| 00:00.04 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-fdzhaghbywwrgali) |
| 01:42.41 | *** join/#asterisk-dev FunnymanVA (~funnymanv@pool-71-171-125-245.clppva.fios.verizon.net) |
| 01:54.51 | *** join/#asterisk-dev file (~file@asterisk/developer-and-muffin-lover/file) |
| 01:54.51 | *** mode/#asterisk-dev [+o file] by ChanServ |
| 02:17.07 | *** join/#asterisk-dev beefcafe (~beefcafe@unaffiliated/beefcafe) |
| 02:17.18 | beefcafe | what should I do to run asterisk 11 as non-root? right now "service asterisk start" shows it started but in fact it didn't. using sudo before that works |
| 03:24.10 | pabelanger | beefcafe: #asterisk |
| 03:24.14 | pabelanger | and don't cross post |
| 03:24.59 | beefcafe | could you help me out there? |
| 03:27.43 | zerohalo | beefcafe: see #asterisk |
| 03:28.50 | beefcafe | seen |
| 03:36.32 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-nklfmgmeuphppnbn) |
| 03:36.32 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 03:52.11 | beefcafe | will there be ipv6 support for chan motif? |
| 04:05.45 | opticron | leedm777, ping |
| 04:06.18 | opticron | I've run into further issues trying to remove the race condition |
| 04:07.01 | opticron | I think I'm most of the way there, but there seems to be some memory corruption that I've caused along the way |
| 04:29.00 | *** join/#asterisk-dev GameGamer43 (uid5533@gateway/web/irccloud.com/x-dxvvxpbtrtlgduah) |
| 06:01.03 | *** join/#asterisk-dev tonyclewis (uid6025@gateway/web/irccloud.com/x-mkvjweygdjdxyrqs) |
| 06:24.31 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 06:24.31 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:58.33 | *** join/#asterisk-dev schmidts (~schmidts@vie-91-186-159-076.dsl.sil.at) |
| 07:58.36 | schmidts | good morning |
| 08:17.00 | *** join/#asterisk-dev saghul (~saghul@ip3e830637.speed.planet.nl) |
| 08:32.23 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 08:32.24 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 09:40.14 | *** join/#asterisk-dev jmls (~jmls@77.107.171.82) |
| 10:24.54 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 10:24.54 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 10:51.49 | *** join/#asterisk-dev aurs (~aurs@110.84-49-69.nextgentel.com) |
| 12:48.24 | *** join/#asterisk-dev jkroon (~jkroon@dsl-244-34-204.telkomadsl.co.za) |
| 13:11.01 | *** join/#asterisk-dev caio1982 (~caio1982@189.125.242.3) |
| 13:24.33 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-nklfmgmeuphppnbn) |
| 13:24.47 | jkroon | hi guys, segfault bug in pbx_dundi, bug report and patch available @ https://issues.asterisk.org/jira/browse/ASTERISK-21205 - any chance of a quick review? |
| 13:42.34 | *** join/#asterisk-dev leifmadsen (~lmadsen@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:42.34 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 13:55.10 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-iasscynfakqnxeut) |
| 14:00.29 | *** join/#asterisk-dev leifmadsen (~lmadsen@asterisk/documenteur-extraordinaire/blitzrage) |
| 14:00.33 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 14:20.07 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 14:20.07 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 14:30.29 | *** join/#asterisk-dev serafie (~erin@24.214.158.242) |
| 14:31.40 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:31.40 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:36.01 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-cesjmkoomsqvhiga) |
| 14:36.01 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:44.45 | *** part/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 15:27.39 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 15:27.39 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 15:30.31 | opticron | leedm777, so I now have a solution to the memory corruption I've been seeing with the caching unit tests |
| 15:31.06 | leedm777 | opticron: cool. I think I also have a solution. |
| 15:31.12 | leedm777 | I wonder how similar they are :-) |
| 15:31.25 | opticron | it was actually a bug in the test |
| 15:31.45 | opticron | end result: everything that the callback needs to function properly needs to be ensured to survive until the callback can no longer be called |
| 15:32.03 | opticron | which means until the last subscription change notification comes through |
| 15:32.13 | leedm777 | ooo |
| 15:32.51 | opticron | so if you have a refcounted object that you're tossing into the void* of the sub, you need to bump the refcount and let the callback take care of the final deref |
| 15:33.18 | opticron | and if it's not refcounted and the callback is the last thing to use it, the callback gets to destroy it on that final message |
| 15:33.42 | opticron | what was your solution? |
| 15:33.43 | leedm777 | actually, since we know that the unsubscribe is going to be the final message |
| 15:33.49 | leedm777 | it can destroy/unref as appropriate |
| 15:34.27 | opticron | right, but the trick there is to tell the callback what the uniqueid for its subscription was so that it knows which unsub message belongs to it |
| 15:34.37 | opticron | which is something that the void* currently has to convey |
| 15:34.59 | opticron | but which really seems like it should be a bit more readily available |
| 15:35.31 | leedm777 | We could pass the subscription to the callback... |
| 15:35.44 | opticron | we could |
| 15:36.11 | opticron | since the dispatch holds a ref to it anyway |
| 15:36.18 | leedm777 | right |
| 15:36.54 | opticron | I like that solution better than my current hackery |
| 15:37.19 | leedm777 | And it's different from the problem that I found :-) |
| 15:37.29 | opticron | :) |
