| 01:01.27 | *** join/#asterisk-dev felipe_ (n=felipe@my.nada.kth.se) |
| 01:07.22 | *** part/#asterisk-dev securevoip (n=securevo@c-76-123-20-170.hsd1.va.comcast.net) |
| 01:12.29 | *** join/#asterisk-dev Deeewayne (n=dwayne@76.29.245.9) |
| 01:12.29 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 01:51.03 | *** join/#asterisk-dev snuff-home (n=snuffy@203-219-29-182-bri-ts6-2600.tpgi.com.au) |
| 01:51.03 | *** mode/#asterisk-dev [+o snuff-home] by irc.freenode.net |
| 01:51.10 | *** join/#asterisk-dev micols (n=mio@rlogin.dk) |
| 02:12.42 | *** join/#asterisk-dev eliel (n=eliels@120-17-235-201.fibertel.com.ar) |
| 02:12.42 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 02:54.50 | *** join/#asterisk-dev moy (n=moy@CPE001cdfec4cee-CM00080dab8485.cpe.net.cable.rogers.com) |
| 03:19.08 | *** part/#asterisk-dev sruffell (n=sruffell@asterisk/the-kernel-guy/sruffell) |
| 03:32.43 | *** join/#asterisk-dev CunningPike (n=CunningP@S01060014bf81366b.vc.shawcable.net) |
| 04:10.57 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 04:10.57 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 05:53.30 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 05:53.30 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 06:10.37 | *** join/#asterisk-dev sergee (n=serg@195.94.224.197) |
| 07:09.48 | *** join/#asterisk-dev tamiel (n=tamiel@213.30.183.226) |
| 07:12.27 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 09:08.04 | *** join/#asterisk-dev sergee (n=serg@195.94.224.197) |
| 09:23.34 | *** join/#asterisk-dev maverick (n=maverick@83.240.229.38) |
| 09:33.02 | *** join/#asterisk-dev michael-i (n=michael-@141.41.40.153) |
| 10:33.41 | *** join/#asterisk-dev ghenry (n=ghenry@pdpc/supporter/monthlybyte/ghenry) |
| 10:34.03 | ghenry | Hi, is a commit message like so "https://issues.asterisk.org/view.php?id=15156" ok? |
| 10:38.45 | tzafrir_laptop | ghenry, err... what commit message? |
| 10:39.00 | ghenry | I mean jsut a link to the bug? |
| 10:39.17 | tzafrir_laptop | Use 'generate commit template' |
| 10:39.52 | ghenry | Sorry? |
| 10:41.27 | tzafrir_laptop | Generally when you mention 'issue #15156' in a commit message, the svnbot will record the commit message in the bug log |
| 10:41.44 | ghenry | ah, cool. |
| 10:41.57 | tzafrir_laptop | And 'closes issue #15156' in a commit message will make it close the bug |
| 10:42.10 | ghenry | nice, ok. |
| 10:42.22 | ghenry | This closes that bug, so I'll do the last one. |
| 10:42.24 | ghenry | thanks. |
| 10:43.13 | ghenry | It didn't mention that at http://www.asterisk.org/developers/getting-started so wasn't sure |
| 10:52.21 | mvanbaak | ghenry: if you are looking at the bug on issues.asterisk.org, there's an option to generate a commit message |
| 10:52.37 | mvanbaak | Please have a look at that |
| 10:53.34 | ghenry | Where abouts? |
| 10:58.01 | tzafrir_laptop | above "relationship" and all the comments |
| 10:58.43 | tzafrir_laptop | that commit message mentions that it closes the issue, but doesn't give any further details |
| 10:58.45 | ghenry | thanks |
| 11:00.23 | ghenry | so when I use "generate commit template" it will use that template as long as my commit includes "issue #blah" ? |
| 11:03.30 | tzafrir_laptop | no. It will just generate a web page with a template for you to base your commit message on (through standard copy/paste) |
| 11:03.55 | tzafrir_laptop | pressing that button will not commit anything or change the state of the bug in any way |
| 11:06.41 | *** join/#asterisk-dev snuff-home (n=snuffy@203-219-29-182-bri-ts6-2600.tpgi.com.au) |
| 11:06.42 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 11:08.11 | ghenry | ok, thanks. Sorry for dump questions. |
| 11:08.14 | ghenry | dumb ;-) |
| 11:12.01 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:14.17 | *** join/#asterisk-dev ccesario (n=ccesario@189.88.2.103) |
| 11:18.22 | mvanbaak | there are no dumb questions, only dumb answers |
| 11:20.59 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:22.51 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:23.15 | *** join/#asterisk-dev oej (n=olle@fw01d.snowmen.se) |
| 11:23.15 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 11:31.22 | ghenry | Hi again. So the stuff I've just done in trunk do I need to put them in the 1.6.0/1.6.1/1.6.2 branches? |
| 11:31.47 | ghenry | Also, is there a doc explaining the differences between .0/.1/.2 1.6 branches? |
| 12:04.32 | *** join/#asterisk-dev qdk (n=qdk@195.242.194.42) |
| 12:13.35 | *** join/#asterisk-dev leifmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:13.35 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 12:23.25 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:34.37 | *** join/#asterisk-dev msetim (n=setim@189.32.44.21) |
| 12:34.47 | msetim | hi guys |
| 12:35.17 | msetim | I'm facing a very random problem and I don't know what can I do to open a good ticket to we work on them. |
| 12:36.19 | msetim | Randomly the queue stop to routing calls to agents. It's very strange because show agents show the agent idle and when I run show queues it's show Busy to same agent... |
| 12:37.04 | *** join/#asterisk-dev bbryant (i=d1c064c1@gateway/web/ajax/mibbit.com/x-16f426c2bf06a0ce) |
| 12:37.04 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 12:38.09 | msetim | The show queues take a long time to set the really agent state... (not in use) |
| 12:41.17 | leifmadsen | msetim: what version of asterisk? Sounds like you need to use state_interface |
| 12:42.48 | msetim | I'm using 1.4.20.1... i found this problem (very similar) but the syntom is a bit different https://issues.asterisk.org/view.php?id=14112 |
| 12:43.20 | leifmadsen | msetim: later versions of 1.4 have state_interface that gets around this issue |
| 12:43.41 | msetim | wow! another information important, when I manually exec reload app_queue the calls are routed correctly |
| 12:44.24 | msetim | right I will do tests against latest version |
| 12:47.02 | *** join/#asterisk-dev eliel (n=eliels@200.61.172.61) |
| 12:47.02 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 12:48.21 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:48.21 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:50.28 | *** part/#asterisk-dev bbryant (i=d1c064c1@gateway/web/ajax/mibbit.com/x-16f426c2bf06a0ce) |
| 13:00.32 | eliel | ghery? |
| 13:02.10 | kpfleming | ghenry |
| 13:02.20 | ghenry | hi |
| 13:03.33 | leifmadsen | howdy! |
| 13:03.43 | eliel | ghenry: hello, please look at the commit message guidelines |
| 13:04.27 | ghenry | I didn't find any. I was pointed to the getting started guide and given various pointers. |
| 13:04.29 | eliel | when closing an issue, you have can generate the commit message template from the issue in the bug tracker |
| 13:04.33 | ghenry | Is there a link |
| 13:04.47 | ghenry | I did do that on the two issues |
| 13:04.50 | eliel | ghenry: i think so, i don't remember where it is :) |
| 13:05.06 | ghenry | :) |
| 13:05.51 | *** join/#asterisk-dev Victor_Yure (n=victor@unaffiliated/victoryure/x-837844) |
| 13:07.12 | eliel | and when you merge from trunk to 1.6.x use mergetrunk6 |
| 13:09.02 | ghenry | ok. Is there a guide for this too? I only saw private branch merging |
| 13:10.59 | leifmadsen | eliel: how much more AstXML documentation needs to be converted? |
| 13:11.15 | leifmadsen | eliel: I think we should open a bug to track the remaining docs that need to be done |
| 13:11.16 | eliel | ghenry: http://www.asterisk.org/developers/svn-branching-merging |
| 13:11.23 | leifmadsen | eliel: I plan on helping there later today |
| 13:11.29 | kpfleming | eliel: ghenry: this would be a good time to ensure that *all* the stuff the developers need to know is documented somewhere |
| 13:11.41 | kpfleming | so anything you come across that is not documented, or you can't find, keep track of it, so we can fix that up |
| 13:11.44 | ghenry | eliel: thanks |
| 13:12.23 | leifmadsen | I noticed some of the links on the asterisk.org/developers page yesterday need to be updated to reflect the new names too |
| 13:12.23 | eliel | leifmadsen: all the manager actions, apps a func from dialplan are already converted, only left an agi command in app_stack.c |
| 13:12.39 | leifmadsen | eliel: really all funcs, apps, and manager commands are done? wow! |
| 13:12.52 | leifmadsen | I figured you still had some outstanding |
| 13:14.30 | eliel | leifmadsen: there is one application there that needs to be converted AGI() and the stuff inside app_rpt |
| 13:14.48 | leifmadsen | we can ignore app_rpt I think... |
| 13:14.58 | leifmadsen | unless someone thinks it needs to be updated |
| 13:15.11 | leifmadsen | brb, gotta go analyze a document |
| 13:15.17 | seanbright | heh |
| 13:15.21 | seanbright | is that a euphamism? |
| 13:15.40 | seanbright | i have to go... "analyze a document" |
| 13:15.44 | seanbright | runs off to work |
| 13:18.08 | leifmadsen | seanbright: lol |
| 13:18.18 | leifmadsen | seanbright: it was not a euphemism :) |
| 13:19.12 | jsmith | leifmadsen *is* a euphemism |
| 13:19.27 | leifmadsen | jsmith isn't wrong |
| 13:22.47 | kpfleming | a euphemism for what, would be the question |
| 13:23.21 | kpfleming | eliel: the app_stack agi command issue is almost fixed... my new_optional_api branch just needs some test reports from mac osx users and it's good to merge |
| 13:24.05 | leifmadsen | kpfleming: you don't even want to know |
| 13:25.28 | eliel | kpfleming: i will try to test it in the macbook |
| 13:26.39 | kpfleming | the test is to make sure that app_stack can be loaded both with and without res_agi loaded first, and do the right thing in both cases |
| 13:28.46 | mvanbaak | kpfleming: team/kpfleming/new_optional_api ? |
| 13:28.49 | leifmadsen | I have OSX 10.4 if a test needs to be done there, however I'm not positive I ever setup a compiling environment |
| 13:28.51 | kpfleming | yep |
| 13:29.07 | mvanbaak | kpfleming: I'll give it a spin on OpenBSD |
| 13:36.19 | *** join/#asterisk-dev putnopvut (n=putnopvu@asterisk/master-of-queues/mmichelson) |
| 13:36.19 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 13:37.28 | eliel | i will test that branch if i am able to checkout the branch... grrr, hate argentinian providers.. |
| 13:41.25 | *** join/#asterisk-dev sst1234 (i=ss@minos.sil.at) |
| 13:41.27 | sst1234 | Hello |
| 13:41.52 | sst1234 | i have a question pending on issue #15151 |
| 13:42.20 | sst1234 | as i have seen in the changelog it hasnt been implemented in version 1.4.25, right? |
| 13:43.13 | sst1234 | cause i have a problem with my pbx system using version 1.4.25 |
| 13:43.42 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 13:43.48 | sst1234 | sometimes asterisk freeze and doesnt handle sip packets anymore and with core show lock i have ssen the following: |
| 13:44.14 | sst1234 | === Thread ID: 1092675920 (do_monitor started at [16595] chan_sip.c restart_monitor()) |
| 13:44.18 | sst1234 | === ---> Lock #0 (chan_sip.c): MUTEX 16264 sipsock_read &netlock 0x7f6d0e9d0700 (1) |
| 13:44.22 | sst1234 | === ---> Lock #1 (chan_sip.c): MUTEX 4728 find_call &p->lock 0x374ffd0 (1) |
| 13:44.24 | sst1234 | === ---> Lock #2 (sched.c): MUTEX 220 ast_sched_add_variable &con->lock 0x1b2eee0 (1) |
| 13:45.10 | sst1234 | could this depend on issue 15151? |
| 13:45.47 | eliel | sst1234: use pastebin |
| 13:46.08 | sst1234 | ok, how ;) |
| 13:47.07 | leifmadsen | www.pastebin.ca |
| 13:47.16 | sst1234 | ah ok |
| 13:47.19 | leifmadsen | M15151 |
| 13:47.21 | MuffinMan | [new] [Asterisk] Channels/chan_sip/Transfers 0015151: [patch] Deadlock On One-legged Transfer [SIP / REPLACES] (Call Pickup) reported by irroot (Karma: +19.00) https://issues.asterisk.org/view.php?id=15151 |
| 13:48.18 | leifmadsen | sst1234: I'd test the patch provided if you think they are related to confirm if it solves your problem |
| 13:48.50 | kpfleming | anyone know a command i can use from the configure script to find out which version (major/minor) of OSX the build is being made on? |
| 13:49.55 | sst1234 | http://www.pastebin.ca/1437776 here are two locks which appears during and at end of the freeze time which is around 10 to 30 sec |
| 13:52.00 | eliel | kpfleming: (uname -r |cut -d"." -f 1) (uname -r |cut -d"." -f 2) (uname -r |cut -d"." -f 3) ? |
| 13:52.06 | mvanbaak | kpfleming: use uname |
| 13:52.16 | mvanbaak | lol, eliel was 1 second faster |
| 13:52.46 | mvanbaak | eliel: you can make that: (uname -r | cut -d "." -f 1,2,3 |
| 13:55.15 | kpfleming | what is the sample output of that command? |
| 13:55.54 | eliel | 9.7.0 is the output of uname -r |
| 13:55.56 | tzanger | kpfleming: that's teh first 3 digits of the kernel version |
| 13:56.12 | tzanger | er no linux and 2.6 or 2.4 etc I think |
| 13:56.20 | tzanger | nope my first guess was right :-) |
| 13:56.37 | tzafrir_laptop | $ uname -r | cut -d. -f3 |
| 13:56.37 | tzafrir_laptop | 26-2-amd64 |
| 13:56.46 | kpfleming | no, i want 10.4 for OSX 10.4.x, not the kernel version |
| 13:56.58 | kpfleming | this is for darwin, not linux :-) |
| 13:57.12 | tzanger | ohhh |
| 13:57.14 | tzanger | sorry |
| 13:57.58 | eliel | uname -r |cut -d"." -f 1,2 |
| 13:58.14 | eliel | the output of this command is 9.7 |
| 13:58.32 | kpfleming | but which osx version is on that system? |
| 13:58.46 | kpfleming | osx is always 10.x.x by definition :-) |
| 13:59.00 | eliel | ooohh, sorry, 10.5.7 |
| 13:59.04 | tzanger | so the next is osxi? |
| 13:59.05 | eliel | hmmm, let me try :) |
| 13:59.40 | kpfleming | tzanger: no, Next died a long time ago :-) |
| 13:59.54 | tzanger | kpfleming: smart ass |
| 14:00.55 | seanbright | or -f1-3 |
| 14:00.56 | sst1234 | leifmadsen: got any other ideas what these error could be about? |
| 14:01.04 | seanbright | is late to the uname/cut party |
| 14:01.06 | leifmadsen | sst1234: I'm not a developer, so I have zero idea |
| 14:01.18 | seanbright | kpfleming: delayed menuselect-related pong |
| 14:01.25 | kpfleming | ha! |
| 14:01.36 | seanbright | had to work up the courage to pong |
| 14:01.54 | kpfleming | so... which would you prefer: <depend>optional_api | res_smdi</depend> or <depend><or>optional_api</or><or>res_smdi</or></depend> |
| 14:02.26 | sst1234 | well anybody else? |
| 14:02.47 | sst1234 | otherwise we have to downgrade to asterisk 1.2 again... |
| 14:02.48 | seanbright | kpfleming: ugh... is there a 3rd option? |
| 14:02.50 | kpfleming | if you offer one |
| 14:02.54 | kpfleming | i don't have a third option, no |
| 14:02.58 | seanbright | will think on it for a few |
| 14:03.04 | eliel | kpfleming: grep -A1 ProductVersion /System/Library/CoreServices/SystemVersion.plist |grep string |
| 14:03.16 | kpfleming | wow |
| 14:03.49 | eliel | yes, owoowowo |
| 14:04.02 | eliel | and the output is: <string>10.5.7</string> |
| 14:04.11 | kpfleming | ok |
| 14:04.20 | kpfleming | so that can be parsed from the configure script then |
| 14:04.22 | kpfleming | fun |
| 14:04.23 | sst1234 | is the address in the error message (0x7f6d0e9d0700) related to the netlock function in any way? |
| 14:04.31 | sst1234 | is it the argument to the function? |
| 14:04.32 | kpfleming | addresses are random |
| 14:04.43 | kpfleming | they are different on every system and from run to run |
| 14:04.57 | kpfleming | and netlock is not a function |
| 14:05.02 | sst1234 | so how do i know whats in it? |
| 14:05.23 | kpfleming | you would be best to open an issue on the issue tracker and follow the reporting guidelines for providing backtraces and/or 'core show locks' output |
| 14:05.54 | eliel | kpfleming: done.... $(sw_vers -productVersion |cut -d"." -f1,2) |
| 14:06.02 | eliel | output is 10.5 |
| 14:06.12 | kpfleming | fabulous! thanks |
| 14:06.13 | tzafrir_laptop | Qwell, btw: it sees that there's a script in dahdi-tool (xpp/waitfor_xpds) that relies on 'which' which is a separate package |
| 14:06.18 | sst1234 | is there a way to generate a backtrace when asterisk dont dies? |
| 14:06.18 | seanbright | kpfleming: short of restructuring how we do all the xml stuff for menuselect... i guess the | is good enough. |
| 14:06.38 | eliel | ../usr/bin/sw_vers |
| 14:06.49 | kpfleming | sst1234: yes, there is a script for it, but 'core show locks' is more useful in that case |
| 14:06.50 | tzafrir_laptop | I didn't realise until now a system could live happily without it |
| 14:06.59 | sst1234 | ok thanks |
| 14:07.52 | tzafrir_laptop | and see also the silly error reporting fix I just commited :-( |
| 14:08.09 | kpfleming | seanbright: well, i could implement the <or> method fairly easily |
| 14:08.12 | sst1234 | then i will have a look at the issue tracker |
| 14:08.18 | sst1234 | thank you |
| 14:08.23 | kpfleming | good luck! |
| 14:08.36 | kpfleming | i'll figure out some method of doing this then |
| 14:08.43 | seanbright | kpfleming: well if we were going to introduce new tags i would want it to be something like... |
| 14:08.51 | kpfleming | without it, the whole optional_api thing doesn't provide much benefit |
| 14:09.04 | seanbright | <depend><or><depend>x</depend><depend>y</depend></or></depend> |
| 14:09.17 | kpfleming | really? |
| 14:09.20 | putnopvut | recursive xml |
| 14:09.20 | seanbright | yeah |
| 14:09.24 | kpfleming | oh, i see |
| 14:09.37 | seanbright | that's more flexible |
| 14:09.45 | kpfleming | so there is an implicit <and> around all the current <depend> tags then |
| 14:09.47 | seanbright | if you wanted to get really crazy with it, anyway. |
| 14:09.51 | seanbright | yeah |
| 14:09.57 | seanbright | one could argue that, anyway :) |
| 14:10.01 | tzafrir_laptop | but is it really needed? |
| 14:10.04 | kpfleming | that would be definitely be a lot more code |
| 14:10.13 | seanbright | tzafrir_laptop: no, we're just talking about it because we're bored. |
| 14:10.27 | kpfleming | tzafrir_laptop: yes, because i don't want people to be forced to build res_smdi if they have no intention of using it |
| 14:10.31 | seanbright | kpfleming: agreed, which is why i said | was good enough :) |
| 14:11.36 | kpfleming | but if their platform can't support the optional_api stuff, then they *have* to build res_smdi (and load it) for app_voicemail to be usable |
| 14:11.44 | kpfleming | so it would be best for menuselect to be able to handle this |
| 14:12.02 | seanbright | well, fwiw, i vote for pipe |
| 14:12.08 | kpfleming | me too |
| 14:12.12 | kpfleming | simple to implement and use |
| 14:12.15 | seanbright | indeed. |
| 14:12.35 | seanbright | now... what if a dependency has a | in it's name? |
| 14:12.37 | seanbright | heh |
| 14:12.40 | kpfleming | not allowed |
| 14:12.47 | seanbright | how convenient. |
| 14:13.19 | kpfleming | i think it's reasonable to assume the dependency name character set is [a-zA-Z0-9_-] |
| 14:13.47 | kpfleming | actually, it's not even case sensitive |
| 14:13.52 | tzafrir_laptop | kpfleming, also: . |
| 14:13.58 | seanbright | tzafrir_laptop: what was that bashism checker script you've mentioned before? |
| 14:14.00 | kpfleming | so [A-Z0-9_-.] |
| 14:14.10 | seanbright | \. |
| 14:14.12 | seanbright | :) |
| 14:14.17 | kpfleming | lol |
| 14:14.18 | tzafrir_laptop | checkbashism |
| 14:14.26 | seanbright | danke. |
| 14:14.26 | kpfleming | what a convenient name |
| 14:14.31 | tzafrir_laptop | part of devscripts in debian/ubuntu |
| 14:15.03 | seanbright | thank you, sir. |
| 14:17.29 | *** join/#asterisk-dev mog (n=mog@c-68-62-174-19.hsd1.al.comcast.net) |
| 14:17.29 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 14:21.01 | putnopvut | Heh, I just noticed that in https://reviewboard.asterisk.org/r/252 I have a return -0; in there |
| 14:21.20 | seanbright | return 3i; |
| 14:21.22 | kpfleming | holy cripes... the xcode download is 1GB |
| 14:21.27 | seanbright | yeah |
| 14:21.33 | seanbright | i think you get 2 full gcc versions with it |
| 14:21.40 | seanbright | at least i did when i downloaded an old old version |
| 14:21.52 | kpfleming | good thing i have lots of room |
| 14:22.11 | seanbright | putnopvut: you should change it to: |
| 14:22.28 | putnopvut | is in suspense |
| 14:22.40 | seanbright | return (1 * 7 / 5 + 3.14159 * 88.2) * 0; |
| 14:23.02 | putnopvut | I wonder if the optimizer would optimize that to just be 0. |
| 14:23.08 | seanbright | one would hope |
| 14:23.28 | seanbright | my favorite is still: return !!some_func_call(); |
| 14:23.42 | putnopvut | I'm a fan of that too. |
| 14:24.12 | seanbright | good times. |
| 14:26.52 | seanbright | <PROTECTED> |
| 14:26.58 | seanbright | ^ notice the space |
| 14:27.00 | seanbright | heh |
| 14:27.26 | kpfleming | yes, with optimization on expressions that can be evaluated at compile time are |
| 14:28.35 | *** join/#asterisk-dev DarkRift (n=dark@bas1-sthubert21-1096591461.dsl.bell.ca) |
| 14:33.56 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 14:33.56 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 14:33.58 | kaii | seanbright: what does "return !!func()" return? given my func() returns true? |
| 14:34.06 | kaii | never tried that |
| 14:34.18 | russellb | kaii: 1 |
| 14:34.22 | seanbright | right |
| 14:34.37 | seanbright | the !! guarantees you either get back a 0 or a 1 |
| 14:34.40 | kaii | expectable.. :) |
| 14:34.47 | seanbright | even if func returns "hello" or 9 or -2 |
| 14:34.48 | kaii | seanbright: ahhh i see |
| 14:34.57 | kaii | makes sense |
| 14:39.04 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 14:45.03 | *** join/#asterisk-dev Deeewayne (n=dwayne@75.76.254.162) |
| 14:45.03 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 14:56.51 | philippel | putnopvut curious if you have had a chance to glance at: https://issues.asterisk.org/view.php?id=15168 yet, I want to blast out to the dev mailing list to get feedback but though it would be best for you to have a look at it first? |
| 14:57.24 | putnopvut | philippel: I have absolutely been swamped here and have not had a chance to look at it. |
| 15:05.16 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:05.16 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:11.04 | *** join/#asterisk-dev CunningPike (n=CunningP@204.239.10.119) |
| 15:11.54 | russellb | ghenry: ping |
| 15:12.00 | ghenry | hi |
| 15:12.26 | russellb | actually, nevermind |
| 15:12.38 | russellb | I was going to say something about your commit message, but it looks like you found the template :-) |
| 15:12.39 | russellb | so nevermind! |
| 15:13.01 | ghenry | Yeah, I've been shouted at already. Feel free. I've no nerves left! ;-) |
| 15:13.06 | ghenry | Joking. |
| 15:13.39 | jpeeler | also the merge scripts don't appear to have been used |
| 15:16.59 | russellb | ghenry: It's okay, it's expected to have some hiccups at first |
| 15:17.09 | russellb | I very much appreciate your help |
| 15:17.17 | ghenry | jpeeler: Yeah, I didn't know about other than cp |
| 15:17.19 | ghenry | sorry |
| 15:17.24 | ghenry | I know now |
| 15:17.31 | jpeeler | heh, as russell said, it's fine |
| 15:19.18 | *** part/#asterisk-dev jsmith (n=jsmith@asterisk/training-and-documentation-guru/jsmith) |
| 15:19.28 | *** join/#asterisk-dev jtodd (i=zidb0j3e@ns.fox-den.com) |
| 15:19.28 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 15:22.00 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 15:22.58 | *** join/#asterisk-dev jsmith (n=jsmith@asterisk/training-and-documentation-guru/jsmith) |
| 15:22.58 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 15:23.11 | jsmith | The_Boy_Wonder: You're my new hero! |
| 15:23.51 | jsmith | The_Boy_Wonder: angler and I were just discussing IAX2 trunking and the timing requirements, and then I saw http://reviewboard.digium.com/r/262/ |
| 15:24.51 | The_Boy_Wonder | HOORAY! |
| 15:25.05 | angler | :) |
| 15:25.12 | jsmith | The_Boy_Wonder: I've got (trivial) patch along those lines coming your way |
| 15:25.21 | The_Boy_Wonder | cool |
| 15:25.37 | jsmith | Well, maybe it'll be quicker for you just to write than me |
| 15:25.43 | angler | jsmith, you doing a patch for the Trunking field? |
| 15:25.59 | The_Boy_Wonder | ok |
| 15:26.09 | jsmith | angler and I decided it would be nice if the output of "iax2 show peer blah" or "iax2 show user blah" would show whether or not trunking was enabled for that peer |
| 15:26.27 | jsmith | (and whether or not it would actually be used, because of the unavailability of a timing device) |
| 15:26.48 | jsmith | Something along the lines of ast_cli(a->fd, " Trunk : %s\n", ast_test_flag(peer, IAX_TRUNK) ? "Yes" : "No"); |
| 15:27.15 | jsmith | That shows whether or not trunking is enabled, but not whether trunking will actually get used |
| 15:27.25 | jsmith | The_Boy_Wonder: Make sense? |
| 15:27.52 | leifmadsen | ok, I'm about to create a new 1.4.26-rc1 -- does anyone have anything pending for 1.4 branch for the next 10 mins? |
| 15:28.06 | leifmadsen | is asking because he is seeing a flurry of activity in the -commits |
| 15:28.08 | The_Boy_Wonder | jsmith: yeah, that makes sense. At the moment there's no way of knowing whether or not a peer was set up for trunking if the timing interface fails |
| 15:28.41 | jsmith | The_Boy_Wonder: Oh? So the IAX_TRUNK flag doesn't get set? |
| 15:28.51 | The_Boy_Wonder | jsmith: basically if the timer fails, the trunk flag will be cleared |
| 15:29.01 | The_Boy_Wonder | jsmith: as it is right now |
| 15:29.17 | jsmith | The_Boy_Wonder: But there's no way to easily tell if the trunk was not set vs. got cleared because of lack of timing, right? |
| 15:29.26 | jsmith | It would be nice if we could differentiate that in the output |
| 15:29.58 | angler | jsmith, ya thats what i ment.... be in the output of the user and peer |
| 15:30.01 | The_Boy_Wonder | jsmith: i agree, the only way to tell the trunk was not set at the moment is a warning message that's output during build_peer |
| 15:30.30 | The_Boy_Wonder | and that warning usually gets lost in the noise of starting asterisk |
| 15:30.52 | jsmith | I 100% agree... how hard would it be to use a second flag called IAX_TRUNK_NOTIMING |
| 15:31.06 | jsmith | So we could differentiate between the two causes |
| 15:31.59 | The_Boy_Wonder | not hard, its just there's only one flag left in the iax2_flags enum |
| 15:32.02 | The_Boy_Wonder | :/ |
| 15:32.20 | jsmith | :-/ |
| 15:32.29 | jsmith | Only a single bit left? |
| 15:32.32 | jsmith | sighs |
| 15:32.34 | The_Boy_Wonder | right |
| 15:32.38 | The_Boy_Wonder | i mean i could use it |
| 15:32.57 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 15:33.03 | The_Boy_Wonder | but, i should probably put a system in place to allow for more in the future if i do |
| 15:33.12 | jsmith | Well, it certainly makes it less simple |
| 15:33.18 | jsmith | But that's probably the ideal approach |
| 15:34.03 | The_Boy_Wonder | cool, well i'll look into it as time permits. this has bugged me in the past as well |
| 15:34.34 | jsmith | The_Boy_Wonder: In the meantime, mind adding the one-liner I pasted above into the output of "iax2 show user blah" and "iax2 show peer blah"? |
| 15:37.43 | jsmith | The_Boy_Wonder: (or would you prefer I do that as a separate patch on the bug tracker?) |
| 15:38.58 | The_Boy_Wonder | jsmith: i'll take care of it, no big deal |
| 15:46.31 | *** join/#asterisk-dev marv[work] (n=timr@router.asteriasgi.com) |
| 16:09.02 | The_Boy_Wonder | jsmith: committed that one line fix |
| 16:13.26 | jsmith | The_Boy_Wonder: Couple of quick questions for you, if you have a second |
| 16:13.44 | The_Boy_Wonder | jsmith: yep |
| 16:15.31 | *** join/#asterisk-dev MikeJ (n=MikeJ@freeswitch/developer/mikej) |
| 16:17.05 | kpfleming | Corydon76-dig: did your link fail for symbols related to __ast_monitor? |
| 16:18.42 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 16:18.42 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 16:18.58 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 16:23.08 | *** join/#asterisk-dev shido6 (n=shido6@96-28-35-120.dhcp.insightbb.com) |
| 16:32.28 | *** join/#asterisk-dev Mw3 (i=mw3@88.151.97.220) |
| 16:41.54 | ajohnson | Corydon76-dig: Would you like access to the machine this problem is happening on? |
| 17:08.24 | *** join/#asterisk-dev moy (n=moy@74.12.123.90) |
| 17:15.01 | Corydon76-dig | ajohnson: Just a backtrace would be fine |
| 17:15.15 | Corydon76-dig | ajohnson: or another valgrind |
| 17:15.19 | ajohnson | bt full on the core file? |
| 17:15.25 | Corydon76-dig | Right |
| 17:15.26 | *** join/#asterisk-dev caio1982_ (i=caio1982@CAcert-br/caio1982) |
| 17:15.39 | ajohnson | ok, heading to meeting, will do in about 15 minutes |
| 17:15.53 | Corydon76-dig | I definitely fixed A bug, but it may not be THE bug |
| 17:17.07 | *** join/#asterisk-dev Victor_Yure (n=victor@unaffiliated/victoryure/x-837844) |
| 17:33.32 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 17:36.43 | angler | The_Boy_Wonder, you rock! |
| 17:45.44 | seanbright | kpfleming: i require your counsel |
| 17:46.28 | russellb | I believe he is lunching. |
| 17:46.38 | russellb | I can serve as a kpfleming proxy, what's up |
| 17:46.49 | seanbright | meh... i already asked you and you weren't helpful |
| 17:46.49 | seanbright | :) |
| 17:46.51 | seanbright | http://pastie.org/private/pcyos2kyy8ljkkpmo8bq |
| 17:46.53 | russellb | harsh. |
| 17:46.59 | seanbright | i want to know if i can shorten that with macro magic |
| 17:47.44 | *** join/#asterisk-dev qdk (n=qdk@81.7.168.130) |
| 17:50.59 | seanbright | russellb: it's hopeless, eh? |
| 17:51.36 | russellb | I don't know. I just don't know of a better way. |
| 17:52.05 | seanbright | preprocessor macros can call other ones, yeah? |
| 17:52.40 | russellb | yes. |
| 17:52.55 | seanbright | but i probably can't "build" a define |
| 17:52.59 | seanbright | like HAVE_##whatever |
| 17:53.24 | seanbright | damn. |
| 17:53.35 | russellb | Tricky. I know you can dynamically build symbols ... but I don't know if you can do that within stuff in the preprocessor |
| 17:54.05 | seanbright | heh... i could be REALLY nasty and have configure pump out something like: |
| 17:54.26 | seanbright | #define HAVE_SYSLOG_FACILITY_LOG_AUTH { "auth", LOG_AUTH }, |
| 17:54.30 | seanbright | heh |
| 17:54.36 | russellb | bleh |
| 17:54.40 | seanbright | yeah |
| 17:54.45 | seanbright | silly programming. |
| 17:59.25 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 17:59.25 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 18:04.45 | seanbright | #pragma GCC poison |
| 18:04.55 | seanbright | awesome. |
| 18:06.24 | Juggie | eugh, pragma reminds me of borland builder :) |
| 18:15.52 | *** join/#asterisk-dev willianmazzardo (n=willianm@201.66.134.150) |
| 18:31.41 | kpfleming | no, you cannot define macros from within macros |
| 18:31.47 | kpfleming | i would have used that yesterday if i could have |
| 18:32.21 | kpfleming | that is one reason why the gcc team spent so much time making 'static inline' functions work as fast as macros, so that macros could define them and not lose performance |
| 18:33.42 | russellb | hacks kpfleming's gibson and downloads his garbage file |
| 18:33.48 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:33.51 | russellb | now I haz your sekretz |
| 18:33.57 | ajohnson | Corydon76-dig: I have a question. Can I upgrade unixODBC independent of the -devel sources package? If not, how do I recompile Asterisk with the unixODBC sources in a different directory? |
| 18:35.38 | eliel | talking about macro magic, i would like to simplify this without using two macros |
| 18:36.45 | *** join/#asterisk-dev VoipForces (n=kvirc@office.prival.ca) |
| 18:36.58 | Corydon76-dig | ajohnson: I don't think installing the source gives you an option about installing headers |
| 18:37.07 | eliel | http://www.pastebin.ca/1438237 |
| 18:37.10 | Corydon76-dig | ajohnson: you have dependency hell? |
| 18:37.12 | VoipForces | Hi, anyone ever worked on a voicemailmain patch to allow cross context voicemail forwards ? |
| 18:37.29 | ajohnson | trying to avoid dependency hell |
| 18:37.36 | ajohnson | I'll try it out right now |
| 18:37.51 | Corydon76-dig | but duplicate includes shouldn't hurt anything, as long as the libraries aren't duplicated |
| 18:37.51 | ajohnson | Is it concerning that this happens in 1.6.2-beta2 and not in 1.6.1? |
| 18:38.28 | Corydon76-dig | We're doing some things differently in 1.6.2, so the source might not match |
| 18:41.12 | Juggie | Corydon76-dig, i noticed something the other day just wondering if there is any technical reason why when VoiceMailMain() is called, if you passed no mailbox it would not check to see if mailbox was set and use that value (the one you set for the mwi support in various channels) |
| 18:41.46 | file | the mailbox is not set at the channel level. |
| 18:41.47 | Juggie | was going to do a patch for it, but it seems so obvious i thought there might be a reason it hasnt been done |
| 18:42.25 | Juggie | oh, its only attached to the peer/dahdi extension etc. |
| 18:42.34 | Corydon76-dig | Juggie: sure... suppose you wanted to check your voicemail and you weren't at your desk, but at somebody else's desk |
| 18:42.52 | file | so the way to do that would be to use setvar and have a dialplan variable which contains the information, and then have the dialplan use it |
| 18:42.58 | kpfleming | right |
| 18:43.01 | Corydon76-dig | should Asterisk stop you, because the logic thinks it's smarter than the PBX admin? |
| 18:43.08 | kpfleming | there's no need for implicit behavior when the dialplan can do it |
| 18:43.44 | Juggie | right, knew i could do that, i just thought since it was already in there but you are right it makes sense that its attached to the phone/peer and not the channel. |
| 18:43.46 | Corydon76-dig | Agreed, especially when the implicit behavior might be undesireable in certain circumstances |
| 18:44.04 | Juggie | Corydon76-dig, well, it would of had to have been a new option |
| 18:44.13 | Juggie | at this point since no parms asks for a vmbox |
| 18:44.18 | *** part/#asterisk-dev VoipForces (n=kvirc@office.prival.ca) |
| 18:44.33 | Corydon76-dig | Juggie: but as soon as you're talking about a new option, you might as well just specify the mailbox directly |
| 18:44.37 | kpfleming | rigth |
| 18:45.05 | Juggie | well, this can get confusing in your dialplan when you have multiple contexts of vmboxes etc |
| 18:45.10 | Juggie | since you have to set mailbox=box@context |
| 18:45.16 | Juggie | i just thought that might be worth reusing |
| 18:46.50 | Juggie | i knew i could do a setvar in the peer/phone/etc config, but that seemed like i was duplicating the same information. |
| 18:49.00 | kpfleming | $(SIPPEER(mailbox)) |
| 18:49.09 | kpfleming | well, it needs the peer name |
| 18:49.30 | Juggie | what about dahdi channels |
| 18:49.55 | kpfleming | if you are going to do any kind of patch, then it would be best to add functionality to the channel drivers to expose the information that is already present |
| 18:50.06 | kpfleming | that would have been required anyway for your proposed patch |
| 18:50.15 | Juggie | would there be a case for exposing mailbox in channel()? |
| 18:50.26 | kpfleming | sure, that could be done |
| 18:50.33 | kpfleming | $(CHANNEL(subscribed_mailbox)) or something |
| 18:50.39 | Juggie | right. |
| 18:50.55 | kpfleming | although keep in mind that channel drivers can support multiple mailboxes being subscribed at once |
| 18:51.12 | Juggie | true... |
| 18:51.14 | Juggie | i didnt think of that. |
| 18:55.29 | ajohnson | I have achieved dependency hell |
| 18:56.28 | Juggie | i would say just call it subscribed_mailboxes and just spit it back in the same format as set in mailbox= so, 'mailbox,mailbox' |
| 18:56.41 | Juggie | leave it up to the admin to sort it out if they use that feature |
| 18:57.15 | kpfleming | although that's not really a channel property, we may have to come up with a better name |
| 18:58.00 | *** join/#asterisk-dev |Cybex| (n=John@80.100.126.176) |
| 19:03.16 | eliel | I have a -users/-dev question regarding the ackcall option in chan_agent |
| 19:03.30 | eliel | it is there, but i think it was done for agentcallbacklog and not for agentlogin |
| 19:06.22 | jsmith | thinks it should be there for agentlogin agents as well |
| 19:08.49 | eliel | i tried using it for agentlogin, but nothing happens.... i will review the code but i think it is not working as expected |
| 19:25.23 | *** join/#asterisk-dev Victor_Yure (n=victor@unaffiliated/victoryure/x-837844) |
| 19:31.27 | seanbright | kpfleming: so can i somehow use static inline functions to clean up that monstrosity? or am i doomed? |
| 19:33.26 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 19:36.00 | seanbright | actually, that's an amazingly stupid question. |
| 19:39.06 | tzafrir_laptop | I connect a loop on a PRI port and get flooded with messages such as: |
| 19:39.13 | tzafrir_laptop | [May 28 19:34:58] WARNING[4474] chan_dahdi.c: PRI Error on span 0: We think we're the CPE, but they think they're the CPE too. |
| 19:39.24 | tzafrir_laptop | Should such messages be rate-limited? |
| 19:39.52 | *** join/#asterisk-dev ghenry (n=ghenry@pdpc/supporter/monthlybyte/ghenry) |
| 19:41.19 | *** part/#asterisk-dev willianmazzardo (n=willianm@201.66.134.150) |
| 19:44.43 | *** join/#asterisk-dev malcolmd (n=malcolmd@pdpc/sponsor/digium/malcolmd) |
| 19:44.44 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 19:47.48 | *** join/#asterisk-dev qdk (n=qdk@81.7.168.130) |
| 20:14.45 | *** join/#asterisk-dev jtodd (i=hhj1q7ua@ns.fox-den.com) |
| 20:14.45 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 20:23.42 | kpfleming | tzafrir_laptop: we don't have any rate limiting for log messages in asterisk today, but it certainly would be useful |
| 20:24.45 | seanbright | we do in some places |
| 20:24.51 | seanbright | but nothing generic |
| 20:25.11 | seanbright | if (atomic_fetch_add()++ % 10 == 0) |
| 20:25.14 | seanbright | that kinda stuff |
| 20:25.55 | russellb | there is a patch on mantis for preventing duplicate log messages |
| 20:26.00 | seanbright | nice |
| 20:26.08 | russellb | the whole "previous message repeated 394234 times" type of thing |
| 20:26.16 | seanbright | awesome |
| 20:26.19 | seanbright | strcmp ftw! |
| 20:26.22 | russellb | :-) |
| 20:26.35 | seanbright | oh, i was thinking. |
| 20:26.38 | russellb | If someone wanted to test and review it, that would be excellent! ;-) |
| 20:26.43 | seanbright | string payload type in event system |
| 20:26.52 | seanbright | mayhaps worth a version field? or is it too late for that? |
| 20:26.58 | seanbright | like... what if the hash function changes |
| 20:27.29 | russellb | It's not too late. |
| 20:27.49 | russellb | For 1.6.1, I intentionally marked res_ais as experimental for the very reason that I wasn't willing to commit to a wire format between versions (yet) |
| 20:28.03 | russellb | so all changes are fair game in trunk as far as I am concerned |
| 20:28.12 | seanbright | excellent. |
| 20:28.15 | russellb | The one big thing to consider with big event API changes is the newcdr branch |
| 20:28.21 | seanbright | ah. yes. |
| 20:28.24 | russellb | That is a _ton_ of code that I _REALLY_ do not want to let bitrot |
| 20:28.34 | russellb | We've got to get that in and not destroy it |
| 20:31.57 | seanbright | that is/was murf's? |
| 20:38.29 | russellb | He was one of the primary developers, yes. |
| 20:38.34 | russellb | The other was bmd from Switchvox. |
| 20:40.56 | seanbright | well in reagards to the event api, i think having a version number is worthwhile. |
| 20:41.04 | seanbright | s/reag/reg/ |
| 21:05.36 | *** join/#asterisk-dev thansen (n=thansen@7.247.sfcn.org) |
| 21:06.36 | thansen | hello all, I recently upgraded to 1.6.1.0 and I think we may have a bug in the MixMonitor app...can anyone look at this with me for a second? |
| 21:07.14 | seanbright | leaking a file handle |
| 21:07.18 | thansen | I don't think the file handle is being closed on the recording |
| 21:07.25 | seanbright | i feel like a fix went in for that recently |
| 21:07.31 | thansen | seanbright: yes! |
| 21:07.47 | seanbright | have you tried the latest 1.6.1 branch? |
| 21:08.00 | thansen | seanbright: can you be so kind as to point me to a patch/bug #? |
| 21:08.09 | thansen | seanbright: no, this is on a production machine :( |
| 21:08.10 | seanbright | well, you could just search on mantis |
| 21:08.24 | seanbright | and i can't get to digium's servers at the moment |
| 21:08.48 | thansen | seanbright: sorry, what's mantis (I don't do any dev with asterisk) |
| 21:09.06 | seanbright | https://issues.asterisk.org/ |
| 21:09.12 | seanbright | but it appears to be down atm |
| 21:10.17 | file | ummm |
| 21:10.19 | file | it's larger then that |
| 21:10.26 | seanbright | what happened? |
| 21:10.33 | file | I can't get to the office ^_^ |
| 21:10.40 | seanbright | uh oh |
| 21:10.47 | seanbright | the russians!! |
| 21:10.56 | seanbright | THE LYBIANS!! |
| 21:10.57 | seanbright | heh |
| 21:11.01 | seanbright | libians. |
| 21:11.03 | seanbright | whatever. |
| 21:11.08 | file | aha there we are |
| 21:11.20 | file | things/people should be reappearing |
| 21:11.28 | seanbright | splendid |
| 21:11.44 | *** join/#asterisk-dev putnopvut (n=putnopvu@asterisk/master-of-queues/mmichelson) |
| 21:11.44 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 21:12.39 | seanbright | thansen: g729? |
| 21:13.24 | thansen | seanbright: no, this wasn't an issue for me in 1.6.0.9 either :( |
| 21:13.35 | thansen | I'm saving them as wav |
| 21:14.29 | thansen | then in the 'h' extension after everything has hung up I launch an php agi script |
| 21:14.57 | thansen | which subsqeuently manually calls system command of sox 'file.wav' 'file.mp3' |
| 21:15.29 | thansen | that script is running fine, but my mp3 file is now always 288 bytes |
| 21:16.22 | thansen | the same size as if I manually run the sox command from the terminal before hanging up the channel |
| 21:17.28 | thansen | this is on a MixMonitor file, Record() is still working fine |
| 21:17.49 | seanbright | hmm |
| 21:17.56 | seanbright | i'm not seeing anything in mantis |
| 21:21.17 | seanbright | thansen: might want to open a report on mantis (https://issues.asterisk.org/) |
| 21:24.49 | kpfleming | Corydon76-dig: so this weak_import thing does not appear to work properly at all |
| 21:25.16 | kpfleming | i can set MACOSX_DEPLOYMENT_TARGET to something higher than 10.2 to get the compiler to accept it, but the linker tries to resolves the weak symbols at link time, not run time |
| 21:26.41 | Corydon76-dig | kpfleming: hrm |
| 21:26.53 | kpfleming | let me pastebin my example |
| 21:27.19 | kpfleming | http://pastebin.ca/1438537 |
| 21:27.29 | kpfleming | compiling that with 'gcc -o test test.c' produces a linker failure |
| 21:28.01 | kpfleming | if compile to a .o file, and use 'nm -mg' to look at the symbols, it is marked as a weak external |
| 21:28.32 | thansen | seanbright: this looks like the only difference that I can see |
| 21:28.33 | thansen | - if (!(datastore = ast_channel_datastore_alloc(&mixmonitor_ds_info, NULL))) { |
| 21:28.33 | thansen | + if (!(datastore = ast_datastore_alloc(&mixmonitor_ds_info, NULL))) { |
| 21:28.33 | thansen | <PROTECTED> |
| 21:28.33 | thansen | <PROTECTED> |
| 21:28.33 | thansen | <PROTECTED> |
| 21:28.44 | kpfleming | that was just a function renaming |
| 21:28.48 | kpfleming | no effective difference |
| 21:29.02 | thansen | so I guess it's something lower level |
| 21:31.23 | thansen | well, and the file permissions changed from 644 to 666 but I don't see how that can affect anything |
| 21:31.39 | thansen | where is ast_free defined? |
| 21:33.49 | kpfleming | utils.h, i think |
| 21:33.53 | kpfleming | it's just a wrapper for free() |
| 21:40.00 | kpfleming | Corydon76-dig: almost there,i think i've found the solution, some linker flags |
| 21:40.23 | Corydon76-dig | Yay |
| 21:40.47 | *** join/#asterisk-dev wedhorn (n=nnnnnnnn@123-243-209-34.static.tpgi.com.au) |
| 21:40.48 | kpfleming | but i need to understand the impact of using '-flat_namespace' when linking asterisk objects, because that is required for using weak symbols |
| 21:43.04 | kpfleming | well... we are already using similar flags to link asterisk in the makefile |
| 21:43.05 | kpfleming | hmm |
| 21:48.49 | thansen | what is ast_frame_free supposed to do? |
| 21:49.49 | thansen | or even better, in what line in app_mixmonitor.c should the file handle be getting released? :) |
| 21:49.51 | kpfleming | all this stuff is documented int the header files |
| 21:50.05 | kpfleming | but the name is pretty self explanatory |
| 21:50.10 | thansen | kpfleming: I'm not really a dev, just trying to fix an urgent issue :( |
| 21:50.23 | kpfleming | well, not doing your homework does not encourage other people to help you |
| 21:50.35 | thansen | I've never looked at the asterisk source code until now |
| 21:51.39 | thansen | I guess 'homework' is relative, at least I'm trying to solve the issue and provide a patch rather than simply opening a bug :) |
| 21:51.52 | seanbright | well opening a bug will get more eyes on it is all |
| 21:51.53 | kpfleming | right |
| 21:52.10 | leifmadsen | ideally all issues should have a bug associated with them |
| 21:52.14 | kpfleming | but looking through the header files for descriptions of functions is pretty easy |
| 21:52.25 | leifmadsen | and if you're going to provide a patch, you'll need to open one at some point (just FYI) |
| 21:52.27 | russellb | or doxygen! :-) |
| 21:52.37 | kpfleming | yep, even better |
| 21:54.35 | kpfleming | seanbright: you have made many more than your quota of commits this week |
| 21:55.37 | russellb | I keep telling him to stop so he doesn't make me look bad |
| 21:56.23 | russellb | thansen: Even though we like to give people a hard time to RTFM, we really do appreciate your help. :-) |
| 21:56.50 | thansen | nods |
| 21:57.11 | russellb | Another thing you can do is try to post the information you have to the asterisk-dev list |
| 21:57.17 | russellb | Then, people can answer asynchronously |
| 21:57.21 | thansen | I don't mind RTFM, I just am really trying to get this solved on a running server |
| 21:57.22 | russellb | like, right now, I don't have time to dig into an issue |
| 21:57.43 | russellb | but I could probably review it and answer when I'm going through the -dev list later on |
| 22:09.57 | ajohnson | Corydon76-dig: Keep in mind that everyone who uses ODBC is going to have a problem in 1.6.2 with the way that it currently is (even if it is a unixODBC issue). |
| 22:10.56 | ajohnson | Well... everyone who uses ODBC and MySQL |
| 22:10.58 | Corydon76-dig | ajohnson: I'm not saying that it isn't ultimately a problem in Asterisk, just that Nick is a good person to involve at this point |
| 22:11.33 | Corydon76-dig | He's been very nice to help us in the past |
| 22:11.34 | ajohnson | Ok, I'm going to get in touch with him. |
| 22:11.59 | ajohnson | I'm wondering if my ODBC voicemail problem could also be a unixODBC issue as well |
| 22:12.17 | Corydon76-dig | which issue number? |
| 22:12.42 | ajohnson | 15186 |
| 22:13.07 | ajohnson | I accidentally put the product version as 1.6.1.0 when it really impacts all versions of 1.6 |
| 22:13.17 | Corydon76-dig | ajohnson: what's weird about this first issue is that Asterisk is in the middle of a loop retrieving multiple fields out of a table, when it suddenly chokes 3/4ths of the way in |
| 22:13.36 | Corydon76-dig | If it was an Asterisk problem, I would have expected it to die on the first or second field |
| 22:13.53 | Corydon76-dig | or last field |
| 22:14.00 | ajohnson | Can you tell me which field number? |
| 22:14.20 | ajohnson | Just wondering if it is the LONGBLOB field |
| 22:14.20 | Corydon76-dig | 34 |
| 22:14.53 | Corydon76-dig | 34 of 44 |
| 22:15.03 | ajohnson | wait, the 34th column of the table? or of a field in Asterisk? |
| 22:15.09 | Corydon76-dig | of the table |
| 22:15.12 | ajohnson | because that table only has 14 columns |
| 22:15.45 | Corydon76-dig | SELECT * FROM sip_peers WHERE name = ? AND host = ? |
| 22:15.51 | Corydon76-dig | the sip_peers table? |
| 22:15.56 | seanbright | kpfleming: will stop |
| 22:16.13 | ajohnson | is that where the problem is? not in storing the message, but querying the sip peers table? |
| 22:16.29 | Corydon76-dig | ajohnson: first issue, not the mwi issue |
| 22:17.00 | ajohnson | we are talking about ODBC voicemail storage, right? |
| 22:17.10 | ajohnson | I don't have an MWI issue |
| 22:17.24 | Corydon76-dig | ajohnson: Oh, right |
| 22:18.30 | ajohnson | well wait a second... why is Asterisk querying for a SIP peer whose name is empty? |
| 22:18.51 | ajohnson | right before storing a voicemail |
| 22:19.33 | Corydon76-dig | ajohnson: again, not talking about the voicemail issue at all yet |
| 22:19.52 | ajohnson | oh |
| 22:20.27 | Corydon76-dig | So is the 34th field of your sip_peers table something weird? |
| 22:20.37 | ajohnson | counting from 1 or 0? |
| 22:20.42 | Corydon76-dig | from 1 |
| 22:20.47 | ajohnson | disallow |
| 22:20.51 | ajohnson | = all |
| 22:21.04 | Corydon76-dig | Odd |
| 22:21.55 | Corydon76-dig | Okay, in the voicemail issue, do you still have access to the core? |
| 22:21.55 | ajohnson | I uploaded a tarball of the core to the issue |
| 22:21.57 | Corydon76-dig | because I'm curious about "p *data" |
| 22:22.01 | ajohnson | :) |
| 22:22.17 | Corydon76-dig | Yeah, but the corefile is useless to anybody other than the machine on which it was generated |
| 22:22.39 | ajohnson | that...make..sense... |
| 22:22.47 | Corydon76-dig | Or an exact duplicate |
| 22:22.54 | ajohnson | so I can download it or I can generate a new one... |
| 22:22.58 | ajohnson | it's very easy to reproduce |
| 22:22.59 | Corydon76-dig | Libraries, architecture, etc. |
| 22:23.06 | MikeJ | Corydon76-dig: its useful elsewhere to help fill empty disks! |
| 22:23.07 | Corydon76-dig | k |
| 22:23.10 | ajohnson | right, no it really does make sense |
| 22:23.41 | Corydon76-dig | MikeJ: I can do that with porn, though. :-P |
| 22:23.48 | MikeJ | thats true |
| 22:24.00 | MikeJ | or tail /dev/urandom > .... |
| 22:24.17 | MikeJ | actually.. |
| 22:24.36 | MikeJ | that probably just spins forever |
| 22:24.38 | MikeJ | thats a cpu test! |
| 22:25.10 | Corydon76-dig | A dd with no count does the same |
| 22:25.25 | Corydon76-dig | except the dd is a little more efficient |
| 22:25.37 | MikeJ | if=/dev/urandom :D |
| 22:25.44 | ajohnson | Corydon76-dig: updated ticket |
| 22:26.36 | Corydon76-dig | ajohnson: I know why it crashed and also how to fix |
| 22:26.50 | MikeJ | <PROTECTED> |
| 22:27.24 | ajohnson | sweet |
| 22:27.31 | ajohnson | assuming it's something wrong with my table |
| 22:29.43 | kpfleming | Corydon76-dig: fix has been tested on 10.4 on this intel mini and works... will commit shortly after one more test |
| 22:29.59 | Corydon76-dig | kpfleming: suh-weet! |
| 22:30.06 | kpfleming | stubs work too |
| 22:32.34 | Corydon76-dig | ajohnson: 15186 is ready for testing |
| 22:32.38 | ajohnson | okie |
| 22:34.23 | ajohnson | woot! |
| 22:40.23 | ajohnson | Corydon76-dig: Thanks for the help. |
| 22:41.33 | *** join/#asterisk-dev malcolmd (n=malcolmd@pdpc/sponsor/digium/malcolmd) |
| 22:41.33 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 22:42.30 | kpfleming | Corydon76-dig: new_optional_api branch updated... should work on PPC as well |
| 22:46.16 | seanbright | gotta love doing all that research and having a 2 liner fix it :) |
| 22:46.26 | kpfleming | and one of those lines was just an optimization |
| 22:46.39 | seanbright | silly programming. |
| 22:47.02 | kpfleming | silly osx linker having to be told to ignore undefined symbols when they are already declared as 'weak' |
| 22:47.23 | seanbright | that's weak |
| 22:47.25 | seanbright | ba dum dum |
| 22:47.33 | seanbright | and on that note... |
| 22:47.37 | seanbright | heads home |
| 22:51.39 | *** join/#asterisk-dev qdk (n=qdk@0x573d8dd1.bynqu1.dynamic.dsl.tele.dk) |
| 23:20.56 | *** join/#asterisk-dev outtolunc (n=me@c-71-198-197-201.hsd1.ca.comcast.net) |
| 23:28.12 | *** join/#asterisk-dev moy (n=moy@CPE001cdfec4cee-CM00080dab8485.cpe.net.cable.rogers.com) |
| 23:34.58 | *** join/#asterisk-dev eliel (n=eliels@120-17-235-201.fibertel.com.ar) |
| 23:34.58 | *** mode/#asterisk-dev [+o eliel] by ChanServ |