| 01:18.56 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 01:55.56 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 03:26.43 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 03:46.46 | *** join/#asterisk-dev babak (uid19622@gateway/web/irccloud.com/x-lworgtolnalzbtpb) |
| 04:58.34 | *** join/#asterisk-dev leedm777 (~leedm777@50-81-28-152.client.mchsi.com) |
| 06:34.45 | *** join/#asterisk-dev oej (~oej@9.210.50.195.sta.estpak.ee) |
| 06:34.45 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 06:35.59 | oej | Morning folks |
| 06:56.00 | *** join/#asterisk-dev hehol (~hehol@gatekeeper.loca.net) |
| 07:12.56 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 07:12.56 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 08:03.39 | *** join/#asterisk-dev frenk77 (~frenk77@gw3.lbox.cz) |
| 08:07.35 | frenk77 | Hello i have some problem with gdb and core dump. If i use gcore -o file PID, then open in gdb it is ok, i see every function any "?? ()" after bactrace adresses. But problem is when asterisk crash by call abort in code or by kill -SIGABRT PID. I open gdb with coredump and it writes many "warning: .dynamic section for "/usr/lib64/asterisk/modules/app_cdr.so" is not at the expected address (wrong library or version mismatch?)" then i have 90% of backtrac |
| 08:28.00 | *** join/#asterisk-dev flo291288 (~flo291288@mx306-1-88-173-34-251.fbx.proxad.net) |
| 08:31.23 | flo291288 | Hello |
| 08:32.44 | flo291288 | wath is the procedure for bloking the loading of an app_module when the dependency (res_module) is LOAD_DECLINE? or FAILURE? Regards. |
| 08:35.18 | flo291288 | is normal when my res_module returne AST_LOAD_MODULE_FAILURE asterisk is killed? |
| 08:47.34 | wdoekes | frenk77: do you have an outdated asterisk-dbg package? |
| 08:54.00 | frenk77 | no it is actual, same machine, same packages. Only difference is type of core generation: gcore of running process is ok, core generated by abort in code or SIGABRT signal is bad. |
| 09:37.42 | flo291288 | the ast_check_module function return 1 when module is not loaded, it's a bug? |
| 10:02.52 | *** join/#asterisk-dev oej (~oej@9.210.50.195.sta.estpak.ee) |
| 10:02.52 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 10:10.33 | frenk77 | nobody know with my problem ? :-/ |
| 11:03.09 | *** join/#asterisk-dev newtonr (RustyNewto@nat/digium/x-apijatbpkxqmpjtn) |
| 11:03.12 | *** mode/#asterisk-dev [+o newtonr] by ChanServ |
| 11:16.26 | frenk77 | Hello i have some problem with gdb and core dump. If i use gcore -o file PID, then open in gdb it is ok, i see every function any "?? ()" after bactrace adresses. |
| 11:16.28 | frenk77 | But problem is when asterisk crash by call abort in code or by kill -SIGABRT PID. I open gdb with coredump and it writes many "warning: .dynamic section for "/usr/lib64/asterisk/modules/app_cdr.so" is not at the expected address (wrong library or version mismatch?)" then i have 90% of backtraces with "?? ()" so i cant debug :( What can cause this ? |
| 12:19.51 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 12:25.27 | newtonr | russellb, for ASTERISK-25370 , the reporter can't reproduce under valgrind, can't get mmlog and we can't reproduce. I'm not sure there is anywhere to go on that one. |
| 12:30.55 | newtonr | I put it in feedback for you to see if you have any further ideas. |
| 12:58.49 | *** join/#asterisk-dev leedm777 (leedm777@nat/digium/x-psfwtjfsfedifhbb) |
| 13:15.02 | *** join/#asterisk-dev Marquis42 (~mrbmw@d192-24-112-5.try.wideopenwest.com) |
| 13:28.30 | *** join/#asterisk-dev babak (uid19622@gateway/web/irccloud.com/x-pcrdycpbgwghaepp) |
| 13:46.31 | toresbe | I'm still struggling with the crash I started seeing after I dist-upgraded, I've kept pulling git to see if something's changed but there's nothing so far |
| 13:46.44 | toresbe | http://pastebin.com/qUcs348m is the backtrace |
| 13:48.48 | frenk77 | can somebody help with my core ? :( |
| 13:49.06 | file | toresbe, that doesn't appear to be anything in our world... looks system level, what exactly I don't know |
| 13:49.14 | file | frenk77, can't say I've seen that... |
| 13:49.36 | frenk77 | ok again |
| 13:49.40 | frenk77 | Hello i have some problem with gdb and core dump. If i use gcore -o file PID, then open in gdb it is ok, i see every function any "?? ()" after bactrace adresses. |
| 13:49.45 | frenk77 | But problem is when asterisk crash by call abort in code or by kill -SIGABRT PID. I open gdb with coredump and it writes many "warning: .dynamic section for "/usr/lib64/asterisk/modules/app_cdr.so" is not at the expected address (wrong library or version mismatch?)" then i have 90% of backtraces with "?? ()" so i cant debug :( What can cause this ? |
| 13:49.46 | file | I saw |
| 13:49.53 | file | I meant I haven't seen that problem and don't know why it would happen |
| 13:50.53 | toresbe | file: this is a stock Debian kernel and a stock Debian glibc - so if it is system level then FWIW you're going to be seeing more of it |
| 13:51.39 | toresbe | http://pastebin.com/VCn69q1n here is ldd, version and a check of md5sum |
| 13:52.49 | frenk77 | ok nobody knows, so i now using abort() in code to generate crash/core....it is possible to call something whitch stops/pause asterisk process/all threads ... so then i can get my core from gcore(which work ok in dgb) |
| 13:53.33 | frenk77 | i mean call something instead of abort |
| 13:55.30 | file | frenk77, there is nothing in Asterisk itself to do that |
| 13:55.52 | file | toresbe, nothing stands out |
| 13:56.13 | file | tzafrir is a Debian user but unfortunately he is idle |
| 13:56.14 | frenk77 | pause function in c ? call some signal ? |
| 13:56.32 | tzafrir | what's the problem? |
| 13:56.39 | file | tzafrir, you appear! |
| 13:56.55 | file | tzafrir, toresbe did a dist-upgrade under Debian and now Asterisk builds are doing this: http://pastebin.com/qUcs348m |
| 13:58.58 | toresbe | tzafrir: hi :) |
| 13:59.27 | toresbe | I could pull from git and try it on a different machine with same Debian to eliminate any potential confounding factors |
| 13:59.31 | tzafrir | libgcc_s.so.1 is from libgcc1 which should be installed |
| 13:59.55 | toresbe | tzafrir: it is; I've confirmed the md5sum with debsums: http://pastebin.com/VCn69q1n |
| 14:00.26 | tzafrir | sid or stretch? |
| 14:00.37 | toresbe | jessie |
| 14:00.43 | toresbe | went from wheezy to jessie |
| 14:07.29 | toresbe | It doesn't always crash the same way fwiw |
| 14:08.02 | toresbe | http://pastebin.com/2zDsqNsJ here's one where I dialled into a MoH |
| 14:08.53 | *** join/#asterisk-dev kharwell (kharwell@nat/digium/x-greqakdedbpwkocy) |
| 14:10.00 | toresbe | sometimes it doesn't even finish launcing before the issue arises |
| 14:12.34 | tzafrir | I can't think of anything specific right now |
| 14:31.46 | wdoekes | toresbe: you did recompile asterisk after dist-upgrade? |
| 14:36.40 | *** part/#asterisk-dev frenk77 (~frenk77@gw3.lbox.cz) |
| 14:42.24 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 14:45.16 | toresbe | wdoekes: yes |
| 14:45.37 | toresbe | wdoekes: the last binary, which also crashed, was a rather old one from SVN |
| 14:53.53 | wdoekes | and this here (Cannot access memory at address 0x3062613231306334), looks like "4c012ab0" got written over your memory (callid? secret?). try valgrind? |
| 14:59.17 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-yhakoanphizevihd) |
| 14:59.29 | *** join/#asterisk-dev rmudgett (rmudgett@nat/digium/x-avdqvthedhiarpgf) |
| 15:16.25 | stefan27 | <@newtonr> russellb, for ASTERISK-25370 , the reporter can't reproduce under valgrind, can't get mmlog and we can't reproduce. I'm not sure there is anywhere to go on that one. |
| 15:16.38 | stefan27 | it's kind of the same for my ticket 25401 :) |
| 15:18.11 | stefan27 | If I could only reproduce it in an environment that's under my control I could upload a lot more debug-data |
| 15:21.22 | newtonr | stefan27, aye - putnovput will take a look when he gets a chance to see if we can do anything with just that trace. |
| 15:21.23 | stefan27 | But it's possible that ASTERISK-25401 which is hard to reproduce is related to ASTERISK-25317, which *is* re-producable. |
| 15:24.05 | stefan27 | I could try re-compiling asterisk and pjprojects with my own debug logs, but it's hard work |
| 15:25.05 | stefan27 | if I can't reproduce the issue it wouldn't help much either |
| 15:33.22 | *** join/#asterisk-dev superscrat (asanders@nat/digium/x-efmzknldtwnlvymo) |
| 15:38.47 | stefan27 | if I want to add code in functions like chan_sip::handle_response_invite(struct sip_pvt p,...), how do I know which locks I have, is the p locked in this case? I notice there's already code in there like if (p->owner) { ... do something with p->owner ... }, which suggests that one can be sure that nothing can NULL out p->owner while this function is executing? |
| 15:40.23 | file | p is locked there, to be sure you end up following the function calls |
| 15:40.27 | file | up up up |
| 15:42.19 | stefan27 | ok cool, I notice some functions have comments in their headers about pre and post conditions, which type of functions have those comments or is that arbitrary? |
| 15:43.01 | stefan27 | oh I gotta run it's soon 6, thanks anyways! |
| 15:43.08 | file | older stuff generally doesn't |
| 16:14.19 | seanbright | i'm in a bit of a bind and don't have time to look this up - what does chan_sip do if Hangup() is called before the channel is answered? |
| 16:14.27 | seanbright | 603? |
| 16:18.02 | file | I think so, yes |
| 16:18.15 | file | 603 or 503 |
| 16:20.58 | seanbright | thanks |
| 16:21.18 | seanbright | file: ^^ |
| 17:15.43 | *** join/#asterisk-dev oej (~oej@89.221.66.182) |
| 17:15.43 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 19:29.52 | *** join/#asterisk-dev bipul (~bipul@unaffiliated/bipul) |
| 19:32.19 | bipul | Hi. |
| 19:32.27 | bipul | Anyone around? |
| 19:33.21 | file | maybe |
| 19:33.57 | bipul | Hello file |
| 19:34.00 | file | hello |
| 19:34.10 | bipul | Good to see, someone responded to me. |
| 19:34.59 | bipul | Actually i have some question. |
| 19:36.17 | bipul | First i would like to contribute into asterisk. I am very old user of it. |
| 19:36.59 | bipul | What are the prerequisite or basic for any individual before jumping into the Asterisk-development? |
| 19:38.08 | file | knowing C really... otherwise it depends on what you want to do |
| 19:38.49 | bipul | Yes, i can write code in C, and basic Linux and networking knowledge apart, how mobile network works. |
| 19:39.03 | *** join/#asterisk-dev oej (~oej@89.221.66.182) |
| 19:39.03 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 19:41.34 | bipul | I wish i can know some groups or community where i can start . |
| 19:45.38 | bipul | file, ? |
| 19:45.58 | file | there's no real starting point, you have to either find some small things to look at or pick a place |
| 19:47.25 | bipul | pk. |
| 19:47.28 | bipul | ok. |
| 21:00.24 | *** join/#asterisk-dev [TK]D-Fender (~joe@64.235.216.2) |
| 21:25.00 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-179-40-172.cust.bezeqint.net) |
| 21:25.00 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 22:40.02 | TazzNZ | hey guys - I have a pcap of a failed call over ipv6 to chan_sip - I suspect it's an Asterisk issue, but would like to bounce it off someone first before reporting the issue - how would I go about doing that ? |
| 22:41.16 | [TK]D-Fender | pastebin is a great start |
| 22:41.24 | [TK]D-Fender | like .. showing us |
| 22:41.27 | [TK]D-Fender | ... or something |
| 22:44.12 | TazzNZ | http://www.calls4less.co.nz/failed_call.pcap <-- there is the file - it's captured on my firewall so it shows the packets twice, once entering the firewall and once exiting |
| 22:45.36 | TazzNZ | 2407:8300:8001 is my local ipv6 segment while 2001:4830:121e is the Asterisk server (freepbx) |
| 22:45.44 | TazzNZ | ipv4 calls work fine |
| 23:23.48 | *** part/#asterisk-dev kharwell (kharwell@nat/digium/x-greqakdedbpwkocy) |
| 23:44.04 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-jjhgglvurarmnfyo) |