IRC log for #asterisk-bugs on 20071030

01:44.13*** join/#asterisk-bugs ivanfm (n=ivanfm@c906b486.virtua.com.br)
01:57.18*** join/#asterisk-bugs blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage)
01:57.18*** mode/#asterisk-bugs [+o blitzrage] by ChanServ
07:00.04*** join/#asterisk-bugs dlynes_laptop (n=dlynes@d154-20-34-39.bchsia.telus.net)
07:47.01*** join/#asterisk-bugs agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it)
08:59.50*** join/#asterisk-bugs dlynes_laptop (n=dlynes@d154-20-34-39.bchsia.telus.net)
10:55.10*** join/#asterisk-bugs caio1982 (i=caio1982@CAcert-br/caio1982)
12:05.13*** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
12:05.13*** mode/#asterisk-bugs [+o russellb] by ChanServ
12:29.51*** join/#asterisk-bugs blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage)
12:29.51*** mode/#asterisk-bugs [+o blitzrage] by ChanServ
13:21.51*** join/#asterisk-bugs oej (n=olle@64.251.114.2)
13:55.29*** join/#asterisk-bugs oej (n=olle@64.251.114.2)
14:13.41*** join/#asterisk-bugs jcmoore (n=jcmoore@unaffiliated/tgrman)
14:42.34*** join/#asterisk-bugs russellb_ (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
14:42.34*** mode/#asterisk-bugs [+o russellb_] by ChanServ
14:42.49*** join/#asterisk-bugs oej (n=olle@64.251.114.2)
14:56.10*** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
14:56.10*** mode/#asterisk-bugs [+o russellb] by ChanServ
14:57.28*** join/#asterisk-bugs putnopvut (i=putnopvu@nat/digium/x-ea38b3acd740874a)
14:57.28*** mode/#asterisk-bugs [+o putnopvut] by ChanServ
15:03.47*** join/#asterisk-bugs russellb_ (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
15:03.47*** mode/#asterisk-bugs [+o russellb_] by ChanServ
15:14.26*** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
15:14.26*** mode/#asterisk-bugs [+o russellb] by ChanServ
15:22.46*** join/#asterisk-bugs oej (n=olle@64.251.114.2)
15:27.40*** join/#asterisk-bugs michael-i (n=michael-@141.41.40.55)
15:45.30*** join/#asterisk-bugs putnopvut (i=putnopvu@nat/digium/x-cd3b7f47687669a5)
15:45.30*** mode/#asterisk-bugs [+o putnopvut] by ChanServ
15:45.44putnopvutQwell: I feel...feverish.
15:45.58putnopvutLike it could develop into something worse, but it might just get better.
15:46.00putnopvutHard to say.
15:46.10filego home! sleep! rest! veg!
15:49.12*** join/#asterisk-bugs mvanbaak (i=michiel@vanbaak.xs4all.nl)
16:01.40*** join/#asterisk-bugs michael-i (n=michael-@141.41.40.55)
16:16.20*** join/#asterisk-bugs fakhir (n=fakhir@unaffiliated/fakhir)
16:50.21blitzrageputnopvut: you should not over do it then...
16:50.37putnopvut?
16:50.43putnopvutOh.
16:50.58putnopvutI'm doing okay so far.
16:51.12putnopvutI fixed a bug of yours, in fact.:)
16:51.16blitzragethat's the nice thing about not working in an office... everyone gets sick every year, and I tend to only get sick every 3 or 4 years, and it's because I do something dumb like wake up in the middle of the night and drink some water that my sick friend was drinking out of earlier in the day
16:51.25blitzrageputnopvut: I saw that, I'll test it out today!
16:51.52blitzragedid you notice if RaiseException(<reason>) has to be one of the EXCEPTION() values?
16:52.04putnopvutI'm not certain about that.
16:52.07blitzragei.e. INVALID, ERROR, ABSOLUTETIMEOUT, and whatever the other one was
16:52.20putnopvutI kind of thought it would set the EXCEPTION() to whatever you put in there.
16:52.25putnopvutBut I don't know for certain.
16:52.27blitzrageyou can set a <reason>, but wasn't sure what the referred to... EXCEPTION(type) maybe?
16:52.35putnopvutAh, I don't know.
16:52.40blitzrageya, that's what I was wondering... if it could be anything or not :)
16:52.45blitzrageI'll try it now that RaiseException() could work
16:54.10putnopvutI *think* you can set it to whatever you want, but the only things that Asterisk will set automatically are the builtin values INVALID, ERROR, etc.
16:54.16blitzragemakes sense
16:54.18blitzrageI'm gonna try it out
16:55.12putnopvutLooking at the code, it looks like it will accept anything.
16:55.39putnopvutAnd you just read it back by checking ${EXCEPTION(type)}
16:55.47blitzrageok, that's kinda what I was thinking it would
16:55.48blitzragedo
16:56.10blitzragehrmmm... neato
16:56.10blitzrage<PROTECTED>
16:56.10blitzrage[Oct 30 13:00:49] WARNING[14958]: cli.c:1397 __ast_cli_register: Command '(null)' already registered (or something close enough)
16:56.38putnopvut?
16:56.48blitzragea warning I get when loading trunk
16:57.01blitzrageapparently there is some (null) command somewhere...
16:57.14blitzragewhich makes no sense to me :)
16:57.18putnopvutBeautiful...
16:59.04blitzragehrmmm....
16:59.53blitzrageok... I have a usage "bug".... let me know what you think
17:00.47blitzrageI call ${DB_EXISTS(foo/bar)} which sets ${DB_RESULT} with the returned value if it exists. Lets say the first time I call that something does exist, so ${DB_RESULT} is set
17:01.10blitzrageThen lets say I call it again, and the 2nd time nothing exists... however, ${DB_RESULT} is still populated.... do you think that should be reset to NULL ?
17:01.56putnopvutThat makes sense to me. Of course, I don't use the DB any,so I don't know if that would break people's implementations.
17:02.12blitzrageya, me either... but you shouldn't be relying on the value to exists still... should you?
17:02.46putnopvutNo, I'd think that DB_EXISTS would always set DB_RESULT to something.
17:02.46putnopvutSomething new that is.
17:02.48blitzragei agree
17:03.02blitzrageis that something I should open a bug for?
17:03.16blitzrage(I'm doing a very bad job of helping to close bugs... but at least I'm testing trunk :))
17:03.26putnopvutAnd that's much appreciated!
17:03.51blitzragebtw: bug in RaiseException()
17:04.01blitzrageit is going to e,2, and not e,1 when called
17:04.12blitzrage(i.e. priority 2, not 1)
17:04.18putnopvutReally...tha tshould be easy enough to fix.
17:04.26blitzragei figured as much
17:05.20blitzrageand you're right -- you can set the <reason> to anything you want and access it with ${EXCEPTION(type)}
17:05.39blitzragewhich is a bit confusing from the documentation since it's not immediately obvious that you can do that...
17:05.48blitzragelet me see if I can make a documentation patch for that
17:06.23putnopvut87608 has the fix to make RaiseException go to e,1 instead of e,2
17:06.49putnopvutI'm headed out for a walk, bbl.
17:07.02blitzrageI just got back from the gym and going to eat after this patch -- ttyl!
17:07.08Corydon76-digputnopvut: you go out for a walk every day?
17:07.36Corydon76-digDoctor's orders?
17:07.38putnopvutI'm trying to.
17:07.46blitzrageyou could go out with Corydon76-dig and hold his hand :)
17:07.51blitzrageit'd be so cute!
17:08.09Corydon76-digMy doctor told me I need to exercise more
17:08.15putnopvutI used to walk everywhere I went in Tuscaloosa, and moving to Huntsville has made me too sedentary.
17:08.24putnopvutSo I'm sneaking in walks whenever I can.
17:08.29Corydon76-digAh
17:08.52blitzragewalking is good
17:09.08putnopvutamen to that.
17:09.11putnopvutI'm out.
17:09.14putnopvutBack in about a half.
17:09.19Corydon76-digYeah, I walk, like on the weekend
17:09.22blitzragethink this wording is ok?
17:09.23Corydon76-digmaybe
17:09.23blitzrage-       "in the current context, setting the dialplan function EXCEPTION().  If the \"e\"\n"
17:09.23blitzrage+       "in the current context, setting the dialplan function EXCEPTION().\n"
17:09.23blitzrage+       "You can access the value of <reason> with ${EXCEPTION(type)}. If the \"e\"\n"
17:10.08Corydon76-digblitzrage: sounds fine
17:10.33blitzrageposting to the tracker unless you just wanna do a cp/paste from that and commit it
17:10.50Corydon76-digMatt N originally intended to have type and extra, but I eliminated the extra parameter because it's really not necessary
17:11.13blitzragepatch:  http://pastebin.ca/755333
17:11.15Corydon76-digtype is enough
17:11.22blitzrageya for sure
17:11.25Corydon76-digPost it to the tracker
17:11.29blitzrageokie
17:11.32blitzragethen I can get karma, heh :)
17:11.46Corydon76-digYeppers.  ;-)
17:14.58blitzrageoh man... did the connection to Digium just die again... ?
17:15.25Qwellblitzrage: I'm still here, but it's being funky
17:15.47blitzragenevermind -- it was my internet
17:15.50blitzrageit hiccup'd
17:15.54Qwellyeah, so did ours
17:16.35blitzragehrmmm... probably hiccup'd here because of my tunnel
17:17.54blitzrageM11117
17:17.55MuffinMan[new] [Asterisk] Core/PBX 0011117: Make it obvious you can get the value of RaiseException(<reason>) from ${EXCEPTION(<type>)} reported by blitzrage  http://bugs.digium.com/view.php?id=11117
17:23.29Corydon76-digblitzrage: we could just change the name of the field to be "reason", which would make it more obvious
17:23.40blitzrageI think so too
17:23.53blitzrageI thought of that last night... then forgot about it today :)
17:24.32blitzrageback in a bit -- shower and food time
17:38.56*** join/#asterisk-bugs zerohalo (n=zeroHalo@pool-71-255-162-167.bstnma.east.verizon.net)
17:48.13*** join/#asterisk-bugs michael-i (n=michael-@Wbdf3.w.pppool.de)
18:45.47*** join/#asterisk-bugs asterisk3700 (n=silver19@adsl-153-178-73.mia.bellsouth.net)
18:45.53asterisk3700Hello to all
18:46.09*** part/#asterisk-bugs asterisk3700 (n=silver19@adsl-153-178-73.mia.bellsouth.net)
19:50.06*** join/#asterisk-bugs russellb_ (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
19:50.08*** mode/#asterisk-bugs [+o russellb_] by ChanServ
19:52.07*** join/#asterisk-bugs russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla)
19:52.07*** mode/#asterisk-bugs [+o russellb] by ChanServ
19:52.58*** join/#asterisk-bugs putnopvut (i=putnopvu@nat/digium/x-f8f37ce9d6229ff7)
19:52.59*** mode/#asterisk-bugs [+o putnopvut] by ChanServ
19:58.04*** join/#asterisk-bugs oej (n=olle@64.251.114.2)
19:58.59Deeewayneputnopvut: no problem with analog only
20:01.32putnopvutReally?
20:01.48putnopvutNot even the no audio problem originally reported?
20:01.54Deeewaynehmmm...I'm not seeing with all SIP either
20:02.03putnopvutDo you have the latest 1.4 svn?
20:02.11Deeewayneyeah
20:05.58Deeewayneputnopvut: config problem, sorry
20:07.22putnopvutOne fundamental problem I'm having with this issue is that I don't understand how the part in __ast_read in channel.c avoids a deadlock.
20:07.56Deeewayneyay! now I have no audio :-D
20:14.55Deeewayneputnopvut: are you getting the "... didn't answer/confirm ..." message before the "freed more times that we locked" message ?
20:16.17putnopvutjust a sec.
20:17.32putnopvutI need to revert my change first.
20:19.01putnopvutDangit. I can't get that error to happen...
20:20.26DeeewayneM11071
20:20.27MuffinMan[new] [Asterisk] Applications/app_queue 0011071: No audio on calls into queues with non-persistent agents reported by callguy  http://bugs.digium.com/view.php?id=11071
20:20.36putnopvutI'm not getting the "freed more times than locked" message anymore.
20:20.53putnopvutApparently there are certain conditions necessary for that to happen, and I'm not meeting them.
20:21.14Deeewaynefor some reason I was thinking ^ the issue was one way audio, but its no audio at all w/o your change, right?
20:21.22putnopvutcorrect.
20:21.40putnopvutWith my change, it's perfect audio to the agent, and crappy audio from the agent.
20:21.53Deeewayne....silly agent....
20:21.59putnopvut(Okay not perfect, but good)
20:22.33putnopvutWait a sec, I think I might know something...
20:23.09putnopvutOr maybe not. This will require experimenting.
20:36.39Corydon76-digputnopvut: I've got some prospective changes for 11071, too
20:36.52Corydon76-digjust not done yet
20:37.14putnopvutAh, just the person I wanted to talk to.
20:37.16Corydon76-digputnopvut: what'd you change?
20:37.23putnopvutI haven't committed anything yet.
20:37.37putnopvutBut what I changed was the get_base_channel method in chan_agent
20:37.47Corydon76-digOkay, how?
20:38.01putnopvutjust a sec.
20:38.06Corydon76-digbecause that might work better than what I was trying to do
20:39.48putnopvutI changed it so that if the base channel was a local channel, then instead of returning p->chan, return p->chan->masq instead.
20:40.07putnopvutThat at least got audio working, but the quality from agent->caller was terrible.
20:41.18putnopvutI had just been curious about how the loop you added in __ast_read of channel.c prevented a potential deadlock?
20:54.56Corydon76-digAh, yes
20:55.34Corydon76-digWhat it's doing is to stop agent_read from getting called... because agent_read recursively calls ast_read
20:55.48Corydon76-digi.e. ast_read->agent_read->ast_read
20:56.12Corydon76-digIt's a classic case of locking order
20:56.50Corydon76-digSometimes the underlying channel is locked first, then looks for the agent channel VS sometimes the agent channel is locked first, then it looks for the underlying channel
20:57.29Corydon76-digso by looking for the underlying channel and ensuring that we don't deadlock with one locking order, we ensure no deadlock is possible
21:01.50putnopvutwhy does agent_read call ast_read?
21:11.27Corydon76-digbecause it's a proxy channel
21:11.43Corydon76-digso it calls it for the underlying channel
21:12.18Corydon76-digSo basically, with the new method, both agent_read and agent_write become dead code
21:12.42Corydon76-digsince we resolve the underlying channel with the new API
21:14.49putnopvutThat makes sense.
21:15.14putnopvutSo my fix makes sense, then, since I'm grabbing the REAL underlying channel instead of a Local channel.
21:15.48putnopvutBut as I said before, it had bad audio quality in one direction, so it needs to be tweaked, I guess. I'm not sure how to go about that, though.
21:41.11*** part/#asterisk-bugs slima (i=slima@unaffiliated/slima)
22:16.02*** join/#asterisk-bugs Juggie (n=Juggie@CPE001601df17fb-CM000a73a18a20.cpe.net.cable.rogers.com)
22:16.02*** mode/#asterisk-bugs [+o Juggie] by ChanServ
22:41.41*** part/#asterisk-bugs zerohalo (n=zeroHalo@pool-71-255-162-167.bstnma.east.verizon.net)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.