| 00:14.58 | leroybuckingham | In theory, I should be able to use something along the lines of netcat ( i know this is getting hacky) and grab the remote port from "sip show peer <peer>", and that should traverse NAT since UDP is connectionless, but what if SIP is set up for TCP? |
| 00:15.09 | leroybuckingham | I know this is low level networking stuff I probably don't need to bug you guys with |
| 02:49.48 | *** join/#asterisk-dev infobot (~infobot@rikers.org) |
| 02:49.48 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- http://xkcd.com/303/ |
| 07:16.59 | *** join/#asterisk-dev schmidts (~schmidts@Lo20-rt01.lm33.sil.at) |
| 07:17.01 | schmidts | good morning |
| 07:28.57 | *** join/#asterisk-dev pnlarsson (~niklas@hulken.infracom.se) |
| 08:01.18 | *** join/#asterisk-dev kriegerod (~krieger@chaf.tenet.odessa.ua) |
| 08:19.08 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:993f:7361:5261:b89a) |
| 08:19.23 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 08:19.23 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 08:20.29 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 09:55.47 | *** join/#asterisk-dev sgimeno (~chatzilla@163.117.211.10) |
| 11:50.49 | *** join/#asterisk-dev actzipild (~actzipild@postfix.eureko.ro) |
| 11:50.57 | *** part/#asterisk-dev actzipild (~actzipild@postfix.eureko.ro) |
| 12:01.45 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:01.45 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:31.37 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 12:31.37 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 12:47.37 | *** join/#asterisk-dev Dovid (Dovid@office.mypbxmanager.net) |
| 13:36.56 | *** join/#asterisk-dev joachim_- (~joachim@post.comvie.no) |
| 13:37.18 | joachim_- | Why wont cisco 7940 phones register after upgrading from asterisk 1.6 to 1.8 ? |
| 13:38.27 | russellb | #asterisk |
| 14:00.17 | *** join/#asterisk-dev espiceland (~erin@nat/digium/x-mvwjirhrgkwyvwoo) |
| 14:29.10 | *** join/#asterisk-dev asterisk-learner (~chatzilla@77.42.241.114) |
| 14:38.00 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |
| 14:55.17 | *** part/#asterisk-dev joachim_- (~joachim@post.comvie.no) |
| 14:55.22 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:55.22 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:59.13 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:59.13 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 15:01.32 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:c9d8:9ca8:42b7:c298) |
| 15:16.43 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 15:31.56 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 15:35.24 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:35.24 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:59.32 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 15:59.34 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 15:59.49 | *** part/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 16:06.54 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 16:06.54 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 16:12.50 | *** join/#asterisk-dev datachomper (~Adium@nat/digium/x-tgwsqwljgnzslavo) |
| 16:12.55 | *** part/#asterisk-dev datachomper (~Adium@nat/digium/x-tgwsqwljgnzslavo) |
| 16:14.08 | *** join/#asterisk-dev chazzam (~chazz@173-24-239-247.client.mchsi.com) |
| 16:29.38 | *** part/#asterisk-dev pnlarsson (~niklas@hulken.infracom.se) |
| 16:31.02 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:50.02 | *** join/#asterisk-dev zerohalo (~zerohalo@173-13-92-27-NewEngland.hfc.comcastbusiness.net) |
| 16:54.35 | *** join/#asterisk-dev rgagnon (~rgagnon@rrcs-71-42-183-54.sw.biz.rr.com) |
| 17:26.42 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 17:33.42 | *** join/#asterisk-dev moy (~moy@CPE002719f00364-CM0026f3a1701d.cpe.net.cable.rogers.com) |
| 17:52.05 | *** join/#asterisk-dev marv (~timr@173.14.206.193) |
| 17:52.55 | *** join/#asterisk-dev sgimeno (~chatzilla@163.117.211.10) |
| 17:55.53 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 18:04.29 | *** join/#asterisk-dev miguel3239 (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 18:54.16 | *** join/#asterisk-dev zerohalo (~zerohalo@173-13-92-27-NewEngland.hfc.comcastbusiness.net) |
| 19:06.16 | *** join/#asterisk-dev kriegerod (~krieger@user-204.45.infomir.com.ua) |
| 19:24.24 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 19:32.32 | rgagnon | anyone ever see this when compiling? /usr/bin/ld: /usr/lib/gcc/x86_64-redhat-linux/4.1.2/../../../../lib64/libresample.a(resample.c.o): relocation R_X86_64_32 against `a local symbol' can not be used when making a shared object; recompile with -fPIC |
| 19:35.00 | russellb | how did you install libresample? |
| 19:39.20 | leifmadsen | if you're on a 64-bit machine you may need to do: ./configure CFLAGS="-fPIC" |
| 19:45.25 | Qwell | seanbright: that patch must have taken you all day |
| 19:46.57 | *** join/#asterisk-dev alecdavis (~alec@210.48.127.227) |
| 19:51.23 | seanbright | Qwell: well i had to type in 'man curl_easy_setopt' |
| 20:08.14 | *** join/#asterisk-dev cyphorious (~cyphoriou@chello062178189196.2.15.tuwien.teleweb.at) |
| 20:17.12 | *** join/#asterisk-dev drivefast (~radu@adsl-99-92-126-154.dsl.lsan03.sbcglobal.net) |
| 20:18.25 | alecdavis | M18904 ready for testing, also on review board. now can use Voicemail and all other DTMF applications without her misreading the digits :) |
| 20:18.28 | MuffinMan | [ready for testing] [Asterisk] Applications/app_disa 0018904: [patch] dsp.c sends multiple DTMF key events up to applications reported by alecdavis https://issues.asterisk.org/view.php?id=18904 |
| 20:18.40 | alecdavis | I hate voicemail, if that matters. |
| 20:18.49 | Kobaz | all you have to do is ask your question |
| 20:18.57 | Kobaz | > drivefast |
| 20:19.05 | drivefast | thanks :) |
| 20:19.58 | drivefast | hello everybody, any one of you can offer some help in building custom modules? it's not the c code that i need help with, but the makefiles |
| 20:20.18 | Kobaz | you need to explain what you're doing and note any specific errors or problems, etc |
| 20:20.28 | drivefast | one of my modules uses an external library and i need to do the magic for checking if it is installed |
| 20:21.28 | drivefast | i added the "-lmydependency" compile option and it works, but it's probably the ugly way to do it |
| 20:21.48 | drivefast | i added that in apps/Makefile |
| 20:22.05 | drivefast | but i definitely have the feeling that theres more to it than that |
| 20:22.12 | Kobaz | are you planning to redistribute your changes? |
| 20:22.17 | drivefast | yes |
| 20:22.25 | Kobaz | part of the asterisk mainline? |
| 20:22.49 | drivefast | i am aware whats up the road - code reviews and stuff, but i have to do a nice build first :) |
| 20:22.51 | drivefast | yes |
| 20:23.08 | Kobaz | yeah, okay, that's good |
| 20:23.19 | drivefast | it's a module that works with a memcache server |
| 20:23.25 | Kobaz | so you're going to want to look into autoconf, since that's what's used to check dependencies |
| 20:23.42 | Kobaz | the ./configure script |
| 20:23.44 | drivefast | i tested it and it's doing well, i just need to clean it up |
| 20:23.58 | drivefast | ok... i hear ya |
| 20:24.20 | drivefast | i looked over there already, it was a bit chinese for me to be honest |
| 20:24.47 | drivefast | tried to imitate what was in for other internal dependencies |
| 20:24.55 | Qwell | drivefast: look at configure.ac |
| 20:25.01 | Qwell | and makeopts.in |
| 20:25.02 | drivefast | s/internal/external |
| 20:26.02 | drivefast | ok, i will look in there and come with more specific questions (or hopefully not :P ) |
| 20:26.09 | *** join/#asterisk-dev p3nguin (~xwQ5kwYl6@cpe-0210-4bff-fe2b-9074.dhcp.a2infotech.com) |
| 20:27.02 | drivefast | another thing that i need to do is to compile in a .c file and a .h file that are making up another library |
| 20:27.42 | drivefast | theyre written by somebody else, so i would rather elave the original files intact |
| 20:27.54 | Kobaz | you have to be careful with directly compiling in code that you want to add to mainline, the license has to be compatible and the author needs to give permission to digium to redistribute |
| 20:28.34 | Kobaz | if it's gpl, you can just link with it |
| 20:28.49 | drivefast | the .c file does compile if i put it in main, but compiles too late for my module to be taken care of |
| 20:29.12 | drivefast | (Kobaz: i will be back to the licensing stuff in a few minutes :) i know that) |
| 20:30.01 | drivefast | again, the code for my module works great, BUT... i copy-pasted all the original code from the static library on top of mine to test it |
| 20:30.16 | drivefast | which is totally the ugly way to do it :) |
| 20:30.33 | Kobaz | so what exactly is the issue, your module needs code from a main c file which is build later? |
| 20:30.40 | Kobaz | s/built |
| 20:32.02 | drivefast | 1mom pls |
| 20:33.11 | drivefast | so my code uses a few utility functions from a 3rd party |
| 20:33.29 | drivefast | that guy distributes the files as a .c and a .h |
| 20:34.15 | drivefast | i can add the .c and the .h in the apps/ directory and it looks to compile fine like that, but them my module fails to load at asterisk startup |
| 20:34.30 | drivefast | complains about the functions in the utility library not found |
| 20:34.35 | drivefast | OR |
| 20:34.58 | drivefast | i can have the .c utilities file in main, and the .h in include/asterisk |
| 20:35.18 | drivefast | but in this case the compile too late - my module gets compiled first |
| 20:35.47 | Kobaz | the stuff in apps/ doesn't get all linked together |
| 20:35.52 | Kobaz | it's dynamicaly loaded at run time |
| 20:35.57 | drivefast | right |
| 20:36.05 | Kobaz | so if you put something in apps/ and you have something in say res/ |
| 20:36.15 | Kobaz | you need to load the module in asterisk |
| 20:36.23 | Kobaz | and then load your own |
| 20:36.34 | Kobaz | or, compile it right into the base, which would be in main/ |
| 20:38.04 | drivefast | mmm... so let's say that my module (the one i wrote) is app/app_mymodule.c |
| 20:38.24 | drivefast | and uses code from someutilities.c |
| 20:38.48 | Kobaz | someutilsities.c belongs in main/ |
| 20:38.54 | drivefast | ok |
| 20:39.36 | drivefast | and how do i tell the linker "for you to create app_mymodule.so, you have to link app_mymodule.o AND main/someutilities.o" |
| 20:40.06 | drivefast | (or am i just getting it totally wrong? :) ) |
| 20:40.14 | Kobaz | the makefile takes care of building main/* first |
| 20:40.35 | *** join/#asterisk-dev hehol (~Adium@ip-78-94-0-76.unitymediagroup.de) |
| 20:41.13 | drivefast | yeah, thats what i thought too |
| 20:42.09 | drivefast | but for example i see, during the make, on the creation of some .so modules, that multiple .o (object?) files are used to create that .so |
| 20:43.05 | drivefast | which is not what i see when it comes to my module... it only says it uses app_mymodule.o to create app_mymodule.so |
| 20:43.18 | Kobaz | the apps/*.so files are dynamically loaded into the core, so anything linked into the core (that's not static) will be available |
| 20:43.44 | Kobaz | the loading is all done runtime for modules |
| 20:44.05 | drivefast | 1 more moment pls - what you say makes sense |
| 20:44.11 | *** join/#asterisk-dev espiceland (~erin@207.98.195.107) |
| 20:52.41 | drivefast | Kobaz: according to what you say, i should have main/someutilities.c include/asterisk/someutilities.h and apps/app_mymodule.c |
| 20:53.01 | drivefast | which is what i did, and the make process went all the way |
| 20:54.05 | drivefast | however, when i start asterisk after that, it immediately complains about my module not finding a function from the someutilities library |
| 20:54.16 | drivefast | and, as a result, my module is not loaded |
| 20:54.23 | drivefast | does it make sense? |
| 20:55.56 | Kobaz | yeah hmm, i've never added a new file to main, you'll probably have to monkey the makefiles |
| 20:56.39 | drivefast | i tried, but the result was the same... (and i'm a good monkey) |
| 20:57.10 | Kobaz | you added to OBJS in main/Makefile ? |
| 20:57.25 | drivefast | lemme check |
| 20:59.36 | drivefast | mmmno, not specifically, let me try and tell you how it worked |
| 21:12.56 | drivefast | nope... didnt help |
| 21:15.37 | drivefast | loader.c: Error loading module 'app_mymodule.so': /usr/lib/asterisk/modules/app_mymodule.so: undefined symbol: PrintUnformatted |
| 21:17.38 | Kobaz | are you functions in someutilities.c defined as static? |
| 21:17.57 | drivefast | checking |
| 21:18.43 | drivefast | hell yeah, a ton of them |
| 21:19.46 | Kobaz | this is starting to turn into something for #c |
| 21:19.48 | Kobaz | heh |
| 21:19.56 | drivefast | surprisingly the one singled out by the loader is not static |
| 21:20.24 | drivefast | lol |
| 21:20.33 | Kobaz | have you verified that your util source is being compiled and linked into the asterisk binary |
| 21:20.46 | drivefast | yes |
| 21:21.16 | drivefast | so i do have a matching .o file (and i had it even before i added it explicitely to OBJS) |
| 21:22.00 | drivefast | i dont know if it matters or not, but my utilities object file gets created later than app_mymodule.so |
| 21:22.41 | drivefast | and the hint is in the messages from the linker |
| 21:23.38 | drivefast | here's what i mean: |
| 21:23.43 | drivefast | [LD] format_mp3.o mp3/common.o mp3/dct64_i386.o mp3/decode_ntom.o mp3/layer3.o mp3/tabinit.o mp3/interface.o -> format_mp3.so |
| 21:23.58 | drivefast | this is one of the lines that i wish i could imitate |
| 21:24.28 | drivefast | i think it says that "building format_mp3.so tool using all those .o files" |
| 21:24.42 | seanbright | linking |
| 21:24.44 | seanbright | not building |
| 21:24.51 | drivefast | i wish i could tell the link editor the same thing... somewhere |
| 21:24.53 | seanbright | but you know that. |
| 21:26.16 | seanbright | i'm fairly certain that you need to export the symbols somehow |
| 21:26.50 | seanbright | take a look at main/asterisk.exports |
| 21:27.30 | drivefast | looking... |
| 21:28.26 | drivefast | whoa... sounds interesting |
| 21:28.33 | drivefast | lemme try something |
| 21:28.58 | seanbright | add PrintUnformatted to that list and rebuild |
| 21:29.05 | seanbright | if it works, you may thank me |
| 21:29.38 | drivefast | i may indeed, im hopeful :) |
| 21:36.41 | drivefast | hm, not yet, but i've one more idea... this seems to be on the right track |
| 21:39.44 | drivefast | make clean; make; make install again... fingers crossed |
| 21:39.47 | Kobaz | seanbright: i haven't had to add anything to any export lists when i added a new function to an existing c file in main/ |
| 21:40.55 | drivefast | Kobaz: could it be because you added it to an existing file?... which would have all the functions exported thru another mechanism?... |
| 21:41.10 | drivefast | im adding a new file and this may be the difference |
| 21:42.49 | Kobaz | not familiar with those parts, not sure |
| 21:43.08 | drivefast | i'm only guessing myself too |
| 21:55.10 | drivefast | nah :( didnt work the way i tried it either... |
| 21:55.23 | drivefast | same error |
| 21:55.56 | drivefast | i got to somehow force the link editor to pick the symbols from the other library... |
| 21:56.39 | seanbright | Kobaz: was it named ast_ something? |
| 21:57.22 | seanbright | drivefast: let me whip up a test... what version are you compiling against? |
| 21:57.46 | seanbright | drivefast: check asterisk.exports again... are your changes actually there? |
| 21:58.08 | seanbright | drivefast: asterisk.exports is overwritten by asterisk.exports.in during the build process |
| 21:58.30 | drivefast | checking |
| 21:58.38 | drivefast | i'm on a 1.8.... |
| 21:58.43 | seanbright | ok |
| 21:58.47 | drivefast | 1.8.0 |
| 21:59.00 | seanbright | i'll test with 1.8 SVN |
| 21:59.19 | seanbright | unless you confirm that your asterisk.exports got blown away during the build |
| 21:59.49 | drivefast | youre right my entry got out of there |
| 21:59.59 | drivefast | let me add it to .in and try again |
| 22:03.37 | seanbright | alrightly |
| 22:03.41 | seanbright | i'll wait here |
| 22:04.43 | drivefast | so ./configure didnt move the entry from asterisk.exports.in into asterisk.exports - maybe make will do it |
| 22:07.44 | drivefast | sean, you're bright, man |
| 22:07.53 | drivefast | my module got loaded |
| 22:08.22 | drivefast | let's see if it passes the unit tests now |
| 22:09.00 | drivefast | aaaand, it does! woo-hoo |
| 22:09.28 | drivefast | thank you thank you thank you |
| 22:11.12 | drivefast | Kobaz, thank you to you too, i will be looking now to put in the checks for the dependencies as you suggested |
| 22:12.09 | drivefast | coming back to the licensing issue now, here's what's stated by the author, in the header of the files: |
| 22:12.36 | drivefast | <PROTECTED> |
| 22:12.37 | drivefast | <PROTECTED> |
| 22:12.37 | drivefast | <PROTECTED> |
| 22:12.37 | drivefast | <PROTECTED> |
| 22:12.38 | drivefast | <PROTECTED> |
| 22:12.38 | drivefast | <PROTECTED> |
| 22:12.38 | drivefast | <PROTECTED> |
| 22:12.38 | drivefast | <PROTECTED> |
| 22:12.39 | drivefast | <PROTECTED> |
| 22:13.29 | Kobaz | seanbright: yeah ast_* |
| 22:14.37 | Kobaz | i think that might be okay |
| 22:14.50 | Kobaz | permission to distribute and sublicense is good |
| 22:15.18 | Kobaz | drivefast: you'll have to update the library to be in check with the asterisk coding standards |
| 22:16.12 | Kobaz | http://www.asterisk.org/developers/coding-guidelines |
| 22:16.16 | drivefast | umm... thats gonna be a little difficult |
| 22:16.23 | russellb | maybe ... in some cases it's not worth it, in case we want to keep it in sync with where it came from |
| 22:16.26 | drivefast | yeah i know them |
| 22:16.27 | russellb | the asterisk module for sure |
| 22:16.38 | russellb | but a supporting library, meh, not as big of a deal |
| 22:16.45 | drivefast | the asterisk module *is* taken care of |
| 22:17.34 | drivefast | but i wouldnt really go into spice up this guy's work with the asterisk standards... reason being some of the very functions overlap |
| 22:18.02 | drivefast | e.g. he also has a safe malloc(), and a safe strncpy() |
| 22:18.03 | Kobaz | you should use internal asterisk functions when possible if there's overlapping functionality in the lib |
| 22:19.21 | drivefast | here's what i plan to do: i will do all the build stuff first, so that it is up to the standards |
| 22:19.26 | drivefast | add the documentation |
| 22:19.41 | drivefast | leave the functions in this extra library as they are |
| 22:20.01 | drivefast | and throw it for review in the lions cage :) |
| 22:23.03 | drivefast | it will take a few more days, i also have a job to take care of in the mean time |
| 22:23.10 | seanbright | drivefast: no sweat (sorry, was afk) |
| 22:23.35 | drivefast | but i will hang out here and keep in touch with you |
| 23:22.44 | russellb | The_Boy_Wonder: ping |
| 23:22.53 | The_Boy_Wonder | russellb: hey |
| 23:22.59 | russellb | re: confbridge reload |
| 23:23.10 | russellb | Did you also make it so "module reload app_confbridge.so" works |
| 23:23.11 | russellb | ? |
| 23:23.16 | The_Boy_Wonder | yes |
| 23:23.18 | The_Boy_Wonder | they both work |
| 23:23.32 | russellb | That's really my preferred way ... the specific reload CLI commands are (IMO) a deprecated method from before we had the generic reload CLI command |
| 23:23.56 | The_Boy_Wonder | well, they both do the same thing |
| 23:24.01 | russellb | right |
| 23:24.14 | russellb | just "module reload ..." seems sufficient |
| 23:24.35 | russellb | or else what's the point of "module reload" |
| 23:25.31 | The_Boy_Wonder | so you think i should take 'confbridge reload' out? i |
| 23:25.37 | The_Boy_Wonder | m just used to sip reload and iax2 reload |
| 23:25.50 | russellb | right, those commands are ancient and have been there since before "module reload" |
| 23:26.07 | russellb | dialplan reload, too .. |
| 23:26.50 | The_Boy_Wonder | so is this a request to take it out? or were you just pointing out the module reload stuff? |
| 23:26.57 | russellb | it's a request to take it out |
| 23:27.02 | The_Boy_Wonder | hehe |
| 23:27.54 | The_Boy_Wonder | done |
| 23:28.00 | russellb | cool |
| 23:28.21 | russellb | was just trying to explain my stance before just flat out saying "plz take it out kthxbye" |
| 23:28.48 | The_Boy_Wonder | alrighty, works for me |