| 15:37.41 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 15:37.41 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 15:37.50 | leedm777 | I've got a patch I'm currently merging into your team branch |
| 15:38.06 | leedm777 | but doesn't address the void* lifecycle problem you found |
| 15:38.12 | opticron | waits for it to break his world |
| 15:38.35 | leedm777 | I hope it actually mends your broken world |
| 15:38.38 | leedm777 | or something |
| 15:38.40 | opticron | :) |
| 15:38.56 | leedm777 | waits for the world to compile |
| 15:39.14 | sruffell | stabs opticron's world |
| 15:39.47 | opticron | does the sruffell shuffle to dodge |
| 15:39.56 | sruffell | well played…well played... |
| 15:46.15 | mjordan | history question: why are endbeforehexten and initiatedseconds global Asterisk options? |
| 15:50.15 | file | mjordan, yes. |
| 15:50.56 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 15:50.56 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 15:52.07 | mjordan | file: exactly |
| 15:54.27 | opticron | Mind. Blown. |
| 15:55.04 | mjordan | and yet another undocumented configuration option found. |
| 16:04.57 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:07.09 | pabelanger | mjordan, stay out of chan_iax2 |
| 16:20.18 | mjordan | ? |
| 16:22.18 | pabelanger | <mjordan> and yet another undocumented configuration option found. |
| 16:24.07 | *** join/#asterisk-dev Gugge (gugge@kriminel.dk) |
| 16:34.38 | *** join/#asterisk-dev tm1000 (tm1000@2600:3c01::f03c:91ff:fe70:fce6) |
| 16:39.18 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 16:39.19 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 17:10.49 | mjordan | pabelanger: interestingly enough, it's in cdr |
| 17:11.09 | Qwell | I am shocked, *shocked*, that CDR has undocumented features. |
| 17:11.36 | *** join/#asterisk-dev sgimeno (~sgimeno@163.117.206.10) |
| 17:45.53 | *** join/#asterisk-dev caio1982 (~caio1982@189.125.242.3) |
| 18:04.53 | Greenlight | Is there much overhead using a frame hook on all channels? The hook itself is only checking for a single frame type which happens early in the call, and ignoring the rest. Just wondering if it's worth having the hook destroy itself, or if it'll not make much different. System has around 400-500 channels at any one time. |
| 18:10.52 | file | shouldn't be |
| 18:13.31 | Greenlight | Okay, thanks, will just leave it as is then :) |
| 18:25.23 | *** join/#asterisk-dev caio1982 (~caio1982@189.125.242.3) |
| 19:10.34 | leedm777 | opticron: I'm going to try to get rid of the out param in creating a caching_topic |
| 19:11.38 | opticron | I welcome that change |
| 19:12.00 | opticron | because having that out param is uuuuugly |
| 20:36.17 | *** join/#asterisk-dev oej (~Adium@h87-96-134-129.dynamic.se.alltele.net) |
| 20:36.17 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 22:08.43 | *** join/#asterisk-dev pabelanger (~pabelange@asterisk/contributor-and-bug-marshal/pabelanger) |
| 22:08.43 | *** mode/#asterisk-dev [+o pabelanger] by ChanServ |
| 22:15.16 | seanbright | anyone at digium want to throw me a discount code for 4 transcoder cards? |
| 22:15.24 | seanbright | buddies... pals... friends... |
| 22:16.20 | leifmadsen | KEEPBNGAWSM |
| 22:16.34 | file | I can give you a discount of 0% |
| 22:16.53 | leifmadsen | I can give you a discount of -10%, balance payable to Leif Madsen |
| 22:17.34 | pabelanger | boo |
| 22:17.52 | seanbright | boo what? |
| 22:18.11 | pabelanger | boo nobody gave out a really coupon |
| 22:18.16 | pabelanger | :) |
| 22:18.26 | leifmadsen | I don't have that kind of access :D |
| 22:18.42 | seanbright | heh |
| 22:19.18 | seanbright | does digium have corporate sales reps? |
| 22:19.32 | opticron | has no idea |
| 22:20.03 | opticron | malcolm might know, but he left about 10 minutes ago |
| 22:33.17 | pabelanger | Hmm, if I could use the transcoding card out side asterisk, I might get one |
| 22:33.21 | pabelanger | but don't think it is possible |
| 22:37.36 | seanbright | why couldn't you? |
| 22:38.04 | pabelanger | not sure, is it possible? |
| 22:38.08 | seanbright | i had a mostly-working utility to do it |
| 22:38.14 | seanbright | was on reviewboard for a period |
| 22:38.19 | pabelanger | interesting |
| 22:39.38 | seanbright | where would you use it outside of asterisk? |
| 22:47.28 | pabelanger | seanbright: don't have PCI access in my vm |
| 22:47.39 | seanbright | oh... |
| 22:47.50 | pabelanger | so, would need to proxy rtp to another system for transcoding |
| 22:47.51 | seanbright | well then how would you use a pci card at all? |
| 22:47.55 | seanbright | ugh |
| 22:48.06 | seanbright | sangoma has an appliance for that i believe |
| 22:48.44 | pabelanger | Ya, haven't really researched it... but was just curious about it |
| 22:48.56 | seanbright | http://www.sangoma.com/products/netborder-transcoding-appliance/ |
| 22:49.03 | seanbright | if you're into that kinda thing... |
| 22:49.50 | pabelanger | Hmm, ya, would like to build my own but let me check it out |
| 22:49.57 | leifmadsen | ya I'd just build my own appliance out of an atom board with a pci slot or something :D |
| 22:50.20 | seanbright | true... that's trivial |
| 22:51.04 | pabelanger | Ya, I don't even know how much CPU saving there is with a DSP card.... but I suspect a lot |
| 22:51.19 | seanbright | for the card and the atom (and a case...) you're looking at 1200 USD to start |
| 22:52.11 | pabelanger | nods |
| 22:52.34 | pabelanger | not at the point of breaking out RTP / transcoding. But might do so in the next few months |
| 23:21.52 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 23:21.53 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |