| 00:17.55 | *** join/#asterisk-dev mjordan (~mjordan@user-69-1-6-49.knology.net) |
| 00:17.55 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 02:04.28 | *** join/#asterisk-dev jmls (~Julian@host217-36-208-155.in-addr.btopenworld.com) |
| 02:52.35 | *** part/#asterisk-dev mjordan (~mjordan@user-69-1-6-49.knology.net) |
| 04:03.17 | *** part/#asterisk-dev Kobaz (~kobaz@its.kobaz.net) |
| 04:03.24 | *** join/#asterisk-dev Kobaz (~kobaz@its.kobaz.net) |
| 04:31.15 | *** join/#asterisk-dev aimka (~aimka@unaffiliated/aimka) |
| 04:31.43 | *** join/#asterisk-dev oej (~olle@ns.webway.se) |
| 04:31.43 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 05:20.17 | *** join/#asterisk-dev gokulnath (~gokulnath@115.111.177.122) |
| 08:18.25 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@local.xorcom.com) |
| 08:18.26 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 08:20.02 | *** join/#asterisk-dev irroot (~irroot@pbx.distrotech.co.za) |
| 09:03.07 | *** join/#asterisk-dev hehol (~Adium@2a01:198:71d:0:21f:d0ff:fea1:568e) |
| 09:54.02 | *** join/#asterisk-dev acidfoo (~nib@modemcable094.94-70-69.static.videotron.ca) |
| 11:12.50 | *** join/#asterisk-dev jmls (~Julian@host217-36-208-155.in-addr.btopenworld.com) |
| 12:26.07 | jmls | no no no no |
| 12:26.20 | jmls | got a demo , getting segfaults |
| 12:26.21 | jmls | :( |
| 12:26.35 | jmls | #6 0x0808668f in internal_ao2_ref (user_data=0x9a780c8, delta=<value optimized out>) at astobj2.c:279 |
| 12:26.36 | jmls | <PROTECTED> |
| 12:26.36 | jmls | <PROTECTED> |
| 12:26.36 | jmls | <PROTECTED> |
| 12:26.36 | jmls | <PROTECTED> |
| 12:27.03 | tzafrir_laptop | pb the full trace somewhere? |
| 12:27.08 | jmls | okies |
| 12:29.42 | jmls | http://pastebin.com/jme4kwTu |
| 12:39.02 | irroot | lo folks |
| 12:40.01 | irroot | jmls you should build with DONT_OPTOMISE |
| 12:40.53 | irroot | jmls confbridge segfault on hangup ? |
| 12:43.00 | irroot | GRRR my techies are talking about getting me asterisk for dummies as gift |
| 12:49.22 | irroot | if (qe->parent->monfmt && *qe->parent->monfmt) { <- hehe guess ast_strlen_zero is better to use |
| 12:49.59 | jmls | the one and only bloody time I didn't set that flag ... |
| 12:50.04 | jmls | set now, recompiling |
| 12:50.16 | jmls | irroot: looks like it |
| 12:50.21 | irroot | jmls beware of the mighty murphy |
| 13:01.55 | jmls | irroot: you any good with networks and problems ? |
| 13:02.03 | jmls | needs some help ;) |
| 13:02.33 | irroot | jmls do i get to use high explosives ?? can i nomintate a network ??/ |
| 13:02.43 | irroot | jmls what up |
| 13:03.49 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-tpfenflnxemrfczt) |
| 13:03.49 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 13:15.58 | irroot | mjordan greets kind sir |
| 13:16.36 | mjordan | irroot: hello |
| 13:55.34 | jmls | anyone got any ideas about what caused the crash, or do we need the non-optimised version ? |
| 13:55.41 | jmls | http://pastebin.com/jme4kwTu |
| 14:01.31 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-tpfenflnxemrfczt) |
| 14:01.32 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-tpfenflnxemrfczt) |
| 14:01.32 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:01.46 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-alnzzajhrvupnvzf) |
| 14:01.46 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:07.13 | jmls | leifmadsen: I will get round to it as soon as this week is over ;) |
| 14:07.21 | irroot | jmls need unoptomized indeed |
| 14:07.35 | irroot | leifmadsen greets |
| 14:08.22 | irroot | im wondering if mnicholson is arround and is keen to test something |
| 14:09.51 | jmls | is hoping that he will never get to have to post an unoptimized bt ;) |
| 14:10.03 | jmls | coz it's all bug free, right ... |
| 14:11.23 | irroot | yes and just this morning i stoped world hunger while starting up my perpetual motion machine running on cold fusion |
| 14:13.53 | jmls | damn you sir. You stole all my ideas |
| 14:14.12 | jmls | wait |
| 14:14.32 | jmls | if you had a pp machine, it wouldn't need to be running on cold fusion |
| 14:14.43 | jmls | wasn't that web design software, anyway ? |
| 14:30.13 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:30.13 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 14:31.59 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 14:32.00 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:37.19 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:37.19 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:37.47 | irroot | putnopvut hi there know you did quite a bit on app_queue ive made some substantial changes to it |
| 14:40.17 | putnopvut | irroot: cool. What have you been changing? |
| 14:41.10 | irroot | putnopvut added a state object shared by members for device state updates more efficient when having multiple logins from members to different queues |
| 14:41.36 | irroot | made locking work at member and queue level so q locks arent held traversing members |
| 14:42.25 | irroot | and the big one is wrap calls to qe->parent-> in locks to make it thread safe and remove locks protecting members replace with per member locks |
| 14:42.37 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:bd2d:c3c:81d:b47e) |
| 14:43.13 | putnopvut | Sounds like some good ideas :) |
| 14:43.18 | irroot | all member add is done by one function now using ast_config struct |
| 14:43.40 | irroot | have dynamic / realtime / static members that play together |
| 14:44.19 | sruffell | tzafrir_laptop: any reason you have to not tag 2.6.0-rc1 ? |
| 14:44.59 | jmls | on one box make menuselect gives me a nice pretty 2-pane coloured view. On another box, I've got the old style. Both are asterisk-10. |
| 14:45.07 | jmls | what package am I missing ? |
| 14:45.17 | putnopvut | jmls: probably libnewt-dev |
| 14:45.30 | jmls | tries that |
| 14:45.58 | putnopvut | Naturally you'll need to re-run the configure script after installing the library. |
| 14:46.00 | jmls | No package libnewt-dev available |
| 14:46.12 | jmls | Package newt-0.52.11-2.el6.i686 already installed and latest version |
| 14:46.17 | Qwell | Silly Fedorians need to add a "el" to the -dev. :) |
| 14:46.26 | irroot | putnopvut unified diff of 5269 lines so far |
| 14:46.31 | putnopvut | Ah, didn't realize he was a fedorian. |
| 14:46.44 | jmls | um, centos ... |
| 14:46.47 | Qwell | same thing |
| 14:46.49 | jmls | I know |
| 14:46.50 | jmls | ;) |
| 14:46.55 | jmls | Centosian |
| 14:47.00 | Qwell | RH calls their packages devel, whereas debian uses dev |
| 14:47.22 | jmls | No package libnewt-devel available. |
| 14:47.23 | irroot | uses own from source distro call me luke i use the source |
| 14:47.31 | Qwell | might just be newt-devel |
| 14:47.32 | jmls | groan |
| 14:47.41 | putnopvut | silly package names! |
| 14:47.47 | jmls | bingo |
| 14:47.53 | putnopvut | It's called bingo? |
| 14:47.54 | tzafrir_laptop | sruffell, I posted on the list a suggestion for userspace part of the pinned-spans |
| 14:48.22 | tzafrir_laptop | Though I figure most of it actually belongs in dahdi-linux, as it is linux-specific |
| 14:48.46 | jmls | ahhh. pretty... |
| 14:49.12 | tzafrir_laptop | I'd like to see basic optional userspace scriptary in place before 2.6.0-rc1 |
| 14:49.23 | sruffell | tzafrir_laptop: I'm operating under the assumption that the user-space part will be for 2.7……2.6 will have just the kernel parts. |
| 14:49.29 | sruffell | hmm... |
| 14:49.31 | sruffell | thinks |
| 14:50.58 | sruffell | My thoughts that we didn't want to start "recommending" any particular way to set up the user space scripts until we've had a little more time…that way we don't end up supporting / teaching some way that wasn't going to be final. Since the kernel space parts are in..nothing *prevents* someone from adding their own user space scripts….I just didn't want to set in stone (soft it may be) how we're going to do that when there might n |
| 14:51.00 | sruffell | discussion. |
| 14:52.17 | tzafrir_laptop | Most of the relevant changes there are scriptary |
| 14:52.30 | tzafrir_laptop | One part is a small change to dahdi_cfg |
| 14:52.48 | tzafrir_laptop | Make it work with specified spans |
| 14:53.01 | sruffell | are you referring to the -k flag? |
| 14:53.10 | sruffell | I think that is probably safe.... |
| 14:53.14 | tzafrir_laptop | In that patch it is actually -C and -S |
| 14:53.20 | jmls | ok. gmime ? |
| 14:53.22 | sruffell | since it doesn't change any of the configuration files that a user would be expected to edit. |
| 14:53.28 | sruffell | looks |
| 14:54.15 | tzafrir_laptop | used by the httpd to detect mime types |
| 14:54.31 | irroot | dont give users a choice :P |
| 14:55.56 | tzafrir_laptop | While we're at it, I have a patch to remove menuselect from dahdi-tools: |
| 14:55.56 | sruffell | irroot: :) yes…as much as possible I think users should have a choice….. |
| 14:55.59 | tzafrir_laptop | https://gitorious.org/~tzafrir/asterisk-tools/tzafrirs-dahdi-tools/commits/nomenu |
| 14:56.06 | sruffell | should *not * have a choice that is. |
| 14:56.19 | sruffell | it should "just work" |
| 14:56.34 | irroot | sruffell users f*** it up |
| 14:56.36 | tzafrir_laptop | There are only two '<depend>' tags. I replaced them with explicit makefile conditions |
| 14:56.50 | sruffell | tzafrir_laptop: ahh... |
| 14:56.51 | tzafrir_laptop | (which is what is already done in xpp/Makefile ) |
| 14:57.02 | sruffell | ok..let me look at that….dropping the external would be "very nice' |
| 14:57.29 | sruffell | I had thought we were holding that up until the config script drops sethdlc on gentoo…but it sounds like you might already have that covered. |
| 14:58.03 | tzafrir_laptop | Looking at the menuselect-in-dahdi-tools thread from January, the only missing feature is a way to disable building of "everything except dahdi_cfg" |
| 14:58.26 | tzafrir_laptop | I guess it can be done by explicitly forcing the "BINS" variable at build time. |
| 14:59.10 | tzafrir_laptop | The config script detects the availability of hdlc, right? So why not use it? |
| 14:59.40 | tzafrir_laptop | (If it's buggy, it should be fixed regardless, as menuselect relies on this right now) |
| 15:00.51 | sruffell | I'm looking for the email thread…but basically gentoo hacks some of the headers..... |
| 15:06.16 | tzafrir_laptop | http://lists.digium.com/pipermail/asterisk-dev/2011-January/thread.html#47672 |
| 15:09.39 | tzafrir_laptop | The configure scripts tries to compile a C program to check for the HDLC headers. |
| 15:10.06 | tzafrir_laptop | We might as well make it slightly more complex to fail the Gentoo case |
| 15:16.34 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:16.44 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:16.54 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:17.03 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:17.14 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:17.23 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:17.32 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:18.30 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:18.40 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:18.50 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:19.01 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:19.11 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:19.21 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:19.30 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:19.39 | *** join/#asterisk-dev Gugge (~gugge@46.30.120.1) |
| 15:32.55 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:32.55 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:40.17 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:00.54 | sruffell | tzafrir_laptop: strange…when I try to look at nomenu I see "This Git object is too large to be displayed in the browser Consider cloning the repository locally and look at the object there" |
| 16:01.02 | sruffell | have you seen that? |
| 16:01.15 | tzafrir_laptop | no |
| 16:01.18 | tzafrir_laptop | let me try |
| 16:06.06 | tzafrir_laptop | Looks fine: https://gitorious.org/~tzafrir/asterisk-tools/tzafrirs-dahdi-tools/commit/20ff1cc32ce741cbf4d3ab6ff18619ee51e0c144 |
| 16:06.50 | sruffell | hmm…strange…that worked. |
| 16:15.07 | *** join/#asterisk-dev irroot (~irroot@197.169.77.91) |
| 16:17.49 | irroot | mnicholson ping |
| 16:20.48 | mnicholson | irroot, pong |
| 16:21.21 | irroot | yo dude got a new app_queue patch up if you dont mind burning it |
| 16:22.06 | irroot | it seperates locking i need to test and eliminate any deadlocks but hoping it does not hurt performance |
| 16:25.39 | mnicholson | irroot, i'll give it a shot |
| 16:28.57 | irroot | mnicholson there is no doubt some deadlocks that have to be iorned out as there multiple layers now queues container/queue/queue members container/member/device and seperate the devices container |
| 16:30.14 | irroot | also put all calls to read a ao2 object in locks as they should be |
| 16:53.17 | *** join/#asterisk-dev irroot (~irroot@197.171.91.242) |
| 16:55.32 | *** join/#asterisk-dev tsearle (~tsearle@host-85-27-35-157.brutele.be) |
| 16:56.38 | mnicholson | irroot, are you building with --enable-dev-mode? |
| 16:57.00 | irroot | should be |
| 16:57.12 | irroot | ill double check clearly you spotted something |
| 16:57.39 | irroot | ill fix it |
| 17:05.38 | tzafrir_laptop | Is twilson around? |
| 17:08.50 | *** join/#asterisk-dev irroot (~irroot@197.169.254.221) |
| 17:09.34 | mnicholson | irroot, I fixed a few bugs in your patch that prevent it from building when --enable-dev-mode is on, and am testing it now |
| 17:10.10 | irroot | cool im testing this side |
| 17:13.02 | mnicholson | irroot, there is a performance regression in here somewhere |
| 17:13.09 | irroot | damm |
| 17:15.03 | irroot | what you throw at it |
| 17:16.01 | mnicholson | irroot, our test involves 380 agents logged in. They are then sent 380 calls. |
| 17:16.12 | irroot | ok cool |
| 17:16.17 | irroot | all realtime ? |
| 17:16.26 | mnicholson | no realtime |
| 17:16.46 | irroot | im useing memcpy to copy the devstate to have it consistant |
| 17:17.33 | irroot | that could be a problem |
| 17:17.36 | mnicholson | no deadlocks though |
| 17:17.46 | irroot | i think i hit one in the realtime |
| 17:17.54 | mnicholson | at least not with one run through |
| 17:18.16 | mnicholson | i see |
| 17:18.17 | irroot | im getting a core and dbg |
| 17:18.55 | irroot | so if i dont memcpy the state dev it may improve |
| 17:19.04 | irroot | esp at that rate of calls |
| 17:19.20 | irroot | pass a ref to the devstate callback like it was |
| 17:19.38 | irroot | all the extra locking can cause performance problems |
| 17:23.10 | mnicholson | i don't understand anything about why the performance degrades, but you sound like you may have a handle on it |
| 17:47.39 | irroot | <PROTECTED> |
| 17:48.38 | mnicholson | irroot, np |
| 18:03.03 | kpfleming | irroot: ping |
| 18:03.12 | irroot | kpfleming pong |
| 18:03.43 | kpfleming | in the future, when you have work that you plan to do that is as extensive as the app_queue changes appear to be, it is recommended that you do that work in a developer branch *only* |
| 18:04.02 | irroot | kpfleming it was not meant to be so big |
| 18:04.13 | kpfleming | it shouldn't go on reviewboard until you have done your own testing and the code has settled down... otherwise time that reviewers spend on it could be wasted when the code gets changed again |
| 18:04.15 | irroot | ill get a branch done |
| 18:04.38 | *** join/#asterisk-dev jmls (~Julian@host217-36-208-155.in-addr.btopenworld.com) |
| 18:04.44 | irroot | was not meant to be a big change :P |
| 18:04.47 | kpfleming | understood... it's always safer to start in a branch and then go to reviewboard after you finish your own testing... that way if it's going to get bigger, that will happen before your testing is done :-) |
| 18:05.05 | kpfleming | also, having it in a branch makes it easier for others to help test it |
| 18:06.51 | file | and for each branch you create you get a scratch card that may give you valuable prizes! or nothing at all. |
| 18:07.31 | irroot | file double points for enableing and leaving automerge on when not needed .... |
| 18:08.06 | irroot | winners will be chosen from automerge updates randomly so the more the higher the chances |
| 18:59.39 | *** join/#asterisk-dev mnicholson_ (~mnicholso@nat/digium/x-hfwhexcnabtoyrpa) |
| 18:59.39 | *** mode/#asterisk-dev [+o mnicholson_] by ChanServ |
| 19:00.26 | irroot | mnicholson_ learnt something valuable dont ever try link or unlink a object or perform any callback while locked :P |
| 19:01.24 | mnicholson_ | irroot: hmm news to use |
| 19:01.48 | irroot | unless you hold the container lock |
| 19:02.29 | irroot | iterator returns object lock it then unlink and you deadlock with a callback in other thread |
| 19:02.30 | mnicholson_ | ahh, unlink grabs the container lock and the object lock |
| 19:02.38 | irroot | ta da boom |
| 19:03.45 | irroot | but you have the obj locked and then container so in a callback in thread 2 you have container and then it will try lock the obj with container not going to end pretty |
| 19:04.26 | kpfleming | that's a violation of locking order anyway... it always has to be container and then object, if you are going to hold both locks |
| 19:04.36 | irroot | indeed |
| 19:06.08 | irroot | as the containers are opaque most the time its not so easy to spot |
| 19:18.38 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-dpmldxfufntpzwvv) |
| 19:18.38 | *** mode/#asterisk-dev [+o mnicholson] by ChanServ |
| 19:30.12 | tzafrir_laptop | gave up on trying to figure out AST-2011-012 . Just doesn't make sense |
| 19:30.17 | tzafrir_laptop | heads home |
| 19:48.12 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 19:48.16 | *** part/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 20:14.15 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 20:14.19 | *** part/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 20:27.39 | *** join/#asterisk-dev otherwiseguy (~otherwise@69.76.221.79) |
| 20:28.22 | otherwiseguy | wdoekes2: congrats on the committing :-) |
| 20:29.57 | wdoekes2 | thank you :) |
| 20:30.11 | wdoekes2 | I'm preparing a commit message for r1395 right now |
| 20:36.03 | otherwiseguy | wdoekes2: good. I'll be happy to be able to close the internal issue linked to that issue that is in my monthly sprint. :-) |
| 20:37.02 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 20:37.07 | *** part/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 20:42.18 | leifmadsen | w00t |
| 20:43.07 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@bzq-218-155-145.cablep.bezeqint.net) |
| 20:43.08 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 20:43.32 | irroot | +1 |
| 20:44.10 | irroot | leifmadsen i have not trolled the bugs in a while seen you been busy got pile in the RSS feed |
| 20:44.13 | irroot | is off to bed |
| 20:44.16 | leifmadsen | :) |
| 20:44.19 | leifmadsen | g'night! |
| 20:44.36 | leifmadsen | ya had to catch up on about 3 weeks worth of issues; got most of the big ones triaged. Still more to go unfortunately. |
| 20:57.41 | *** join/#asterisk-dev kb1_kanobe (~kris@bdr2.fieldrd.scrd.ca) |
| 20:58.46 | *** join/#asterisk-dev citywok (~citywok@67-134-194-33.dia.static.qwest.net) |
| 21:16.57 | kb1_kanobe | Hey all. Code formatting question: I'm looking at doing some work in chan_unistim.c and was going to tidy up the tabs/whitespace to meet the Coding Guidelines recommendations. However, I was wondering about the prefered placement of comments? |
| 21:17.48 | kb1_kanobe | Some structs have a line of code with a descriptive comment on the next line, others have the code and comment on the same line. In all cases the comment is a brief single line description. |
| 21:19.24 | kb1_kanobe | Is there a prefered layout? Personally I like the all-on-one-line for readability as the structs are quite large. |
| 21:22.33 | jmls | gasp. going to throw away my 30+ slide presentation about asterisk, replace it with a 10:20:30 (10 slide, 20 minutes, no text < 30 points) and do the rest of the hour session as a live demo .... I hope there is no segfaults :-D |
| 21:22.47 | jmls | wish me luck |
| 21:33.37 | wdoekes2 | kb1_kanobe: I haven't come across descriptive comments *after* code lines in the asterisk source yet. |
| 21:35.20 | wdoekes2 | i.e. I'm used to them coming before or on the same line |
| 21:35.21 | kb1_kanobe | My bad, that should have implied 'before'. |
| 21:36.06 | kb1_kanobe | In chan_unistim.c I'm comparing the presentation of 'struct unistim_line' to 'static struct unistim_device' |
| 21:36.26 | kb1_kanobe | if there's no big preference I can always just leave well enough alone. |
| 21:37.28 | otherwiseguy | kpfleming and others: the user enumeration due to nat= setting differences discussion on the -dev mailing list seems to have kind of died. The "last options standing" seem to be moving the global default to nat=yes and warning if the global and peer settings are different OR sending replies to both the port in the via and the port from which we received the request. Anyone have any more discussion as to what they think would be best? |
| 21:38.17 | otherwiseguy | On the plus side, sending to both ports no matter what completely fixes the issue. On the negative side, all I can really come up with is that it feels really icky. |
| 21:38.17 | wdoekes2 | kb1_kanobe: in sip.h I see them only on the same line, so that'd be my preference as well. however: I have no authority in the matter |
| 21:38.50 | otherwiseguy | nat=yes as the default could be done no matter what, though. |
| 21:40.11 | kb1_kanobe | wdoekes2: the commits specific to chan_unstim.c have been pretty quiet since 2008 so I might just jump in and reformat it all as single-line. |
| 21:47.12 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-alnzzajhrvupnvzf) |
| 22:38.43 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 22:38.44 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 22:48.57 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 22:48.57 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 23:00.46 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 23:00.46 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 23:16.03 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 23:16.03 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 23:24.52 | *** join/#asterisk-dev chazzam (~chazz@173-24-236-90.client.mchsi.com) |
| 23:37.26 | *** join/#asterisk-dev ChrisInSydney (~Chris@60-242-81-231.tpgi.com.au) |
| 23:44.44 | *** join/#asterisk-dev datachomper (~Adium@75.76.6.182) |
| 23:48.40 | *** part/#asterisk-dev kb1_kanobe (~kris@bdr2.fieldrd.scrd.ca) |
| 23:53.50 | *** part/#asterisk-dev datachomper (~Adium@75.76.6.182) |