| 00:14.49 | *** join/#asterisk-dev jtodd (i=fn1y88n9@asterisk/community-director-and-tie-dye-shirt-lover/jtodd) |
| 00:14.49 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 00:48.21 | *** join/#asterisk-dev Deeewayne (n=dwayne@c-71-228-179-90.hsd1.al.comcast.net) |
| 00:48.21 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 01:33.14 | *** join/#asterisk-dev ebroad (n=eb@pool-173-77-203-142.nycmny.fios.verizon.net) |
| 02:01.19 | *** join/#asterisk-dev jtodd (i=rbw3mnza@asterisk/community-director-and-tie-dye-shirt-lover/jtodd) |
| 02:01.19 | *** mode/#asterisk-dev [+o jtodd] by ChanServ |
| 02:15.48 | *** join/#asterisk-dev moy (n=moy@CPE001cdfec4cee-CM00080dab8485.cpe.net.cable.rogers.com) |
| 02:57.06 | *** join/#asterisk-dev ZPertee (n=ZPertee@rrcs-74-219-221-11.central.biz.rr.com) |
| 03:17.10 | *** join/#asterisk-dev pyite (n=pyite@unaffiliated/pyite) |
| 03:39.41 | *** part/#asterisk-dev ZPertee (n=ZPertee@rrcs-74-219-221-11.central.biz.rr.com) |
| 04:35.50 | ebroad | mvanbaak, ping |
| 04:40.03 | *** join/#asterisk-dev JunK-Y (n=junky@modemcable206.69-20-96.mc.videotron.ca) |
| 05:33.30 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@212.179.75.202) |
| 05:33.30 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 06:00.54 | *** join/#asterisk-dev |Cybex| (n=John@atwork-26.r-212.178.82.atwork.nl) |
| 06:20.39 | mvanbaak | just missed them |
| 07:19.38 | *** join/#asterisk-dev tamiel (n=tamiel@213.30.183.226) |
| 07:36.44 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@local.xorcom.com) |
| 07:36.44 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 08:48.10 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 08:49.45 | *** join/#asterisk-dev fiddur (n=fiddur@dhcp08.textalk.com) |
| 09:03.28 | atis_work | oej's reccomendation on using "bugs fixed" to measure stability is so cool :p |
| 09:04.06 | atis_work | 100-200: new release, 200-500: somewhat stable, 500-1000: stable :D |
| 09:04.55 | *** join/#asterisk-dev kazaa_lite (n=msaleem@94-193-98-124.zone7.bethere.co.uk) |
| 11:55.31 | *** join/#asterisk-dev leifmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 11:55.31 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 12:07.54 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:09.27 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:09.27 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:34.49 | *** join/#asterisk-dev oej (n=olle@1x-193-157-197-57.uio.no) |
| 12:34.49 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 12:45.21 | *** join/#asterisk-dev oej (n=olle@1x-193-157-197-57.uio.no) |
| 12:45.21 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 12:49.51 | *** join/#asterisk-dev PoWeRKiLL (n=powerkil@bzq-79-181-102-215.red.bezeqint.net) |
| 12:50.48 | seanbright | leifmadsen: you got a google wave invite? you son of a bitch |
| 12:51.02 | leifmadsen | seanbright: I did :) it sucks |
| 12:51.09 | leifmadsen | I don't know what people are even going to do with it |
| 12:51.22 | seanbright | it sucks? |
| 12:51.29 | seanbright | eff you |
| 12:51.31 | seanbright | heh |
| 12:51.58 | seanbright | well when you get more invites i expect one |
| 12:52.02 | seanbright | or you're off the christmas list |
| 12:52.57 | leifmadsen | I was on the list? cool :) |
| 12:53.09 | seanbright | well it can't hurt your chances... that's all i'm saying |
| 12:53.11 | seanbright | :) |
| 13:02.33 | leifmadsen | haha |
| 13:51.25 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 14:01.44 | *** join/#asterisk-dev Deeewayne (n=dwayne@75.76.254.162) |
| 14:01.44 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 14:07.07 | tzafrir_laptop | Deeewayne, here? |
| 14:08.00 | tzafrir_laptop | or, FWIW, anyboy else: the topic of #asterisk needs a bit of updating: |
| 14:08.15 | tzafrir_laptop | versions of asterisk, asterisk-addons and libpri |
| 14:09.23 | Deeewayne | tzafrir_laptop, I'm here |
| 14:09.55 | russellb | leifmadsen: ^^^ |
| 14:10.16 | leifmadsen | tzafrir_laptop: to what? |
| 14:10.38 | leifmadsen | did Matt publish new libpri's again without letting me know? :) |
| 14:10.38 | Deeewayne | leifmadsen, "happy unbirthday file" |
| 14:10.42 | tzafrir_laptop | libpri: 1.4.10.2 |
| 14:11.30 | tzafrir_laptop | oh, right. all asterisk "releases" are RCs |
| 14:11.45 | leifmadsen | tzafrir_laptop: yes, we don't change the topic for RCs there |
| 14:11.48 | leifmadsen | so I think those are right |
| 14:26.43 | *** part/#asterisk-dev |Cybex| (n=John@atwork-26.r-212.178.82.atwork.nl) |
| 14:28.29 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 14:28.29 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 14:29.46 | *** join/#asterisk-dev putnopvut (n=putnopvu@asterisk/master-of-queues/mmichelson) |
| 14:29.46 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:29.52 | *** join/#asterisk-dev sruffell (n=sruffell@asterisk/the-kernel-guy/sruffell) |
| 14:29.52 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:33.04 | *** join/#asterisk-dev hugorebelo (n=hugorebe@200-171-132-124.completo.com.br) |
| 14:36.43 | *** join/#asterisk-dev moy (n=moy@74.12.134.3) |
| 14:53.44 | *** join/#asterisk-dev |Cybex| (n=John@80.100.126.176) |
| 14:58.06 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 14:58.06 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:04.04 | *** join/#asterisk-dev eliel (n=eliels@200.61.172.61) |
| 15:04.04 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 15:09.32 | *** join/#asterisk-dev ebroad (n=EB@72.11.213.195) |
| 15:11.41 | *** join/#asterisk-dev anonymouz666 (n=anonymou@187-28-37-118.poolip.RJO.embratel.net.br) |
| 15:16.56 | ebroad | mvanbaak, do you have a minute or two to spare? |
| 15:17.04 | ebroad | chan_skinny question |
| 15:20.50 | mvanbaak | sure |
| 15:20.52 | mvanbaak | shoot |
| 15:20.58 | anonymouz666 | question: In func_volume we have checks for DTMFs * and # to increase/decrease volume... but how to use that...? How do I trigger these DTMFs? |
| 15:21.15 | mvanbaak | just hit those keys on the ohene |
| 15:21.23 | mvanbaak | s/ohene/phone/ |
| 15:21.35 | anonymouz666 | mvanbaak: you use it? |
| 15:21.41 | mvanbaak | no |
| 15:21.44 | anonymouz666 | doesn't seem to work |
| 15:21.49 | anonymouz666 | only by dialplan |
| 15:28.55 | anonymouz666 | mvanbaak: put ast_log() inside the function to increase the volume and nothing is printed...so the function is really not being called |
| 15:38.55 | ebroad | mvanbaak, chan_skinny, line 4344, shouldn't we only be calling ast_rtp_instance_new_source() if we are remotely bridging |
| 15:39.01 | ebroad | ? |
| 15:40.33 | mvanbaak | nope |
| 15:40.43 | mvanbaak | on a SRCUPDATE we should call it as well |
| 15:41.12 | *** join/#asterisk-dev CunningPike (n=CunningP@204.239.8.157) |
| 15:41.16 | mvanbaak | (or the same bug is present in all rtp using modules) |
| 15:41.36 | ebroad | hmm, im getting a segfault on any inbound call from a sip device to a skinny devic |
| 15:41.48 | ebroad | instance->engine->newsource is empty |
| 15:54.11 | mvanbaak | meh |
| 15:54.29 | mvanbaak | I have no problem getting and answering calls that come in over IAX2 |
| 15:54.46 | ebroad | hehe, pigs will fly when cisco supports iax2 |
| 15:58.35 | ebroad | the issue only seems to be local, i.e. calls from a trunk or from a softphone on the outside(nat=yes) work fine. it looks like an issue with directmedia |
| 15:58.50 | mvanbaak | ITSP <-- IAX2 --> asterisk <-- skinny --> 7960 |
| 15:59.24 | ebroad | cheap cordless --> spa-2102 --> asterisk --> 7940(skinny) |
| 15:59.28 | ebroad | segfault |
| 15:59.51 | ebroad | ill have to wait until i get home to get a backtrace |
| 16:00.08 | ebroad | or try to get my wife to try it |
| 16:03.01 | Corydon76-dig | You don't have ssh access? |
| 16:03.11 | ebroad | i don't think the 7940's can do ssh |
| 16:04.09 | ebroad | i have access to my * box, but i need a phone on the local segment to trigger this |
| 16:05.26 | Corydon76-dig | Dunno, if my production boxes segfault, they always leave behind good cores |
| 16:05.51 | ebroad | i trashed my cores yesterday while i was doing some testing |
| 16:22.35 | mvanbaak | ebroad: I never tested that. I dont have any sip devices here at home |
| 16:22.56 | ebroad | ok |
| 16:23.00 | mvanbaak | well, not connected to my asterisk. I have a couple but they are all connected to my work pbx |
| 16:23.28 | ebroad | im going to see what i can do from the 7940 web interface |
| 16:23.54 | ebroad | the sip firmware has telnet and a test feature which will simulate key press |
| 16:26.41 | mvanbaak | reprovision the 7940 to make it autoanswer your calls |
| 16:26.59 | ebroad | rue |
| 16:27.01 | ebroad | *true |
| 16:27.25 | ebroad | i was also getting a segfault when directmedia=yes in skinny.conf for the line |
| 16:27.29 | ebroad | on outbound calls only |
| 16:40.15 | seanbright | why does no one sell the power supply with the phone (7960) on ebay? |
| 16:40.22 | ebroad | poe |
| 16:40.28 | seanbright | ah |
| 16:41.09 | ebroad | if you have $200 or so to spare, get a catalyst 3550 pwr |
| 16:41.23 | seanbright | http://tinyurl.com/yetz7zp <- not the 7960 i was looking for |
| 16:41.44 | ebroad | lol |
| 16:53.21 | *** join/#asterisk-dev jetsnoc (n=brian@72.22.242.66) |
| 16:54.11 | *** join/#asterisk-dev shido6 (n=shido6@74-132-202-71.dhcp.insightbb.com) |
| 17:07.17 | mvanbaak | maybe it's the new design for the cisco phones |
| 17:10.01 | seanbright | heh |
| 17:11.08 | *** join/#asterisk-dev errotan (n=errotan@62.201.122.189) |
| 17:16.34 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 17:27.43 | ebroad | mvanbaak, do you know how to configure auto answer via the SEPmac.cnf.xml file? |
| 17:34.31 | *** join/#asterisk-dev |Cybex| (n=John@80.100.126.176) |
| 17:44.13 | *** join/#asterisk-dev Praise (n=Fat@unaffiliated/praise) |
| 18:17.02 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:26.42 | *** join/#asterisk-dev Victor_Yure (n=victor@unaffiliated/victoryure/x-837844) |
| 18:59.33 | ebroad | mvanbaak, ping |
| 19:00.32 | file | make |
| 19:00.42 | file | this is not my terminal window |
| 19:01.07 | *** join/#asterisk-dev Deeewayne (n=dwayne@75.76.254.162) |
| 19:01.07 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 19:01.09 | Marquis42 | file: Look on the bright side, you didn't think it was a window asking for a password. I've done that. :( |
| 19:01.29 | file | 1234! |
| 19:01.38 | Marquis42 | lol |
| 19:02.04 | ebroad | Marquis42, i think we all have, SANS ISC actually had a poll some time back asking if those passwords were ever used |
| 19:02.43 | Marquis42 | ebroad: Oh yeah? At least I'm not alone. :) |
| 19:03.32 | ebroad | Marquis42, http://isc.sans.org/poll.html?pollid=215&results=Y |
| 19:06.06 | jsmith | How do we know that file's password isn't "make"? |
| 19:06.09 | jsmith | ducks |
| 19:07.20 | Marquis42 | I believe a 'zing!' is in order. ;) |
| 19:08.10 | file | my password is... |
| 19:08.12 | file | not short. |
| 19:17.22 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 19:23.13 | Corydon76-dig | gains root to file's box with "not short" |
| 19:28.04 | mvanbaak | ebroad: pong |
| 19:28.11 | ebroad | got a backtrace |
| 19:28.25 | ebroad | i used xml push to dial |
| 19:28.29 | ebroad | its pretty cool |
| 19:29.24 | ebroad | mvanbaak, http://www.pastebin.ca/1604907, this is with directmedia=yes under the line config in skinny.conf |
| 19:29.32 | ebroad | dialing a sip peer |
| 19:30.39 | ebroad | in ast_rtp_instance_early_bridge_make_compatible() it looks like glue1->get_codec is null |
| 19:31.41 | ebroad | adding an ast_verbose(codec is %d, glue1->get_codec(c1)) will cause it to segfault on the ast_verbose() statment |
| 19:36.58 | atis_home | russellb: 15976 |
| 19:37.06 | russellb | M15976 |
| 19:37.09 | MuffinMan | [closed] [Asterisk] Resources/res_musiconhold 0015976: abort in filestream_destructor / ast_filestream_frame_freed / moh_files_generator reported by atis (Karma: +18.00) https://issues.asterisk.org/view.php?id=15976 |
| 19:37.13 | atis_home | that's why i started to test that patch |
| 19:37.22 | russellb | right |
| 19:37.42 | atis_home | however that crash somehow stopped, i'm unsure why |
| 19:37.53 | russellb | magic, i guess |
| 19:37.59 | atis_home | it couldb be that i removed res_timing_pthread or stopped using core show channels for debug |
| 19:39.15 | atis_home | i just tested for few times, got few backtraces as i uploaded in 15609 and reverted |
| 19:39.55 | atis_home | but i still ocassionally get Exceptionally long... |
| 19:39.59 | atis_home | if it's related at all |
| 19:41.29 | russellb | right, a number of people are seeing that |
| 19:41.32 | russellb | i don't think it's related |
| 19:41.35 | russellb | i need to deal with that next |
| 19:41.59 | atis_home | is that something serious? possible audio problems? |
| 19:42.35 | atis_home | i just have scheduled 1.6.1 release going production saturday :p |
| 19:47.11 | russellb | i don't know yet |
| 19:52.27 | atis_home | uhh, you commited it.. i guess i have to test it overnight |
| 19:53.45 | russellb | i think that crash you saw is the sign of a different problem, if you can reproduce it on the latest code so that I can follow line numbers and such, I'll take a look |
| 19:54.44 | atis_home | russellb: isn't it also going to 1.6? |
| 19:54.46 | russellb | these bugs are driving me nuts |
| 19:54.47 | russellb | nods |
| 19:56.03 | atis_home | perhaps next week.. this week i'm trying to get my 1.6.1 perfectly stable :) |
| 19:56.16 | atis_home | not to make it crash with your patches :D |
| 19:57.06 | russellb | ha |
| 19:57.08 | russellb | k. |
| 20:03.41 | *** part/#asterisk-dev shido6 (n=shido6@74-132-202-71.dhcp.insightbb.com) |
| 20:07.46 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-218-155-147.cablep.bezeqint.net) |
| 20:07.46 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 20:09.00 | atis_home | russellb: oh, could the cause of crash be that ast_closestream in 1.6.1.6 doesn't have AST_SCHED_DEL? |
| 20:09.37 | russellb | probably not |
| 20:09.39 | atis_home | what you commited just didn't merge with 1.6.1.6 |
| 20:09.48 | russellb | that would be why it didn't merge, anyway, yeah |
| 20:10.28 | atis_home | well, i'll leave that patch for testing later on vanilla trunk |
| 20:11.14 | atis_home | now i already have 126kB of patches running :) |
| 20:16.25 | tzafrir_laptop | would appreciate a response of some sort to http://lists.digium.com/pipermail/asterisk-dev/2009-October/040067.html |
| 20:17.33 | kpfleming | i doubt you will get one, and you should probably just proceed. the people who are most likely to be able to analyze that post and respond are involved in other projects that are higher priority for them, unfortunately. |
| 20:18.55 | *** join/#asterisk-dev leifmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 20:18.55 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 20:19.36 | ebroad | The_Boy_Wonder, ping |
| 20:21.19 | russellb | ebroad: i pinged first!!! |
| 20:21.23 | russellb | (on another network :-p) |
| 20:21.48 | ebroad | russellb, hehe, i wanted to ask dave about http://lists.digium.com/pipermail/asterisk-dev/2009-October/040069.html |
| 20:22.04 | ebroad | 4 bytes is 4 bytes :) |
| 20:22.15 | russellb | nods |
| 20:27.04 | *** part/#asterisk-dev |Cybex| (n=John@80.100.126.176) |
| 20:37.03 | ebroad | leifmadsen, ping |
| 20:37.12 | leifmadsen | ebroad: pong |
| 20:37.35 | ebroad | whats the url of the mantis workflow |
| 20:37.51 | leifmadsen | http://www.asterisk.org/developers/doxygen |
| 20:37.58 | ebroad | thanx |
| 20:38.04 | The_Boy_Wonder | ebroad: hey |
| 20:38.07 | leifmadsen | oops |
| 20:38.09 | leifmadsen | ebroad: http://www.asterisk.org/doxygen/trunk/ |
| 20:38.22 | leifmadsen | the wordflow is under Related Pages |
| 20:38.43 | ebroad | leifmadsen, thanx i got it |
| 20:38.44 | leifmadsen | 11th last in the list (near the bottom) |
| 20:38.50 | leifmadsen | coolio! |
| 20:38.55 | ebroad | The_Boy_Wonder, http://lists.digium.com/pipermail/asterisk-dev/2009-October/040069.html |
| 20:39.14 | ebroad | should we save 4 bytes |
| 20:39.16 | leifmadsen | is reading about the erlang model... weee! |
| 20:39.31 | The_Boy_Wonder | ha, sure |
| 20:39.40 | The_Boy_Wonder | its just in one branch |
| 20:40.08 | ebroad | true, i guess its technically a feature as well |
| 20:40.31 | ebroad | though it should backport nicely to 1.6.1 and maybe 1.6.0 |
| 20:40.35 | The_Boy_Wonder | yep, i added it into the CHANGES log for 1.6.3 |
| 20:40.40 | ebroad | i saw that |
| 20:41.09 | The_Boy_Wonder | it's not very invasive, so you have an argument there |
| 20:41.52 | ebroad | true, and it does correct some broken functionality |
| 20:42.05 | ebroad | leifmadsen, i think ill give marshaling a shot |
| 20:42.10 | leifmadsen | ebroad: yay! |
| 20:42.29 | leifmadsen | I will get you "the power" |
| 20:42.37 | The_Boy_Wonder | oh nice! |
| 20:42.39 | leifmadsen | first... you get the money |
| 20:42.56 | ebroad | leifmadsen, given the amount of mantis entries, you guys could use some help |
| 20:43.02 | leifmadsen | :D |
| 20:43.04 | ebroad | leifmadsen, ah yes the money |
| 20:43.14 | The_Boy_Wonder | wait, money!?! |
| 20:43.27 | leifmadsen | The_Boy_Wonder: yes, I'm charging for privileges now |
| 20:43.31 | The_Boy_Wonder | LOL |
| 20:43.58 | leifmadsen | 1 penny of every bug opened goes to children in Africa |
| 20:43.58 | ebroad | leifmadsen, $25 per patch |
| 20:44.24 | ebroad | i think we have a deal |
| 20:44.56 | leifmadsen | ebroad: access updated! |
| 20:45.00 | leifmadsen | let me know when/if you have any questions |
| 20:45.05 | ebroad | leifmadsen, much appreciated |
| 20:45.07 | ebroad | oh i will |
| 20:45.10 | leifmadsen | ebroad: thanks! |
| 20:45.26 | ebroad | "Logged in as: ebroad (Elazar - manager)" |
| 20:45.30 | ebroad | sweeet! |
| 20:45.37 | The_Boy_Wonder | oh SNAP |
| 20:47.20 | Corydon76-dig | thought an ebroad was a virtual girlfrined |
| 20:47.52 | ebroad | Corydon76-dig, the broad jokes are very old ;) |
| 20:48.18 | ebroad | though i seem to have a lot of street namesakes |
| 20:48.27 | Corydon76-dig | I'll be here all week. Try the veal |
| 20:50.13 | ebroad | who handles chan_spy |
| 20:50.17 | ebroad | generally |
| 20:50.22 | Corydon76-dig | Anybody |
| 20:50.40 | Corydon76-dig | Well, file, but file's time is already full |
| 20:50.59 | file | full of implosions! |
| 20:51.05 | ebroad | M16045 |
| 20:51.22 | file | raises eyebrow |
| 20:51.25 | MuffinMan | [new] [Asterisk] Applications/app_chanspy 0016045: crash when spying - was working fine in beta2 reported by bluecrow76 (Karma: +10.00) https://issues.asterisk.org/view.php?id=16045 |
| 20:51.34 | seanbright | ugh |
| 20:51.35 | ebroad | there we go |
| 20:51.39 | seanbright | effing chanspy |
| 20:51.41 | seanbright | just remove it |
| 20:51.57 | ebroad | double free |
| 20:52.04 | Corydon76-dig | only if I can remove chan_sip |
| 20:52.18 | ebroad | if you write chan_sip2 |
| 20:52.46 | Corydon76-dig | ebroad: do you do anything with databases? |
| 20:53.26 | ebroad | Corydon76-dig, i know a thing or two, i work for a medium size shop with a small it dept, so i have to be jack of all trades(master of none) |
| 20:53.58 | Corydon76-dig | ebroad: do you use database transactions? |
| 20:54.11 | ebroad | what platform? |
| 20:54.18 | Corydon76-dig | Any |
| 20:54.29 | ebroad | yes, on MS SQL |
| 20:54.45 | ebroad | basic transactioning though, commit and rollback |
| 20:55.24 | Corydon76-dig | Okay, there's an ability to use transactions in 1.6.2 with func_odbc |
| 20:56.09 | Corydon76-dig | This was a request... it's done, but I'm not really sure that the syntax and description make sense or are helpful |
| 20:56.13 | ebroad | within the dial plan? |
| 20:56.30 | Corydon76-dig | Right |
| 20:56.47 | ebroad | isnt that a little heavy for something that is somewhat time sensitive |
| 20:57.17 | Corydon76-dig | So what I really need is somebody who does that kind of thing to look at the docs and ask the kind of questions that somebody would ask when they're trying to get this to work, so we can be proactive and answer those in the documentation first |
| 20:58.20 | ebroad | architecturally its simple though |
| 20:58.36 | Corydon76-dig | I've personally always managed without transactions, even accounting systems... |
| 20:58.52 | ebroad | ie. insert into table (column) value (value) |
| 20:58.56 | Corydon76-dig | but some people wanted us to support them, so... |
| 20:59.01 | ebroad | if everything goes fine, commit |
| 20:59.08 | ebroad | if something blows up, roll back |
| 20:59.27 | Corydon76-dig | Right, that's the idea |
| 20:59.56 | Corydon76-dig | So the questions are twofold... what haven't I documented well, and am I missing anything critical? |
| 21:01.04 | ebroad | forgive my ignorance, where is the func_odbc doc |
| 21:01.07 | Corydon76-dig | So if you looked at the ODBC() dialplan function and you could figure out how to use it with transactions, then all is good |
| 21:01.24 | Corydon76-dig | well, there's func_odbc.conf.sample |
| 21:01.47 | ebroad | im wondering what they are putting in the database at say the beginning of the call that cant be put when the call is done |
| 21:01.52 | Corydon76-dig | but that only specifies queries... the dialplan is where the transactions are built |
| 21:02.34 | ebroad | the only real reason for transactions is if you believe something might blow up in the middkle |
| 21:02.49 | Corydon76-dig | That's an instance question and I like to leave those to PBX admins |
| 21:03.22 | ebroad | right, but if something is going to blow up in the middle of a call, then you have bigger problems |
| 21:04.01 | Corydon76-dig | Heh... the argument you have is, "why would you ever need transactions in the dialplan?" |
| 21:04.14 | ebroad | exactly |
| 21:04.20 | ebroad | i want to know why |
| 21:04.24 | ebroad | that aside |
| 21:04.25 | Corydon76-dig | I personally don't, and I guess you don't either |
| 21:04.28 | ebroad | ill check out the docs |
| 21:04.50 | Corydon76-dig | Thank you |
| 21:06.18 | leifmadsen | perhaps thats a good question to ask on the -dev list. Sometimes people request functionality when they themselves don't quite understand why they wouldneed it |
| 21:06.28 | ebroad | true |
| 21:07.03 | leifmadsen | if we had a usage example, that typically makes things simpler to understand why something is useful, and how best it should be executed |
| 21:07.31 | ebroad | Corydon76-dig, is there an entry in mantis? |
| 21:07.50 | Corydon76-dig | ebroad: No, because it all resides in res_odbc |
| 21:07.58 | ebroad | ah |
| 21:12.03 | ebroad | documentation looks good to me, defaulting forcecommit to yes would be backwards compatible which is good |
| 21:12.28 | ebroad | when transactioning is not in use, queries are usually commited immediately |
| 21:12.34 | ebroad | depending on the server config |
| 21:12.46 | ebroad | so you can pull the ? mark |
| 21:13.14 | Corydon76-dig | The ? mark? |
| 21:13.25 | ebroad | Default to committing uncommitted transactions? |
| 21:13.41 | ebroad | unless that was an unintended typo |
| 21:13.49 | Corydon76-dig | ebroad: that's a question for the admin to answer |
| 21:13.53 | ebroad | ah |
| 21:14.20 | ebroad | right, but the average user will expect odbc to commit right away |
| 21:14.45 | Corydon76-dig | No, that's not the autocommit flag |
| 21:14.57 | ebroad | ah |
| 21:14.57 | ebroad | ok |
| 21:14.58 | Corydon76-dig | That's what happens to transactions that aren't explicitly committed |
| 21:15.11 | ebroad | then i would default it to no |
| 21:15.15 | Corydon76-dig | Right |
| 21:16.18 | Corydon76-dig | Okay, so I need to clarify that it's NOT the autocommit flag |
| 21:16.22 | ebroad | yes |
| 21:16.57 | Corydon76-dig | Basically, any query can happen either within or outside of a transaction... it's the other commands before it which make that determination |
| 21:17.21 | Corydon76-dig | If you start a transaction before a query, and you're using the same DB handle, then the query happens within the transaction |
| 21:17.37 | Corydon76-dig | If not, it's autocommitted right away |
| 21:17.59 | ebroad | on ms sql anything within the begin transaction block will be transactioned |
| 21:18.21 | Corydon76-dig | Right, until you commit or rollback, and that's the same here |
| 21:18.27 | ebroad | yes] |
| 21:18.34 | ebroad | that is correct |
| 21:19.00 | ebroad | begin transaction is sql92 i believe |
| 21:19.32 | ebroad | so all rdbms will never transaction unless you specify it |
| 21:21.28 | Corydon76-dig | In our case, we use the ODBC() function to start a transaction. |
| 21:21.35 | jsmith | ebroad: Yes, that's my understanding |
| 21:21.55 | jsmith | ebroad: Most RDBMS systems will *only* do transactions if you tell them to |
| 21:21.55 | Corydon76-dig | Set(ODBC(transaction,<conn-name>)=tx-id) |
| 21:22.24 | Corydon76-dig | And then you use "tx-id" for the commit |
| 21:23.26 | Corydon76-dig | The idea is that you could potentially run parallel transactions, in a single dialplan, if you really wanted to, even on the same database (using different connection handles, though) |
| 21:23.49 | Corydon76-dig | err, in a single call, that is. |
| 21:24.06 | Corydon76-dig | Why you would WANT to, OTOH, isn't really addressed. Just that you can. |
| 21:24.36 | ebroad | right, but begin transaction will lock the table |
| 21:24.39 | ebroad | or query |
| 21:24.43 | ebroad | i.e. |
| 21:24.58 | ebroad | odbc begin transaction |
| 21:25.04 | ebroad | dial(somedude) |
| 21:25.09 | ebroad | commit |
| 21:25.16 | Corydon76-dig | There's locks involved, but what they are is backend-defined |
| 21:25.27 | ebroad | if my call with somedude is 20 minutes long |
| 21:25.37 | ebroad | then the table will be locked for 20 minutes |
| 21:25.52 | Corydon76-dig | I don't think that's necessarily true |
| 21:26.32 | Corydon76-dig | What could happen in the meantime, though, is that somebody else inserts a record that conflicts with your insert, and the commit could fail |
| 21:26.49 | Corydon76-dig | but that all depends upon the level of isolation |
| 21:27.05 | ebroad | i think an insert or update would fail |
| 21:27.14 | ebroad | always |
| 21:27.23 | ebroad | what select returns depends on isolation |
| 21:27.42 | Corydon76-dig | Then MS SQL Server does some things differently |
| 21:27.45 | ebroad | you know, i could see this being used in a realtime billing scenario |
| 21:28.02 | Corydon76-dig | When I worked with Informix, you could do this, and it would never lock the whole table |
| 21:28.38 | Corydon76-dig | just the longer that you held that transaction, the more possible that some other query could conflict |
| 21:29.00 | Corydon76-dig | whoever commits first is who takes precedence |
| 21:29.07 | ebroad | correction, ms sql will lock the row |
| 21:29.25 | ebroad | ms sql doesnt do that |
| 21:29.43 | ebroad | i.e. if balance = $8 |
| 21:29.53 | ebroad | i lock, update balance to 10 |
| 21:30.06 | Corydon76-dig | Well, no explicit lock, though |
| 21:30.21 | Corydon76-dig | There ARE locks in Informix, but you need to explicitly request them |
| 21:30.31 | ebroad | you attempt to update as well, depending on isolation, your update will fail, since balance = 8 when you started |
| 21:30.48 | ebroad | or your update will commit after mine |
| 21:31.26 | jsmith | Many modern RDBMSs use MVCC |
| 21:32.08 | ebroad | which makes sense |
| 21:33.46 | ebroad | apparently sql 2005 and later supports mvcc |
| 21:33.50 | ebroad | http://en.wikipedia.org/wiki/Multiversion_concurrency_control#Databases_with_MVCC |
| 21:34.15 | ebroad | at that point transactioning is moot |
| 21:35.30 | ebroad | uhg, i don't like skinny |
| 21:47.28 | *** join/#asterisk-dev telecos (n=sergio@198.167.219.87.dynamic.jazztel.es) |
| 21:48.27 | *** join/#asterisk-dev jsmith (n=jsmith@asterisk/training-and-documentation-guru/jsmith) |
| 21:48.27 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 21:53.57 | *** join/#asterisk-dev denon (i=denon@sassinak.net) |
| 21:54.16 | *** part/#asterisk-dev ebroad (n=EB@72.11.213.195) |
| 21:55.32 | *** join/#asterisk-dev ebroad (n=EB@72.11.213.195) |
| 22:15.45 | *** join/#asterisk-dev Victor_Yure (n=victor@unaffiliated/victoryure/x-837844) |
| 22:30.31 | kazaa_lite | are there some restrictions in calling sleep from a codec? |
| 22:30.51 | kazaa_lite | i call sleep and my codec hangs up |
| 22:31.17 | Corydon76-dig | You shouldn't ever pause within a codec |
| 22:31.30 | atis_home | russellb: you still there? |
| 22:31.43 | Corydon76-dig | If you need to wait, return a null_frame |
| 22:32.19 | kazaa_lite | ummm.... right |
| 22:32.50 | kpfleming | yeah, codec modules cannot block or sleep |
| 22:33.50 | Corydon76-dig | and that's an AST_FRAME_NULL, not NULL |
| 22:34.06 | russellb | atis_home: yup |
| 22:34.10 | kpfleming | actually, there's a global ast_null_frame somewhere |
| 22:34.38 | Corydon76-dig | kpfleming: only in some versions, I think |
| 22:34.47 | atis_home | russellb: i can give you access to our testing server, which constantly has ~120 testing calls for now |
| 22:35.26 | Corydon76-dig | Ah, 1.4 and forward |
| 22:35.28 | russellb | atis_home: and shows this issue? |
| 22:35.57 | russellb | atis_home: aragon said something about usage of "show channel" in his setup. Are you using that, too? |
| 22:35.58 | atis_home | i'm right now grepping today's log, that message shows up several times per hour |
| 22:36.07 | atis_home | russellb: not anymore |
| 22:36.18 | russellb | any sort of CLI or manager activity? |
| 22:36.42 | atis_home | very low manager, - pause/unpause, and some CLI |
| 22:37.10 | atis_home | CLI uses only "core show locks" and "core show uptime" |
| 22:37.15 | russellb | okay. |
| 22:37.58 | russellb | you say several times an hour, roughly how many is several? |
| 22:41.05 | *** join/#asterisk-dev leifmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 22:41.05 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 22:41.38 | atis_home | well, it's actually every 2-20 minutes |
| 22:41.47 | atis_home | and if they start coming up, they repeat often |
| 22:41.55 | atis_home | i guess, it depends on load |
| 23:31.46 | *** join/#asterisk-dev jmls (n=jmls@host217-36-208-155.in-addr.btopenworld.com) |