| 05:53.46 | *** join/#asterisk-dev pnlarsson (~niklas@fw1.gml.g.icnet.infracom.se) |
| 06:33.51 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:78cb:b72a:f9b1:8ed4) |
| 08:47.40 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 10:50.19 | *** join/#asterisk-dev sgimeno (~sgimeno@163.117.206.10) |
| 11:24.06 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@local.xorcom.com) |
| 11:24.06 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 11:33.32 | *** join/#asterisk-dev michael_work (~michael@bzq-82-168-31-134.red.bezeqint.net) |
| 11:40.33 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 12:09.25 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 12:58.44 | wdoekes | where is putnopvut? |
| 13:10.57 | file | he's been out, and it's a holiday in the US today |
| 13:26.02 | pnlarsson | wiki.asterisk.org == not responding |
| 13:47.03 | wdoekes | okay, tnx |
| 14:09.21 | michael_work | wdoekes, ping afain :P |
| 14:09.26 | michael_work | again |
| 14:14.56 | wdoekes | pong |
| 14:16.30 | michael_work | wdoekes, about ASTERISK-22083 |
| 14:26.32 | wdoekes | I thought there was going to be a question at this point |
| 14:26.35 | wdoekes | or a statement |
| 14:26.49 | wdoekes | :) |
| 14:31.25 | Corydon76-home | file: it's not a bank holiday in Canada, like it is in the UK? |
| 14:32.43 | [TK]D-Fender | Nope |
| 14:45.05 | leifmadsen | Corydon76-home: our holiday was last weekend for Victoria Day (LONG LIVE THE QUEEN) |
| 14:45.25 | Corydon76-home | Yes, yes, long live file... |
| 14:45.39 | leifmadsen | touche :) |
| 14:45.49 | michael_work | wdoekes, the answer to question is in the bug :) |
| 14:45.54 | michael_work | you asked to provide info |
| 14:46.08 | michael_work | but i'm noob and not sure how i do it with gdb |
| 14:46.23 | michael_work | as well i said there my theory of what's happening |
| 14:46.46 | michael_work | btw how can i register new account on jira as i want to stop using my boss's one :P |
| 14:47.06 | Corydon76-home | leifmadsen: btw, loving my new Ubiquiti PicoStation |
| 14:47.20 | leifmadsen | oic |
| 14:47.22 | leifmadsen | googles |
| 14:47.23 | leifmadsen | :) |
| 14:47.40 | Corydon76-home | It reaches all the way to the hammock in the backyard now |
| 14:47.45 | wdoekes | michael_work: your boss is Shlomi? |
| 14:47.57 | leifmadsen | Corydon76-home: ooooh |
| 14:48.01 | leifmadsen | I need one of these heh |
| 14:48.01 | michael_work | wdoekes, yes why? |
| 14:48.04 | wdoekes | that does clear up some confusion |
| 14:48.21 | michael_work | that's why i want my own account :P |
| 14:48.22 | wdoekes | I thought you were a different michael |
| 14:48.24 | leifmadsen | Corydon76-home: my backyard isn't that big, so the basement AP actually seems to reach |
| 14:48.41 | Corydon76-home | leifmadsen: I now have one AP for the entire property |
| 14:48.42 | michael_work | wdoekes, i confused few people here already :P |
| 14:49.26 | Corydon76-home | and it's 802.11n, so it scales properly with multiple devices |
| 14:50.37 | wdoekes | michael_work: the bug never mentions cachertclasses=yes nor realtime, until your last post. it's good to get that cleared up |
| 14:50.46 | wdoekes | so in your case, you can probably work around it by setting cachertclasses=no |
| 14:51.28 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 14:51.36 | wdoekes | unless that creates a performance hit, in which case we'd like to know that too |
| 14:52.18 | michael_work | wdoekes, i'm not 100% it's the case though. as well as it would hurt performance :( |
| 14:52.40 | wdoekes | as for the answer to the gdb question: a 'thread apply all bt' on the core dump, and then check the ```moh_scan_files (class=``` across the threads |
| 14:52.48 | michael_work | the problem is that i need to find the way to easily recreate it in order to poke around :P |
| 14:53.09 | wdoekes | you can easily recreate it by adding some sleep between the free's |
| 14:53.14 | wdoekes | (probably) |
| 14:53.53 | michael_work | by check moh classes you mean it would print all the threads and i just need to copy-paste in text and grep it? or it's also some magical command |
| 14:54.06 | wdoekes | grep indeed |
| 14:54.47 | michael_work | the problem with sleeps is that it wouldn't help as there would be sleeps in parallel thread. though i can create random between 1-3 secs sleep or somethign :) |
| 14:54.49 | wdoekes | ('thread apply all bt full' for a more verbose bt) |
| 14:54.58 | michael_work | thanks |
| 14:55.03 | wdoekes | michael_work: no, that shouldn't be a problem |
| 14:55.10 | michael_work | i have 3 or 4 cores i saved to check |
| 14:55.14 | wdoekes | you only need to get them into the same for loop at the same time |
| 14:57.11 | michael_work | as i understand it fall when it tries to free something that was freed else where |
| 14:57.17 | michael_work | so yes sleep would be okay |
| 14:57.51 | michael_work | it would be in sleep on second while second call would get there and would try to free something it was freed |
| 14:58.04 | michael_work | i'll try to find some time to check it |
| 15:26.46 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 16:08.24 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 18:04.44 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 18:18.26 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 19:47.03 | *** join/#asterisk-dev danjenkins (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 21:04.04 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 21:37.01 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@bzq-218-28-58.cablep.bezeqint.net) |
| 21:37.01 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |