| 00:53.56 | *** join/#asterisk-bugs Entomologist (~nobody@nat/digium/x-tgyfyzynkjalqoqn) |
| 00:57.29 | *** join/#asterisk-bugs Entomologist (~nobody@nat/digium/x-mglsghbduxquamee) |
| 01:28.34 | *** join/#asterisk-bugs n0tk (~n0tk@gateway/tor-sasl/n0tk) |
| 01:41.29 | *** join/#asterisk-bugs zerohalo (~zerohalo@74.61.196.236) |
| 04:06.24 | *** join/#asterisk-bugs tzafrir_laptop (~tzafrir@bzq-218-155-145.cablep.bezeqint.net) |
| 08:31.15 | *** join/#asterisk-bugs Entomologist (~nobody@nat/digium/x-wyxhxzdmjaazoxes) |
| 11:41.21 | *** join/#asterisk-bugs irroot (~gregory@2002:294f:dde2:1:b5ce:5681:7113:dadc) |
| 11:43.07 | *** part/#asterisk-bugs irroot (~gregory@2002:294f:dde2:1:b5ce:5681:7113:dadc) |
| 13:00.25 | *** join/#asterisk-bugs serafie (~erin@nat/digium/x-cytnqgsksxvhpqxu) |
| 13:07.20 | *** join/#asterisk-bugs mjordan (~mjordan@nat/digium/x-etdspjtkvwntnonh) |
| 13:07.20 | *** mode/#asterisk-bugs [+o mjordan] by ChanServ |
| 13:25.11 | *** join/#asterisk-bugs putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 13:25.11 | *** mode/#asterisk-bugs [+o putnopvut] by ChanServ |
| 14:28.16 | jsmith | As promised yesterday, a couple of backtraces from more segfaults we're seeing: |
| 14:28.18 | jsmith | http://fpaste.org/Ow7r/ |
| 14:28.26 | jsmith | http://fpaste.org/zOhJ/ |
| 14:29.26 | jsmith | and http://pastebin.com/RqQh8pYk |
| 14:31.58 | Qwell | always love the ast_frdrup() double frees. |
| 14:32.11 | Qwell | err, not double free. whatever. |
| 14:38.43 | jsmith | putnopvut, mjordan: Any thoughts on those three backtraces? |
| 14:41.10 | putnopvut | Welp, certainly different from the ones you showed before. I'm looking deeper into them right now. |
| 14:41.57 | putnopvut | All appear to be mixmonitor related. |
| 14:42.30 | putnopvut | I can try my handy-dandy "run some calls through mixmonitor while using valgrind" test to see if I can find anything odd. What arguments do you pass to MixMonitor? |
| 14:46.05 | mjordan | jsmith: are you using chan_agent? |
| 14:47.06 | jsmith | mjordan: Yes, I'm quite sure we are |
| 14:47.27 | mjordan | hm. |
| 14:47.31 | jsmith | putnopvut: I'll try to figure out |
| 14:53.03 | mjordan | jsmith: I can't seem to find the issue right now, but there was something reported somewhat recently about memory corruptions that only happened when chan_agent was in use. Given chan_agent's relative flakiness, that's not terribly surprising - although its odd that it would be causing issues consistently in mixmonitor |
| 14:53.12 | mjordan | still, something you may want to consider if you're looking for a work around |
| 14:56.17 | jsmith | I've got several hundred agents hotdesked in a call center -- not exactly simple to just stop using chan_agent, unfortunately :-( |
| 14:56.59 | jsmith | If it were my own custom installation, I'd go for it -- but this is using some custom front-ends and configuration generation scripts, and I'm afraid I'd break them if I tried to not use chan_agent |
| 14:58.56 | mjordan | kk |
| 15:18.26 | *** join/#asterisk-bugs Juggie (~Juggie@unaffiliated/juggie) |
| 15:18.26 | *** mode/#asterisk-bugs [+o Juggie] by ChanServ |
| 15:33.42 | putnopvut | As I feared. My simple valgrind tests of "login agent. Call agent with mixmonitor. Have agent transfer the call" got nothing out of valgrind :( |
| 16:49.14 | *** join/#asterisk-bugs leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 16:49.14 | *** mode/#asterisk-bugs [+o leifmadsen] by ChanServ |
| 17:06.49 | *** join/#asterisk-bugs b3nt_pin (~quassel@stjhnf0148w-142134201057.dhcp-dynamic.FibreOp.nl.bellaliant.net) |
| 20:29.21 | *** join/#asterisk-bugs n0tk (~n0tk@gateway/tor-sasl/n0tk) |
| 20:37.37 | *** join/#asterisk-bugs jrose_atDigium (~jrose_atD@nat/digium/x-trhonelaoqxybyxm) |
| 21:48.42 | *** part/#asterisk-bugs mjordan (~mjordan@nat/digium/x-etdspjtkvwntnonh) |
| 22:53.15 | *** join/#asterisk-bugs tzafrir_laptop (~tzafrir@212.179.75.202) |