.17 | Rico29 | can I do anything else to get more debugging stuff ? |
| 13:42.07 | malcolmd | there are some documents on the Wiki covering Asterisk debugging: https://wiki.asterisk.org/wiki/display/AST/Debugging |
| 13:45.23 | Rico29 | yes, i've read it |
| 13:45.40 | Rico29 | i have full debugging in my logfile, but I can't get a backtrace because asterisk is not crashing |
| 13:45.57 | Rico29 | and I did not see anything interesting il logfile |
| 13:49.05 | schmidts | Rico29 afaik in the wiki is also a description how to use gdb with a running process to create a backtrace |
| 13:49.25 | schmidts | btw its easy ;) gdb /usr/sbin/asterisk -p PID |
| 13:51.44 | Rico29 | ok, thanks a lot, I'll try to do a backtrace |
| 13:52.39 | schmidts | sorry i was wrong its not -p its --pid=PID |
| 13:52.45 | Rico29 | ok |
| 13:53.33 | schmidts | hmm or just without any option it should also work gdb /usr/sbin/asterisk PID :D |
| 13:54.14 | *** join/#asterisk-dev pabelanger (~pabelange@nat/digium/x-fhnnrqmzusuqycys) |
| 13:54.14 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 14:00.11 | Rico29 | schmidts, it will not stop or restart asterisk ? it's in a production environment |
| 14:04.10 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:04.10 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:09.15 | wdoekes2 | Rico29: yes.. it will hang while you do nothing |
| 14:09.41 | *** join/#asterisk-dev Victor_Yure (~aaa@unaffiliated/victoryure/x-837844) |
| 14:09.57 | wdoekes2 | gdb `which asterisk -p `pidof asterisk` -ex 'thread apply all bt' -ex 'thread apply all bt full' -ex detach -ex quit > gdb.txt |
| 14:10.42 | wdoekes2 | this will dump info and detach immediately so asterisk can continue |
| 14:10.55 | Rico29 | ok |
| 14:11.06 | wdoekes2 | s/asterisk -p/asterisk` -p/ |
| 14:11.47 | Rico29 | yes I saw |
| 14:11.48 | Rico29 | thanks |
| 14:11.51 | Rico29 | i'll try |
| 14:12.02 | Rico29 | so when I do a "core show locks" it will give me a PID |
| 14:12.08 | Rico29 | and i'll try with this pid |
| 14:12.11 | schmidts | no |
| 14:12.17 | Rico29 | oops |
| 14:12.21 | schmidts | ps faux | grep asterisk will give you a pid ;) |
| 14:12.36 | wdoekes2 | or: pidof asterisk |
| 14:12.37 | Rico29 | ok |
| 14:12.41 | schmidts | with core show locks you will get thread ids, you can find these in the gdb.txt file then |
| 14:12.45 | Rico29 | so what will "core show locks" give me ? |
| 14:12.54 | Rico29 | ok !! |
| 14:12.57 | Rico29 | great |
| 14:15.05 | wdoekes2 | (alternatively, you can kill -ABRT `pidof asterisk` and restart a new asterisk) |
| 14:15.18 | wdoekes2 | (then you get a core dump you can look at in all peace) |
| 14:15.50 | wdoekes2 | (it will of course kill all running calls) |
| 14:17.03 | Rico29 | iep |
| 14:17.09 | Rico29 | ok |
| 14:24.27 | *** join/#asterisk-dev akshayb (~akshay@117.203.10.133) |
| 14:30.14 | *** join/#asterisk-dev rgagnon (~rgagnon@99-185-128-25.lightspeed.austtx.sbcglobal.net) |
| 14:30.17 | *** part/#asterisk-dev rgagnon (~rgagnon@99-185-128-25.lightspeed.austtx.sbcglobal.net) |
| 14:48.22 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 14:48.22 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 14:57.02 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:57.02 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 15:11.38 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 15:24.06 | *** join/#asterisk-dev moy (~moy@bas1-montreal42-1177928259.dsl.bell.ca) |
| 15:51.08 | *** join/#asterisk-dev irroot (~irroot@41.125.11.130) |
| 15:52.36 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 15:52.40 | *** part/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 15:53.43 | *** join/#asterisk-dev zerohalo (~zerohalo@173-13-92-27-NewEngland.hfc.comcastbusiness.net) |
| 15:54.28 | *** join/#asterisk-dev zerohalo (~zerohalo@173-13-92-27-NewEngland.hfc.comcastbusiness.net) |
| 16:19.52 | acidfoo | wierd stuff, (asterisk 1.8.5) into channel.c: ast_channel_early_bridge() doesn't check if parameter can be NULL .... and into app_dial.c in the function dial_exec_full() there is a call: ast_channel_early_bridge(chan, NULL); |
| 16:20.05 | acidfoo | well, eating time, ill try to understand after |
| 18:29.06 | *** join/#asterisk-dev jkroon (~jkroon@dsl-242-3-12.telkomadsl.co.za) |
| 18:30.11 | *** join/#asterisk-dev luckman212 (~irc@2001:470:1f07:1225:8048:3812:9d88:a161) |
| 18:34.08 | jkroon | russellb, kpfleming - would either of you mind helping me yet again? http://pastebin.com/bP8Pe7xS - both cases segfault, not sure why. |
| 18:34.41 | jkroon | from the looks of it an iax2 registration refresh that re-adds the extension to regcontext and ends up exploding for some reason. |
| 18:35.28 | kpfleming | jkroon: what branch/version? |
| 18:35.52 | jkroon | 1.8.5.0 once more. |
| 18:36.03 | jkroon | patch added from last night (morning for you probably) |
| 18:37.09 | kpfleming | no, it was evening |
| 18:37.41 | kpfleming | there's nothing obvious there... you'll probably have to recompile with DONT_OPTIMIZE to get a more useful backtrace |
| 18:38.31 | jkroon | crap. OK, let me first get a reliable non-production reproducable going. |
| 18:44.55 | *** join/#asterisk-dev irroot (~irroot@197.168.41.98) |
| 18:50.01 | *** join/#asterisk-dev hehol (~Adium@2a01:198:71d:0:21f:d0ff:fea1:568e) |
| 18:55.42 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-218-155-145.cablep.bezeqint.net) |
| 18:55.43 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 19:31.43 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 19:31.43 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 19:59.32 | *** join/#asterisk-dev Victor_Yure (~aaa@unaffiliated/victoryure/x-837844) |
| 20:32.57 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 20:32.57 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 21:21.02 | jkroon | kpfleming, [Jul 13 23:19:25] WARNING[13636]: pbx.c:2044 add_exten_to_pattern_tree: Found duplicate exten. Had p¼üY found uls |
| 21:21.17 | jkroon | that imho indicates badness. |
| 21:21.38 | kpfleming | indeed it does |
| 21:23.03 | jkroon | ok, what can I do to assist in tracking and fixing? |
| 21:23.29 | jkroon | that definitely comes from chan_iax's regcontext, but I've also seen one core dump via chan_sip. |
| 21:23.40 | kpfleming | i don't honestly know right now... you've got corruption happening in structures somehow. that sort of thing is hard to track down. |
| 21:23.54 | jkroon | valgrind? |
| 21:24.11 | kpfleming | maybe... but if it's just memory being overwritten, valgrind can't catch that. |
| 21:29.49 | jkroon | seems we got lucky |
| 21:29.58 | jkroon | pasting then i can help starting to look at code. |
| 21:30.30 | jkroon | http://pastebin.com/5w6L7hav |
| 21:31.40 | jkroon | so it seems the breakage case is register, then deregister (I just force deregistered it via the CLI) and then re-register. |
| 21:33.33 | jkroon | or more specifically, ast_add_extension(); ast_context_remove_extension(); ast_add_extension() ... |
| 21:53.51 | *** part/#asterisk-dev irroot (~irroot@197.168.41.98) |
| 21:59.56 | *** join/#asterisk-dev wasanzy (~emmanuel@196.201.43.55) |
| 22:01.27 | *** join/#asterisk-dev Victor_Yure (~aaa@unaffiliated/victoryure/x-837844) |
| 22:02.17 | jkroon | can someone please confirm for me that ast_hashtab_remove_this_object_nolock should be calling tab->compare() to compare obj with the objects in the bucket instead of doing a pointer comparison? |
| 22:05.53 | jkroon | none of the LOG_ERROR messages that should fire that I can see in oast_context_remove_extension_callerid2 that will likely fire if the above is wrong is firing, so I'm not sure about it either way. |
| 22:08.33 | jkroon | nm, there is a _via_lookup that covers that case. |
| 22:30.54 | jkroon | ok, add_exten_to_pattern_tree with findonly=1 if the looked up exten is a prefix on another one, and hits a previously deleted node. |
| 22:35.38 | jkroon | and the if() statement on line 2043 does not honor the deleted flag. |
| 22:36.18 | jkroon | same on 2067 but that's not the branch i'm hitting. |
| 22:39.11 | *** join/#asterisk-dev seraphie (~erin@207.98.195.107) |
| 23:32.56 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 23:32.56 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |