| 00:04.43 | *** join/#asterisk-dev qdk (n=qdk@0x573fe57e.bynqu2.static.dsl.tele.dk) |
| 01:05.39 | *** part/#asterisk-dev simond (n=simon@syria.uc.org) |
| 01:51.21 | JerJer | hmmm - this may have been fixed by now (lets hope) but I just got informed of a system getting hit with a very easily reproducable remote DoS with SIP |
| 01:52.25 | JerJer | on 1.4.19 |
| 01:54.39 | *** join/#asterisk-dev mvanbaak_ (n=michiel@vanbaak.xs4all.nl) |
| 01:54.40 | *** mode/#asterisk-dev [+o mvanbaak_] by ChanServ |
| 01:54.40 | JerJer | in a register |
| 01:56.05 | JerJer | Via: SIP/2.0/UDP localhost:5060;branch=z9hG4bK-2358824857;rport. |
| 01:58.58 | JerJer | causes asterisk to spew [Jul 7 01:45:36] WARNING[17832]: chan_sip.c:9232 check_via: 'localhost' is not a valid host |
| 02:00.25 | JerJer | with less than 3 megabit of bandwidth it brought down a dual xeon box |
| 02:04.40 | JerJer | then wouldn't start processing SIP normally again until the IP address that was spewing invalid register messages was blocked upstream |
| 02:08.42 | *** join/#asterisk-dev IgorG (n=IgorG@195.162.49.23) |
| 02:28.03 | tzanger | nice |
| 02:28.13 | tzanger | got any production boxes I could try that out on for you? |
| 02:41.20 | *** join/#asterisk-dev jeffg (i=gS0UW6aq@pdpc/supporter/active/jeffg) |
| 03:46.57 | *** join/#asterisk-dev obnauticus (n=obnautic@c-67-160-181-107.hsd1.wa.comcast.net) |
| 03:59.40 | *** join/#asterisk-dev bkruse (n=bkruse@76.73.154.120) |
| 03:59.41 | *** mode/#asterisk-dev [+o bkruse] by ChanServ |
| 03:59.50 | bkruse | mvanbaak_: ello |
| 04:12.37 | *** join/#asterisk-dev cristina_crow (n=cvintila@212.146.94.66) |
| 04:13.40 | *** part/#asterisk-dev cristina_crow (n=cvintila@212.146.94.66) |
| 05:21.18 | *** join/#asterisk-dev tzafrir (n=tzafrir@local.xorcom.com) |
| 05:21.18 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 05:21.19 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@local.xorcom.com) |
| 05:21.19 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 05:24.04 | *** join/#asterisk-dev bkruse (n=bkruse@76.73.154.120) |
| 05:24.04 | *** mode/#asterisk-dev [+o bkruse] by ChanServ |
| 05:24.56 | *** part/#asterisk-dev bkruse (n=bkruse@76.73.154.120) |
| 05:52.02 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.133) |
| 06:05.50 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 06:19.08 | mvanbaak | mornin |
| 06:35.51 | *** join/#asterisk-dev monstertruck (n=monstert@191-172.dsl.scrtc.com) |
| 06:37.33 | monstertruck | quick question guys, cant find this anywhere in the iax2 draft |
| 06:37.42 | monstertruck | i get this HANGUP from * : |
| 06:37.55 | monstertruck | Tx-Frame Retry[000] -- OSeqno: 004 ISeqno: 002 Type: IAX Subclass: HANGUP |
| 06:37.56 | monstertruck | <PROTECTED> |
| 06:37.56 | monstertruck | <PROTECTED> |
| 06:38.10 | monstertruck | and reply with this ACK: |
| 06:38.22 | monstertruck | Rx-Frame Retry[ No] -- OSeqno: 002 ISeqno: 004 Type: IAX Subclass: ACK |
| 06:38.22 | monstertruck | <PROTECTED> |
| 06:38.35 | monstertruck | as far as i can tell, the ACK is well formed |
| 06:38.53 | monstertruck | why does * feel the need to retransmit the HANGUP 3 more times? |
| 07:18.24 | mvanbaak | does the remote box get the ACK in time ? |
| 07:36.26 | monstertruck | client and server are running in the same machine |
| 07:36.47 | monstertruck | and the response is immediate |
| 07:38.22 | mvanbaak | then I dont know. |
| 07:38.25 | mvanbaak | sorry |
| 07:38.36 | monstertruck | thanks anyways |
| 08:09.03 | *** join/#asterisk-dev bkruse (n=bkruse@76.73.154.120) |
| 08:09.03 | *** mode/#asterisk-dev [+o bkruse] by ChanServ |
| 08:22.13 | *** join/#asterisk-dev qdk (n=qdk@85.235.253.139) |
| 09:01.14 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 09:01.36 | puzzled | hi |
| 09:51.30 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 10:22.16 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 10:49.54 | *** join/#asterisk-dev _ys (i=yuri@91.151.196.254) |
| 11:01.05 | *** join/#asterisk-dev friendly12345 (n=friendly@ppp59-167-145-230.lns4.mel6.internode.on.net) |
| 11:10.47 | *** part/#asterisk-dev jmls (n=jmls@host217-36-208-155.in-addr.btopenworld.com) |
| 11:11.40 | *** join/#asterisk-dev jmls (n=jmls@host217-36-208-155.in-addr.btopenworld.com) |
| 11:14.32 | *** part/#asterisk-dev jmls (n=jmls@host217-36-208-155.in-addr.btopenworld.com) |
| 11:31.55 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:36.37 | *** join/#asterisk-dev DagMoller (n=aguirre@unaffiliated/dagmoller) |
| 11:55.22 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 11:55.23 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 12:00.15 | snuff-home | ello russellb |
| 12:02.19 | Guest32081 | russellb: ! |
| 12:02.27 | Guest32081 | wow |
| 12:02.34 | russellb | waves |
| 12:02.39 | russellb | bkruse: o.O |
| 12:02.58 | bkruse | russellb: I am already at work! |
| 12:03.20 | russellb | hardcore. |
| 12:03.25 | russellb | i'm in the atlanta airport ... |
| 12:03.27 | russellb | tired as hell |
| 12:04.07 | bkruse | russellb: I bet man, when you going to be back? Bad flight scheduling.... |
| 12:04.31 | bkruse | Ya, keith said i am a "real employee" now, so I have to work "full time" and "come in on saturday" |
| 12:04.53 | russellb | heh. |
| 12:04.53 | snuff-home | hehe |
| 12:05.03 | russellb | i'll be at work by lunch sometime |
| 12:05.25 | bkruse | russellb: You need to sleep. |
| 12:05.28 | snuff-home | mm.. in the physical sense anyway right russ :) |
| 12:05.43 | russellb | snuff-home: that is correct :) |
| 12:07.11 | snuff-home | i figure i'll do some more docxml for mvanbaak.. nice easy thing |
| 12:09.52 | bkruse | snuff-home: yay! |
| 12:10.13 | snuff-home | app_page and 1 or 2 others tonight.. |
| 12:10.39 | bkruse | i'll hopefully get time to work on it at the end of the day, or during if I am waiting on something, get the parsing stuff straightened out, then it will just be converting apps/res/static documentation into the XML format |
| 12:10.43 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:10.45 | snuff-home | mm i made my own dialplan for app_page when i couldn't get it to work properly ;) |
| 12:11.14 | bkruse | after/during/before that we can do more schema (like return types of functions, that's awesome.) |
| 12:11.19 | snuff-home | nothin like dial(sip/blah,G(),40 |
| 12:12.04 | russellb | don't do too many |
| 12:12.08 | russellb | we have to decide on the format first |
| 12:12.13 | russellb | don't want any wasted effort :) |
| 12:12.46 | snuff-home | well its not like its hard to convert stuff.. |
| 12:13.24 | russellb | ok, as long as you know it's subject to change |
| 12:14.01 | snuff-home | mm its all good.. i figure it can't change drastically.. least i hope :) |
| 12:15.03 | snuff-home | most apps aside from app_dial are relatively simple to xml-ify |
| 12:15.44 | bkruse | snuff-home: Right, and me and you and mvanbaak have been making changes, adding/removing things from the schema, and it was pretty minimal |
| 12:16.22 | tzafrir_laptop | What exactly is the format? What happens when the documents format changes? |
| 12:16.34 | tzafrir_laptop | (the exact ABI format) |
| 12:16.51 | bkruse | tzafrir_laptop: Format is defined in the doc/xmldocumentation.txt |
| 12:16.51 | russellb | an XML document that gets loaded at runtime |
| 12:17.48 | tzafrir_laptop | What is the impact of a rouge module with a missing > ? |
| 12:17.58 | russellb | fails to load |
| 12:17.59 | tzafrir_laptop | Or a buggy one? |
| 12:18.12 | russellb | of course, that would be checked before a release is made |
| 12:18.31 | russellb | buggy? |
| 12:18.32 | tzafrir_laptop | You assume all modules are internal |
| 12:19.02 | russellb | i was thinking we would have a directory where you could install more documentation in their own XML files |
| 12:19.05 | bkruse | they can still register documentation the old way, this method is only used if NULL is passed |
| 12:19.12 | russellb | so, if your docs for an external module are broken, it won't break what we distribute |
| 12:19.25 | bkruse | That's good too |
| 12:19.52 | *** join/#asterisk-dev Rem| (n=remi@bg-fw2out.monmouth.com) |
| 12:24.46 | snuff-home | mmm reminds me i forgot about the 'required=' tag |
| 12:25.42 | bkruse | ? |
| 12:25.57 | bkruse | required= was a good addition :) |
| 12:26.49 | snuff-home | mm need to fix my app_fax patch on bugs then |
| 12:30.28 | snuff-home | ok i'm not going to tackle app_directory.. too much if something changes :) |
| 12:32.10 | bkruse | hehe, I was going to tackle main/pbx.c builtin apps |
| 12:32.11 | bkruse | wow.... |
| 12:32.56 | snuff-home | also 'enum' wouldn't be lot of fun to do either.. it has a bit of stuff.. |
| 12:33.32 | bkruse | we will leave that for mvanbaak :D |
| 12:36.10 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:36.10 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 12:39.32 | snuff-home | ack bkruse chanspy also one avoiding until concrete format ;) |
| 12:39.49 | bkruse | hehehe, very true |
| 12:40.21 | snuff-home | so glad dont have to use * for recording/spying seems to be lot of ppls troubles with 1.4 |
| 12:52.12 | snuff-home | mm anyone ever tried 'echo' application with video? |
| 12:53.08 | snuff-home | bkruse: with these changes to app code.. wont the ASTERISK_FILE_VERSION have to go up? |
| 13:01.04 | moy | howdy, any one can tell me when should I use |= AST_SOFTHANGUP_DEV vs ast_queue_hangup inside a CD?? I even see at least 1 piece of code in chan_zap that uses both (useless as far as I can see) |
| 13:10.35 | *** join/#asterisk-dev jeffg (i=cCYj32pg@pdpc/supporter/active/jeffg) |
| 13:11.55 | bkruse | snuff-home: What do you mean? |
| 13:12.40 | snuff-home | generally when u change stuff in one of those app files and then try to compile with mtx_profile etc.. throws toys out of pram |
| 13:12.54 | Corydon76-dig | moy: |= if you're doing it to yourself. ast_queue_hangup if you're doing it to another channel (a bridged channel) |
| 13:14.50 | moy | Corydon76-dig, thanks |
| 13:16.12 | bkruse | oh....hmm |
| 13:16.35 | snuff-home | that was a while ago.. so dont quote me :) |
| 13:16.44 | snuff-home | last app then i will upload to bug |
| 13:20.07 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.228.111) |
| 13:25.53 | *** join/#asterisk-dev RoyK (n=roy@ti211210a080-3097.bb.online.no) |
| 13:26.59 | snuff-home | bkruse: updated bug 13003 with extra stuff i did.. attached as new patch |
| 13:27.22 | bkruse | nice nice nice |
| 13:44.26 | bkruse | geez, i think I finally fixed it |
| 13:47.53 | snuff-home | mm time to go snoozin.. later ppl |
| 13:48.06 | bkruse | cya snuff-home |
| 13:48.26 | bkruse | M13003 |
| 13:50.35 | *** join/#asterisk-dev jainr (n=jainr@65.244.37.75) |
| 13:55.37 | *** join/#asterisk-dev putnopvut (n=putnopvu@216.207.245.1) |
| 13:55.37 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:00.46 | bkruse | putnopvut: ! |
| 14:00.59 | putnopvut | Hi bkruse. How was thy weekend? |
| 14:01.31 | bkruse | putnopvut: It was awesome! you? |
| 14:03.04 | putnopvut | Good. I went kayaking, celebrated my mom's birthday, and went to a really long baseball game. |
| 14:03.51 | bkruse | putnopvut: Kayaking!? Nice! |
| 14:04.21 | putnopvut | Yeah, it was fun, but a lot of work. I sure slept well after that :) |
| 14:05.56 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 14:23.09 | caio1982 | does anyone here feels fine about changing this in trunk? http://pastebin.com/f5e6946d8 |
| 14:23.20 | caio1982 | a too tiny patch to post on mantis :) |
| 14:23.53 | caio1982 | it's a documentation change! /me waits for doc people's alarm ring :) |
| 14:27.05 | putnopvut | caio1982: looks good except that the second part of the patch needs to use the word "a" instead of "an" before "dialplan reload" |
| 14:27.55 | caio1982 | oh of course, it was 'extensions' before, sorry :) |
| 14:28.44 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.138) |
| 14:29.42 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 14:31.11 | putnopvut | caio1982: do you mind of I commit that now? |
| 14:31.45 | caio1982 | putnopvut: please :) |
| 14:34.11 | *** join/#asterisk-dev joetester2 (n=chatzill@216.191.34.13) |
| 15:07.44 | Juggie | putnopvut, you own kayaks? |
| 15:07.57 | putnopvut | No, I rented one. |
| 15:08.16 | Juggie | ah |
| 15:08.23 | Juggie | my gf's sister owns 4. |
| 15:08.33 | Juggie | they are nice too, like 14' footers... tons of fun |
| 15:08.38 | putnopvut | cool. I dont' really live close enough to water to justify owning them. |
| 15:09.05 | Juggie | we live in canada, lots of water up here :) |
| 15:21.49 | *** join/#asterisk-dev Schmooze (n=Schmooze@rrcs-67-53-189-178.west.biz.rr.com) |
| 15:21.52 | Qwell | bkruse: doc language - cool |
| 15:21.53 | *** join/#asterisk-dev shido6 (n=shido6@209.114.208.192) |
| 15:22.56 | bkruse | Qwell: :D |
| 15:28.16 | *** join/#asterisk-dev outtolunc (n=me@c-67-170-211-86.hsd1.ca.comcast.net) |
| 15:31.53 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:31.53 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:31.58 | *** join/#asterisk-dev _ys (n=yuri@80.70.236.69) |
| 15:33.16 | Qwell | bkruse: have you guys considered functions? |
| 15:33.59 | bkruse | Qwell: func_shell.c is done |
| 15:34.11 | Qwell | ahh |
| 15:34.28 | bkruse | anything in MOD_DIRS gets parsed |
| 15:37.13 | *** join/#asterisk-dev jtodd (n=jtodd@blob.fox-den.com) |
| 15:37.13 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 15:37.31 | bkruse | jtodd: ! |
| 15:37.38 | jtodd | Greetings. |
| 15:37.58 | *** join/#asterisk-dev jeremy_g (n=jerms@static-213-115-44-90.sme.bredbandsbolaget.se) |
| 15:38.24 | file | waves to jtodd |
| 15:38.36 | jtodd | waves back. |
| 15:38.36 | ctooley | jtodd, good morning |
| 15:40.13 | *** join/#asterisk-dev sruffell (n=sruffell@216.207.245.1) |
| 15:40.13 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 15:44.34 | *** join/#asterisk-dev qdk (n=qdk@0x573fe57e.bynqu2.static.dsl.tele.dk) |
| 15:51.11 | *** join/#asterisk-dev bkw_ (n=brian@adsl-70-234-182-25.dsl.tul2ok.sbcglobal.net) |
| 16:08.34 | *** join/#asterisk-dev Cresl1n (n=matt@216.207.245.1) |
| 16:08.34 | *** mode/#asterisk-dev [+o Cresl1n] by ChanServ |
| 16:15.32 | JerJer | is there a chan_spi !? |
| 16:16.08 | file | serial peripheral interface? |
| 16:16.58 | ctooley | I'm more curious what the special ascii characters between the i and the ! are? Those could be hiding a lot of details. |
| 16:16.58 | JerJer | i think someone means chan_SIP |
| 16:18.25 | JerJer | ohhhh - chan_spy |
| 16:19.53 | JerJer | i picked the wrong week to stop sniffing glue |
| 16:20.36 | seanbright | classic. |
| 16:20.50 | caio1982 | isnt it that quote from Airplane? |
| 16:20.52 | seanbright | roger, roger. |
| 16:20.56 | seanbright | what's your vector, victor? |
| 16:20.57 | caio1982 | oveur |
| 16:21.03 | caio1982 | haha |
| 16:21.03 | JerJer | apparently I is a Y in some other language ?!?!?! loverly |
| 16:21.35 | seanbright | JerJer: don't forget, in the latin alphabet, jehovah begins with an 'i' |
| 16:21.55 | jeremy_g | is this dev or hollywood |
| 16:22.19 | seanbright | neither |
| 16:22.22 | seanbright | both |
| 16:23.25 | caio1982 | not hollywood, they were flying to chicago :P |
| 16:25.01 | JerJer | file: yeah that's what I was initially thinking - since the support request mentioned agents and not getting the data they need |
| 16:26.02 | JerJer | but its a simple chan_spy issue on BE |
| 16:26.41 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 16:26.41 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 16:28.01 | JerJer | caio1982: :) |
| 16:31.18 | JerJer | seanbright: nice - another good movie reference |
| 16:35.14 | seanbright | bbryant: ping? |
| 16:36.41 | caio1982 | should not 'core show globals' be 'diaplan show globals'? |
| 16:38.42 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 16:38.43 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 16:40.52 | seanbright | bbryant: i updated 12876, but i might not be 100% correct. |
| 16:52.03 | kpfleming | tzafrir: yeah, on 12911, sorry about that. clearly my first fix was not correct :-( |
| 17:01.13 | bbryant | seanbright: thanks |
| 17:10.01 | *** join/#asterisk-dev Laureano (n=Laureano@190.245.115.150) |
| 17:10.05 | Laureano | Hello guys! |
| 17:10.18 | putnopvut | Hi Laureano! |
| 17:10.44 | Laureano | ABI stands for: Application Binary Interface, right? |
| 17:11.37 | putnopvut | That's correct. |
| 17:14.08 | Laureano | It took a long time to figure out that. I should searched Google first. Thanks putnopvut. |
| 17:15.09 | *** join/#asterisk-dev ctooley (n=ctooley@209.33.108.119) |
| 17:15.09 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 17:26.22 | putnopvut | What's the vim setting to enforce case-sensitive searches? |
| 17:28.23 | putnopvut | Found it |
| 17:28.26 | putnopvut | :set noignorecase |
| 17:29.53 | russellb | neat. |
| 17:33.55 | *** join/#asterisk-dev jets (n=brian@pdpc/supporter/active/jets) |
| 17:35.33 | *** join/#asterisk-dev mvanbaak (n=michiel@vanbaak.xs4all.nl) |
| 17:35.34 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 17:42.19 | jpeeler | :set noic for short! |
| 17:44.18 | *** join/#asterisk-dev Corydon76-lap (n=Corydon7@pdpc/supporter/bronze/Corydon76-home) |
| 17:44.18 | *** mode/#asterisk-dev [+o Corydon76-lap] by ChanServ |
| 17:46.38 | bkruse | might go early to the meeting so I won't be at the door, haha |
| 17:47.06 | file | <PROTECTED> |
| 17:47.18 | file | ... silly |
| 17:51.06 | Corydon76-lap | file is always at the door! |
| 17:52.31 | *** join/#asterisk-dev joetester2 (n=chatzill@216.191.34.13) |
| 18:07.09 | *** join/#asterisk-dev jameswf-home (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 18:09.54 | jameswf-home | man take a few days off foer a holiday and missed more bitching.... That settles it no more vacations |
| 19:25.14 | russellb | orly? |
| 19:25.53 | Corydon76-lap | pokes jsmith-teaching |
| 19:26.01 | jsmith-teaching | Corydon76-dig: Yes? |
| 19:26.17 | Corydon76-lap | Hmm, thought you were teaching... |
| 19:26.25 | jsmith-teaching | Corydon76-dig: Lab time! |
| 19:26.37 | Corydon76-lap | Aha |
| 19:44.15 | *** join/#asterisk-dev d1mas (n=chatzill@ip195.117.adsl.wplus.ru) |
| 19:46.03 | d1mas | hello. Anyone with good knowledge of SDP around? |
| 19:48.24 | file | a specific question may get a specific answer |
| 19:48.57 | russellb | ~ask |
| 19:48.58 | jbot | i heard ask is Questions in the channel should be specific, informative, complete, concise, and on-topic. Don't ask if you can ask a question first. Don't ask if a person is there; just ask what you intended to ask them. Better questions more frequently yield better answers. We are all here voluntarily or against our will. |
| 19:49.01 | russellb | :) |
| 19:51.01 | d1mas | okay... First, the problem: since 1.6, Asterisk starts ignoring re-invites with SDP version (one of values in the o= parameter) equal to the last seen version. I'm trying to use Kapanga sofrphone and faxing. When Asterisk sends T38 re-invite to Kapanga, softphone replies with proper OK but its version is the same as in the OK which was used to initially answer the call. As result, Asterisk... |
| 19:51.03 | d1mas | ...ignores that SIP packet and nothing happens. |
| 19:51.47 | d1mas | The question: who is wrong according to SIP/SDP standards? Should I blame Asterisk :) or send report to Kapanga ? |
| 19:54.18 | d1mas | what does jbot say about ~answer? ;) |
| 19:55.04 | file | have you tracked down the code that is causing it to happen, gotten the revision that it was introduced in, and read the commit message/comments in code to see if they yield some info? |
| 19:56.22 | d1mas | file: well... probably not to the same level of details you expect... |
| 20:11.54 | d1mas | ok, tracked down to the level of commit - 98978. So nothing new here - I already knew this change was introduced when adding SIP session timers... |
| 20:16.11 | jsmith-teaching | d1mas: You might try asking about it on the -dev list. I know that the guy that wrote the SIP timers patch is pretty good about responding on the list, and is very knowledgeable about the SIP protocol |
| 20:16.48 | d1mas | Raj? Sure I will ask. thanks |
| 20:18.43 | jsmith-teaching | Yes, Raj. |
| 20:25.49 | tzafrir_laptop | Qwell, while dahdi-perl is only installed if the xpp is selected in utils, I think it is useful for other drivers as well |
| 20:26.50 | tzafrir_laptop | For instance, I think it would be useful to change the init.d script so that if MODULES is not set, the list of modules as detected from DAHDI::Hardware will be loaded |
| 20:27.00 | tzafrir_laptop | This is a perl one-liner |
| 20:27.23 | kpfleming | then why don't you just let the kernel's autoload feature handle it? |
| 20:27.50 | tzafrir_laptop | if we let autoload handle it, we don't need that part of the init script |
| 20:28.04 | kpfleming | you only need that part of the script for non-autoload cases |
| 20:28.31 | tzafrir_laptop | This is useful for times when the init.d script is loaded "manually" (not at boot) |
| 20:28.31 | anonymouz666 | kpfleming: the DNS SRV stuff you did went to 1.4? |
| 20:28.41 | tzafrir_laptop | and never guaranteed to give the same order anyway |
| 20:28.50 | kpfleming | i have no idea what you are asking about... i haven't worked on SRV in many months |
| 20:29.17 | russellb | anonymouz666: svn log is your friend :) |
| 20:29.25 | kpfleming | it is just as guaranteed as using DAHDI::Hardware is |
| 20:30.06 | tzafrir_laptop | What I mean is: whatever you write in MODULES and the autoload modules |
| 20:30.40 | tzafrir_laptop | The idea is not to default to modprobe "all the modules" |
| 20:30.48 | kpfleming | i don't understand should load *any* modules if nothing has been explicitly requested |
| 20:31.03 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 20:31.13 | kpfleming | if the user hasn't filled in MODULES, then they shouldn't be expecting the init script to do anything |
| 20:31.23 | tzafrir_laptop | Because with the default/sample config we ship we have most modules included in MODULES |
| 20:31.32 | tzafrir_laptop | And most people never bother changing that |
| 20:31.50 | kpfleming | then let's change it |
| 20:31.55 | kpfleming | now's our chance to make it better |
| 20:31.55 | tzafrir_laptop | I want the "working default" to be only the required modules |
| 20:32.06 | kpfleming | that is what happens when you use kernel autoloading |
| 20:32.58 | *** join/#asterisk-dev mbwjr12 (n=kvirc@pal-179-077.itap.purdue.edu) |
| 20:32.59 | tzafrir_laptop | You want the default to be empty? |
| 20:33.21 | mbwjr12 | hey all, can any chan_mobile devs tell me which platform they develop on? distro, etc. |
| 20:33.59 | kpfleming | if the script can't provide any useful benefit without having something in MODULES, i see no point is listing all modules in MODULES |
| 20:34.18 | kpfleming | if we do that, then we don't even need the init script, we can just let kernel module autoloading take care of it |
| 20:34.24 | kpfleming | what distros ship with autoloading turned off? |
| 20:34.27 | tzafrir_laptop | BTW: there's a bigger issue on the mailing list :-) |
| 20:34.37 | moy | hello people ... any rule of thumb about when to lock dahdi_pvt in chan_dahdi? I see some places where members like mate, echocanon or subs are read and or written without the pvt lock ( I already read russellb post about locking order) |
| 20:34.41 | kpfleming | haven't gotten there yet |
| 20:39.28 | sruffell | kpfleming, tzafrir_laptop: so /etc/sysconfig/dahdi and /etc/default/dahdi are removed from the dahdi init script by design? Or should that be added back in to mimic the zaptel behavior? |
| 20:39.34 | russellb | moy: that's tough to answer. probably in a lot of places where it is not done now. |
| 20:39.50 | russellb | moy: you need to lock it if you need to ensure that another thread is not reading/writing to/from it at the same time ... |
| 20:40.05 | tzafrir_laptop | sruffell, replaced with /etc/dahdi/init.conf |
| 20:40.06 | russellb | moy: and that is more easily answered on a case by case basis |
| 20:40.13 | tzafrir_laptop | A single place to edit |
| 20:40.25 | sruffell | tzafrir_laptop: and so the MODULES should be defined there? |
| 20:40.31 | tzafrir_laptop | yes |
| 20:40.45 | moy | russellb, yeah, so it seems. Weeell, let's see how it goes :) ... thanks a lot |
| 20:41.21 | russellb | np, sorry i couldn't be more helpful |
| 20:41.22 | sruffell | ok...and btw...until we come up with a way to control the order of registration of drivers to dahdi....I don't think we can rely on autoloading. |
| 20:42.03 | kpfleming | i have built dozens of systems and never used the init script, but if that's what makes people happy, then i'm ok with it |
| 20:42.27 | kpfleming | but i'd much prefer that we never assume anything, and if the init script is going to be used, it *must* be configured to be told what modules to load and in which order |
| 20:43.02 | kpfleming | otherwise we might change the script in a future release to load in a different order (for a perfectly valid reason), and then users who upgrade are right back in the same situation |
| 20:43.03 | sruffell | unless you hand load the drivers, the init script and the module blacklist is the only way to prevent those cases where boards of different types register in different order....unless I'm missing something... |
| 20:43.29 | kpfleming | they always register in PCI bus enumeration order, which hasn't changed very often |
| 20:43.43 | kpfleming | at least in my experience |
| 20:43.48 | kpfleming | but i haven't used lots of different types of servers |
| 20:44.21 | sruffell | udev_start will start a process for each driver that is being registered....so while all the boards serviced by one driver will always load in the same order...that isn't true when you have different drivers (say wctdm24xxp and wcte12xp) |
| 20:44.46 | kpfleming | that's not kernel autoloading, as far as i know |
| 20:44.49 | kpfleming | that's udev autoloading |
| 20:44.53 | jsmith-teaching | sruffell: That's it! Let's roll *everything* into one driver :-p |
| 20:44.55 | kpfleming | kernel autoloading happens even before init gets started |
| 20:46.05 | sruffell | then I'm not aware of the difference. I'm under the impression that the kernel always needs a userspace helper...and on most modern systems, it's udev that scans the PCI bus and automatically loads the drivers. |
| 20:46.22 | kpfleming | maybe i'm mistaken |
| 20:46.40 | kpfleming | wouldn't be the first time |
| 20:47.32 | kpfleming | so how about this: install the init script, with an empty MODULES default, and also install a blacklist that blacklists all the hardware drivers by default |
| 20:47.35 | sruffell | start_udev is called early in rc.sysinit (at least on my Fedora Core 8 system) and that is what loads the modules for devices that aren't blacklisted. |
| 20:47.51 | kpfleming | then the user can choose the method they prefer... fill in MODULES, or remove/edit the blacklist |
| 20:48.23 | sruffell | If you did that, then you would always need to run the init script in order to get it to work...and never use the autoloader (but I think that would be best until we can control how the drivers register with dahdi) |
| 20:48.40 | kpfleming | until we can properly address oron's suggestion for span/channel naming (which i agree with 100% btw), module load order has to be deterministic |
| 20:49.00 | sruffell | ok....then that is what I'll do...thanks. |
| 20:50.01 | kpfleming | if we wanted to get really fancy, 'make config' could use some perl script and the DAHDI:: modules to find out what is on the system and prompt the user to select a module load order, then populate MODULES itself |
| 20:51.07 | tzafrir_laptop | the script: xpp/dahdi_drivers |
| 20:52.05 | tzafrir_laptop | The version there uses some @INC tricks to be able to run with the packaged modules without installing |
| 20:52.14 | kpfleming | right |
| 20:52.19 | tzafrir_laptop | The actual code there are the last two lines |
| 20:53.04 | sruffell | I like that idea. |
| 20:53.41 | tzafrir_laptop | debian and gentoo have facilities to load modules early, before hotplug. AFAIK RH/Fedora has none |
| 20:53.58 | kpfleming | but they respect the blacklist, right? |
| 20:54.18 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 20:54.18 | tzafrir_laptop | blacklist means a module won't be modprobed, right? |
| 20:54.30 | sruffell | blacklist means a module won't be loaded by it's pci alias. |
| 20:54.40 | kpfleming | yes, the blacklist is handled by modprobe itself |
| 20:54.43 | sruffell | you can still modprobe the driver even if it's in the v |
| 20:54.46 | sruffell | v -> blacklist |
| 20:54.47 | kpfleming | right |
| 20:54.53 | kpfleming | it will never be automatically loaded |
| 20:54.57 | kpfleming | only on explict request |
| 20:56.36 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 21:12.34 | *** join/#asterisk-dev RoyK (n=roy@ti211210a080-3097.bb.online.no) |
| 21:14.09 | snuff-home | btw kpfleming see my comment in 12932 ? |
| 21:14.23 | kpfleming | yes, working on it now |
| 21:14.28 | kpfleming | thanks for letting me know |
| 21:14.36 | snuff-home | :) |
| 21:15.05 | kpfleming | does the union hack actually make it work properly? it's ugly, that's for sure |
| 21:16.05 | snuff-home | from what i remembered it still worked.. but i still had other issues compiling |
| 21:16.31 | snuff-home | i swear the change was part of trunk for a little bit before astobj2 sip |
| 21:16.55 | kpfleming | could have been |
| 21:17.11 | kpfleming | i'll put it into aligner2 and you can test that compiling |
| 21:17.22 | snuff-home | k |
| 21:17.26 | kpfleming | i won't mess with chan_iax2 right now, as long as you can do your codec testing using SIP |
| 21:17.57 | snuff-home | mm i understand.. i only have chan_sip selected anyway.. |
| 21:18.31 | kpfleming | ok |
| 21:18.35 | kpfleming | will commit in a minute |
| 21:18.38 | snuff-home | we'll need to work through some other problems before it will compile after this one md5.c comes to mind |
| 21:19.06 | kpfleming | alright |
| 21:19.22 | kpfleming | i'm headed to a meeting and wont' be back online until tomorrow morning, but i'll do what i can |
| 21:19.27 | snuff-home | char[64] cast to int16[] |
| 21:19.37 | kpfleming | nice |
| 21:19.47 | kpfleming | well, that's pretty easy to work around |
| 21:20.03 | snuff-home | mm its ok.. i have other stuff to do actual work n all :) |
| 21:20.15 | kpfleming | yeah |
| 21:20.41 | snuff-home | have to play with pc at lunch break since i haven't got vpn details yet |
| 21:28.18 | kpfleming | tzafrir_laptop: i'm going to play with implementing some of oron's suggestions in the morning, to see how much work it would really be |
| 21:33.51 | *** part/#asterisk-dev Cresl1n (n=matt@216.207.245.1) |
| 21:36.36 | *** part/#asterisk-dev Rem| (n=remi@bg-fw2out.monmouth.com) |
| 22:24.15 | *** join/#asterisk-dev obnauticus (n=obnautic@c-67-160-181-107.hsd1.wa.comcast.net) |
| 22:31.44 | *** join/#asterisk-dev RoyK (n=roy@ti211210a080-3097.bb.online.no) |
| 22:31.59 | seanbright | it's quiet... |
| 22:33.47 | russellb | mhm |
| 22:34.10 | seanbright | busy day in alabama? |
| 22:34.13 | Qwell | jbot: say something random |
| 22:34.13 | jbot | something random |
| 22:36.43 | putnopvut | jbot: say Qwell's a nub |
| 22:36.44 | jbot | Qwell's a nub |
| 22:36.57 | Qwell | jbot: throw stuff at putnopvut |
| 22:36.57 | jbot | ACTION winds up and throws stuff at putnopvut over the moon. |
| 22:37.06 | putnopvut | lol |
| 22:37.23 | putnopvut | there are probably more optimal trajectories for the stuff. |
| 22:37.39 | seanbright | is anthm the freeswitch guy? |
| 22:37.54 | putnopvut | yes sir. |
| 22:38.02 | seanbright | ahh |
| 22:38.04 | seanbright | cool |
| 22:38.35 | Qwell | I think...it is meant... |
| 22:38.38 | Qwell | jbot: throw putnopvut |
| 22:38.38 | jbot | ACTION winds up and throws putnopvut over the moon. |
| 22:38.42 | Qwell | to be used thusly |
| 22:38.46 | putnopvut | right |
| 22:38.58 | putnopvut | jbot: throw the moon over the moon |
| 22:38.59 | jbot | ACTION winds up and throws the moon over the moon over the moon. |
| 22:39.39 | seanbright | jbot: fist Qwell |
| 22:39.40 | jbot | ACTION uses Qwell as a handpuppet |
| 22:39.45 | seanbright | weeeeeee |
| 22:40.37 | putnopvut | jbot: hello |
| 22:40.37 | jbot | Howdy Bub |
| 22:41.26 | Corydon76-lap | jbot: throw exception |
| 22:41.27 | jbot | ACTION winds up and throws exception over the moon. |
| 22:46.05 | *** join/#asterisk-dev DarkRift (n=dark@bas10-montreal02-1177583238.dsl.bell.ca) |
| 23:31.56 | *** part/#asterisk-dev Deeewayne (n=Deeewayn@216.207.245.1) |
| 23:53.42 | *** join/#asterisk-dev RoyK (n=roy@ti211210a080-3097.bb.online.no) |