| 00:10.57 | *** join/#asterisk-dev sustav (~alfa@nat/digium/x-spvdqsclejdzxrib) |
| 00:14.37 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 00:14.37 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 00:21.22 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-aewxovuztlpdmeeu) |
| 00:21.22 | *** mode/#asterisk-dev [+o mnicholson] by ChanServ |
| 01:22.21 | *** join/#asterisk-dev otherwiseguy (~otherwise@69.76.221.79) |
| 04:27.24 | *** join/#asterisk-dev irroot (~gregory@197.170.73.99) |
| 06:46.14 | *** join/#asterisk-dev oej (~olle@87.96.134.129) |
| 06:46.14 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:07.51 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-tlawacfyvfoiyzha) |
| 07:07.51 | *** mode/#asterisk-dev [+o mnicholson] by ChanServ |
| 07:29.29 | *** join/#asterisk-dev schmidts (~schmidts@Lo20-rt01.lm33.sil.at) |
| 07:29.32 | schmidts | good morning |
| 08:45.53 | *** join/#asterisk-dev vrban (~kris@2001:6f8:9c8:0:224:e8ff:fe20:d141) |
| 08:57.00 | *** join/#asterisk-dev oej (~olle@static-213-115-251-100.sme.bredbandsbolaget.se) |
| 08:57.00 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 09:02.55 | schmidts | good morning oej |
| 09:08.35 | oej | Good morning |
| 09:26.46 | *** join/#asterisk-dev apten (~apten@carbon.gonicus.de) |
| 09:30.19 | oej | In my AMI discovery tour I've realized that AMI is very serial. One action at a time in one Asterisk. |
| 10:11.22 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 10:11.23 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 10:41.35 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:79f7:5864:38ec:b42c) |
| 11:16.28 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@212.179.75.202) |
| 11:16.28 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 11:19.08 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 11:19.08 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 11:43.30 | WIMPy | I guess it may be related to the log level thing. 'dialplan' show doesn't output anything any more. |
| 11:44.56 | *** join/#asterisk-dev irroot (~gregory@196-210-202-232.dynamic.isadsl.co.za) |
| 11:49.20 | *** join/#asterisk-dev irroot (~gregory@196-210-202-232.dynamic.isadsl.co.za) |
| 12:01.13 | wdoekes2 | WIMPy: dialplan show works over here. are you connected over remote? or -c? what is your verbosity level? |
| 12:02.13 | WIMPy | remote, verbose 1 |
| 12:04.55 | wdoekes2 | can't reproduce |
| 12:06.18 | WIMPy | Is that a restart thing again? |
| 12:06.21 | irroot | SAVE ME FROM STUPID PEOPLE |
| 12:06.49 | WIMPy | ok, I'll shut up. |
| 12:07.48 | irroot | afternoon folks |
| 12:18.42 | WIMPy | Ok, I did everything anew. But I can't seem to increase verbose or debug levels. |
| 12:19.04 | schmidts | hi irroot |
| 12:19.44 | schmidts | wdoekes2 have you seen my message in kamailio, do you have an idea what i can do to solve this problem? |
| 12:21.06 | WIMPy | Yes, If I increase verbose/debug in asterisk.conf, that works, as soon as I restart the remote console, but 'core set verbose' doesn't do anything. |
| 12:23.09 | wdoekes2 | I saw it, but I wasn't sure what happens exactly: could you describe it in a bit more detail? (I can think of a related issue, but that hasn't got anything to do with registers) |
| 12:24.22 | wdoekes2 | WIMPy: I'm running 'make samples' config right now, what do I need to break it? |
| 12:25.18 | WIMPy | Todays trunk, not sure, what else. |
| 12:27.58 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:27.58 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 12:28.30 | WIMPy | I'm still on tho that auto-answer with park(). Todays result is different and seems more logical to me, but the funny thing is that not doing an Answer() first takes an extra second. That feels fishy |
| 12:29.27 | leifmadsen | WIMPy: have you tried playing an audio file without Answer() first and see how that handles? |
| 12:29.41 | leifmadsen | I suspect you'll get the same type of clipped audio |
| 12:31.11 | WIMPy | The interesting thin is that I got the beginning cut off on an ISDN phone as well, when I posted the issue, which doesn't make sense. But that doesn't happen any more. |
| 12:31.56 | leifmadsen | huh |
| 12:32.04 | leifmadsen | was it consistent? |
| 12:32.09 | WIMPy | I only happens on SIP now, but the interestin bit is that not using Answer creates a rather long additional pause befor the digit is played, but still only partially. |
| 12:32.42 | WIMPy | Will try with some other playback. |
| 12:33.52 | WIMPy | The delay is long enough to show in the log. |
| 12:34.05 | leifmadsen | what shows in the log? |
| 12:34.25 | WIMPy | The extra delay. |
| 12:35.21 | WIMPy | Holy shit. |
| 12:36.33 | WIMPy | When I do a Playback(digits/2) before Park(), I hear "2" correctly, but the "1" played by Park() is cut off. Using Answer() before does not change any of that. |
| 12:36.44 | WIMPy | SIP only. |
| 12:38.11 | leifmadsen | ok then something funky must be going on with Park() |
| 12:38.16 | leifmadsen | what version of Asterisk are you using? |
| 12:38.35 | WIMPy | Todays SVN trunk. |
| 12:42.17 | leifmadsen | WIMPy: can you try prior to r332100 out of curiosity? |
| 12:42.27 | leifmadsen | lets say... 332044 |
| 12:42.35 | WIMPy | I just tried with Zoiper/IAX. That's ok. |
| 12:43.34 | leifmadsen | what (soft)phone were you using? |
| 12:43.48 | leifmadsen | still seems odd the phone would make a difference... but maybe? |
| 12:43.52 | WIMPy | Snom 360 |
| 12:44.08 | WIMPy | It's the only SIP one. |
| 12:45.29 | leifmadsen | what about via Ekiga or Jitsi? |
| 12:47.52 | WIMPy | Hmm. Maybe it's an SRTP issue? I will try to disable that. |
| 12:51.13 | leifmadsen | hmmm perhaps? |
| 12:51.21 | leifmadsen | it sounds like something non-standard must be going on |
| 12:51.43 | leifmadsen | because I've never run into cut off audio on Park() like that, especially after other audio had already been played |
| 12:51.52 | WIMPy | Still quite interesting to happen in the middle of a call. |
| 12:51.54 | leifmadsen | would be ideal to figure out exactly what the scenario is |
| 12:51.56 | leifmadsen | agreed |
| 12:53.54 | WIMPy | Same without SRTP. |
| 12:57.34 | leifmadsen | ok, then I'd try with a couple of other types of (soft)phones and see which reproduce the problem (if any). After that, I'd suggest rolling back to that version I suggested to determine if a regression happened after 332100 (which contained some restructuring of parking stuff) |
| 12:58.11 | WIMPy | Disabling G.722 didn;t change it, either. |
| 13:06.18 | WIMPy | Interestingly enough it seems to work with other phones. |
| 13:07.37 | WIMPy | Could it be timestamps perhaps? |
| 13:09.25 | WIMPy | can't see how else the start of a new sample could possibly make a difference to the phone. |
| 13:12.20 | *** part/#asterisk-dev vrban (~kris@2001:6f8:9c8:0:224:e8ff:fe20:d141) |
| 13:19.56 | irroot | "An expert is a person who has made all the mistakes that can be made in a very narrow field" -Nils Bohr |
| 13:22.10 | WIMPy | Somehow svn has effed up :-( |
| 13:22.34 | irroot | looks arround was not me |
| 13:26.31 | *** join/#asterisk-dev cVsup (~cVsup@187.127.102.184) |
| 13:26.48 | cVsup | how can setup fxs as extension? |
| 13:37.03 | leifmadsen | cVsup: please use #asterisk for support issues |
| 13:37.10 | leifmadsen | irroot: hey hey! :0 |
| 13:37.17 | irroot | leifmadsen got memo wilco |
| 13:37.21 | leifmadsen | bah.... I never hold the shift key down long enough on smileys |
| 13:37.23 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-bohxxizvymjlpomd) |
| 13:37.23 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 13:37.43 | irroot | have infact already worked on things for trunk along those lines thx |
| 13:38.00 | leifmadsen | irroot: coolio -- I think some more explainations and documentation about that ignorebusy option needs to be implemented. It might also be interfering with the ringall strategy |
| 13:38.16 | leifmadsen | irroot: ok coolio -- might be something we need to put in 1.8 and 10 as well to fix a possible regression |
| 13:39.19 | irroot | leifmadsen doing some testing of the work ive done ironing out the strangeness busy with mail issues this week and not much done on * |
| 13:48.32 | WIMPy | leifmadsen: I'm on 332944 now as you suggested. The issue persists, but there's a big pause between the samples now. |
| 13:48.47 | leifmadsen | WIMPy: ok then go back to the top of trunk |
| 13:48.58 | leifmadsen | it wasn't that change that did anything then |
| 13:49.02 | leifmadsen | if anything, it made it better |
| 13:49.20 | leifmadsen | regardless, sounds like a problem with the Snom phone |
| 13:50.27 | WIMPy | Yes, but the phone can't know when Asterisk starts a new dialplan application. |
| 13:50.54 | WIMPy | Or shouldn't be able to... |
| 13:51.31 | leifmadsen | well it's not that it's starting the new dialplan application -- it must have something to do with the way it is handling the rtp media |
| 13:51.53 | leifmadsen | at this point I have no idea, so someone else will have to surmise some ideas |
| 13:52.27 | WIMPy | Yes, but only at the start of Park(). I have no clue, either as to how that can be possible. |
| 13:52.44 | WIMPy | Well, apart from timestamps maybe, as suggested above. |
| 13:54.46 | schmidts | WIMPy you know i would general say its your phone :P |
| 13:55.32 | WIMPy | But it cannot possibly do it on its own. |
| 13:58.30 | WIMPy | The channel is put in to hold state after the announcement so that can't be related, either. |
| 14:04.26 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 14:04.26 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 14:04.40 | oej | malcolmd: Good morning! |
| 14:05.00 | oej | Looking at the IRC messages one could come up with a strange theory. |
| 14:05.13 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-mwqfrvxlyxzseqoe) |
| 14:05.15 | oej | Malcolmd joined the chat, kpfleming immediately left. . . |
| 14:05.49 | malcolmd | aloha :D |
| 14:06.11 | malcolmd | kpfleming on his way to the office from home, malcolmd having just arrived at the office. magical teleportation powers at work? perhaps |
| 14:11.46 | oej | We need to adopt the media plane in ASterisk 12 for magical teleportation. And we do need to write an RFC for the SDP syntax. |
| 14:11.54 | oej | Lots of work to do. |
| 14:14.59 | opticron | matter stream via sdp? that's going to make asterisk all gross |
| 14:16.02 | WIMPy | Reminds me of the beer transmisstion protocoll discussions back in the mailbox days. |
| 14:16.34 | oej | BTMP is a good protocol. |
| 14:20.28 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 14:27.58 | WIMPy | BTW: Who had the idea to make prilocaldialplan default to 'national'? |
| 14:29.30 | jsmith | WIMPy: Not I, but I always immediately change it to "unknown" |
| 14:30.20 | WIMPy | Yes, there are a lot of users comming here to ask why they can't set their callerID. |
| 14:35.38 | *** join/#asterisk-dev irroot (~gregory@196-210-140-50.dynamic.isadsl.co.za) |
| 14:47.54 | *** join/#asterisk-dev Juggie (~Juggie@unaffiliated/juggie) |
| 15:10.22 | *** join/#asterisk-dev joshaidan (~brianj@24.109.210.41) |
| 15:11.06 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:11.06 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:25.01 | *** join/#asterisk-dev zerohalo (~zerohalo@74-94-141-204-ma.hfc.comcastbusiness.net) |
| 15:32.02 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 15:32.02 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 15:36.10 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 15:36.10 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:45.03 | *** join/#asterisk-dev chazzam (~chazz@50-81-150-34.client.mchsi.com) |
| 15:47.35 | *** join/#asterisk-dev oej (~olle@87.96.134.129) |
| 15:47.36 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 16:02.26 | *** part/#asterisk-dev apten (~apten@carbon.gonicus.de) |
| 16:11.26 | pabelanger | For those playing along at home, I could not get asterisk 1.4.1 configure script to finish :( |
| 16:11.33 | pabelanger | yes, 1.4.1 |
| 16:15.04 | mjordan | otherwiseguy: I'll post the patch and let you take a look |
| 16:16.05 | otherwiseguy | mjordan: we don't send NO_USER_RESPONSE in an outbound dial situation with a timeout, though. Asterisk sends AST_CAUSE_PROTOCOL_ERROR. |
| 16:16.29 | otherwiseguy | It's in retrans_pkt. |
| 16:16.58 | mjordan | yeah, I changed that |
| 16:16.59 | otherwiseguy | We treat 408s as NO_USER_RESPONSE, which is a little weird. |
| 16:17.07 | mjordan | PROTOCOL_ERROR is wrong, imo |
| 16:17.16 | otherwiseguy | chan_sip is not particularly consistent. |
| 16:17.35 | *** join/#asterisk-dev eliel_ (~quassel@200.32.121.66) |
| 16:17.50 | otherwiseguy | but the point is, HANGUPCAUSE is set, and someone can check that without creating a new dialstatus. |
| 16:18.02 | otherwiseguy | and cdr value |
| 16:18.03 | mjordan | in particular, that piece of code was pretty wrong - above it, we would (normally) set the hangup cause to NO_USER_RESPONSE, then return PROTOCOL_ERROR, ignoring the previously set hangup cause |
| 16:18.29 | mjordan | are you sure? If you outbound dial several extensions, how do you know what the hangup cause was? |
| 16:18.56 | mjordan | the CDR, I agree, is a 'nice to have' - but I added it to maintain a level of consistency |
| 16:19.37 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 16:19.37 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 16:19.48 | mjordan | https://reviewboard.asterisk.org/r/1608/ |
| 16:20.44 | mjordan | and granted, you already get the hangup cause code if you use CEL, so if you're really just looking for statistics on what failed and why, CEL is still always going to be better. But this at least lets you do some stuff after the Dial command in the dialplan |
| 16:24.49 | otherwiseguy | I assumed that hangupcause would be set based on the winning call, but I just did a call to an unlistening server and a phone that just didn't answer, and hangupcause is still 111, so I guess that doesn't work. |
| 16:25.46 | *** join/#asterisk-dev tzafrir_laptop (~tzafrir@212.179.75.202) |
| 16:25.46 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 16:26.04 | mjordan | I don't *think* this is too intrusive, but I'm open to criticism on that. I definitely don't want this to perturb a lot of people's existing CDR / dialplan stuff |
| 16:34.04 | otherwiseguy | I tried with busy, and hangupcause gets set to 18 even though it was dialed with a non-listening address as well. using hangupcause looks kind of racy. Is there any reason existing DIALSTATUSs can't be used CHANUNAVIL? CONGESTION? |
| 16:34.19 | otherwiseguy | mjordan: ^ |
| 16:34.56 | mjordan | that's odd. When I tested with busy, it wasn't a hang up cause of 18 |
| 16:35.26 | otherwiseguy | I'm using 1.8, so it might be different than older versions. |
| 16:35.45 | mjordan | backing up a second - you used a SIP phone and dialed a non-listening address, so the INVITE timed out, right? |
| 16:35.54 | otherwiseguy | yes |
| 16:36.03 | mjordan | that should be 18, which should map (post patch) to timed out |
| 16:36.07 | mjordan | why would that be busy? |
| 16:36.48 | pabelanger | So... /bin/sh: astdb2sqlite3: not found when running asterisk-10.0.0-rc2 |
| 16:37.08 | pabelanger | this is because astdb2sqlite3 is not in $PATH |
| 16:37.19 | otherwiseguy | I did Dial(SIP/unused@unlistening.address&Local/busy@default) with busy@default being a call to Busy(). Got hangupcause 18. |
| 16:37.30 | otherwiseguy | Doing just Dial(SIP/unused@unlistening.address) I got 111. |
| 16:37.37 | otherwiseguy | w/o patch |
| 16:37.47 | WIMPy | votes for 0x29 temporary failure |
| 16:39.31 | mjordan | So, the hangupcausecode is going to be checked before it gets returned in the dialplan, as it will get checked for each channel |
| 16:40.01 | otherwiseguy | also, it drives me absolutely nuts that the dialstatus variables are just hardcoded strings throughout app_dial without really being defined anywhere. |
| 16:40.15 | mjordan | I don't know why busy would map to 18, that doesn't make much sense to me. Pre-patch, timeout gets forced to 111, which doesn't map to 408, which is supposed to be timeout |
| 16:40.28 | mjordan | otherwiseguy: agreed, but that has issues all over the codebase |
| 16:40.45 | otherwiseguy | mjordan: yeah, it definitely should map to the same thing 408 does. |
| 16:41.26 | mjordan | so - what you see as the hangupcausecode for multiple dials is probably set similarly to how the DIALSTATUS gets set - there's a precedence given based on the channels that hung up |
| 16:41.32 | mjordan | (although I'm not 100% on that) |
| 16:41.51 | mjordan | but take a look at the patch on lines 1087 of app_dial |
| 16:42.00 | otherwiseguy | hangucpause = random() |
| 16:42.12 | mjordan | random() % 111 |
| 16:42.23 | mjordan | or something like that |
| 16:43.10 | mjordan | BUSY gets first billing, then CONGESTION, etc. - so if you have any channels that returned a hangup type of BUSY (which is mapped in the handle_cause method), we return a DIALSTATUS of BUSY. Etc. |
| 16:44.07 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:44.32 | mjordan | it *probably* should check for the hangup cause with the greatest number of failures and return that, but that isn't what its doing now, so I didn't make that part of this patch |
| 16:45.32 | mjordan | SO: this explains why if you dial two channels, one times out and the other is busy, why you'd get a BUSY response. Not sure about the hangupcausecode. |
| 16:49.13 | *** join/#asterisk-dev jrose_atDigium (~jon@nat/digium/x-slyekteowtngarya) |
| 16:51.45 | *** join/#asterisk-dev sustav (~alfa@nat/digium/x-iqmttwqdbqvlptlk) |
| 16:51.49 | otherwiseguy | this time when I dialed, it showed up as 111. So hangupcause seems a little unreliable. |
| 16:53.32 | otherwiseguy | mjordan: waiting for the SIP timeout is annoying. Ok, when I just dial a non-answering device, I get CHANUNAVAIL, not NOANSWER. |
| 16:54.41 | otherwiseguy | non-listensing, that is. |
| 16:55.06 | anonymouz666 | IAX2 deadlock asterisk 1.8 in production, I reproduce exactly the same setup injecting tons of calls and the deadlock does not happen |
| 16:55.14 | anonymouz666 | nice |
| 16:55.45 | mjordan | otherwiseguy: prepatch, that sounds right |
| 16:56.48 | mjordan | lunching be back in a bit |
| 16:56.49 | anonymouz666 | without 'core show locks' and debug threads, just using the backtrace-threads is it possible to fix the deadlock? |
| 16:57.36 | kpfleming | sometimes |
| 16:59.01 | anonymouz666 | it's better to enable dont optimize and debug threads then |
| 16:59.12 | anonymouz666 | already happened 3 times |
| 17:17.41 | tzanger | I think that's another long-standing irk I will address.. when you call and it's unavailable (congestion) the cli still says circuit-busy |
| 18:19.04 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 18:19.04 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 18:40.42 | *** join/#asterisk-dev luckman212 (~irc@pool-108-41-8-176.nycmny.fios.verizon.net) |
| 18:44.40 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 18:44.40 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 18:50.23 | *** join/#asterisk-dev joshaidan (~brianj@S0106000c6e79821d.tb.shawcable.net) |
| 19:03.05 | *** join/#asterisk-dev irroot (~gregory@197.172.84.224) |
| 19:14.08 | *** join/#asterisk-dev sezuan (bouncer@irc.scheff32.de) |
| 19:14.11 | pabelanger | mjordan: otherwiseguy: ASTERISK-18959 |
| 19:14.27 | pabelanger | potential blocker for asterisk 10 IMO |
| 19:18.13 | mjordan | Was there a database for it to convert? |
| 19:19.06 | mjordan | ah, nm, I see the copy old astdb. |
| 19:19.30 | mjordan | What happens if you follow the instructions in the ERROR messages? |
| 19:28.13 | pabelanger | mjordan: still fail |
| 19:28.20 | mjordan | okay |
| 19:28.26 | pabelanger | the utility is installed outside of the path |
| 19:28.34 | pabelanger | EG: chroot installs / live_ast |
| 19:31.51 | mjordan | pabelanger: how may other tests need to be run against the astdb upgrade? |
| 19:33.41 | pabelanger | mjordan: depends on how many different versions of asterisk we want to upgrade from |
| 19:34.06 | pabelanger | I'm picking random 1.4, 1.6.0, 1.6.1, 1.6.2, 1.8 release and going from there |
| 19:34.38 | kpfleming | there is no need to check a lot of diffferent 'original' versions, since their behavior in regards to astdb is identical |
| 19:35.08 | pabelanger | kpfleming: okay, I wasn't sure about that |
| 19:49.58 | *** join/#asterisk-dev irroot (~gregory@197.169.71.128) |
| 19:57.29 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 19:57.29 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 20:02.51 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 20:02.51 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 20:03.47 | *** join/#asterisk-dev joshaidan (~brianj@S0106000c6e79821d.tb.shawcable.net) |
| 20:22.17 | *** join/#asterisk-dev vpopov (~happylife@46.251.80.89) |
| 20:25.09 | *** join/#asterisk-dev malcolmd (~malcolmd@pdpc/sponsor/digium/malcolmd) |
| 20:25.09 | *** mode/#asterisk-dev [+o malcolmd] by ChanServ |
| 20:33.18 | jrose_atDigium | ASTERISK-18807 |
| 20:33.27 | jrose_atDigium | ... no rbot here eh |
| 20:33.48 | jrose_atDigium | https://issues.asterisk.org/jira/browse/ASTERISK-18807 |
| 20:34.12 | jrose_atDigium | I'm wondering what people think should happen when a dialplan has this particular mistake.... |
| 20:34.33 | jrose_atDigium | currently you can just have the same label in two different hints on the same extension. |
| 20:34.49 | jrose_atDigium | This patch would change it so that if that happens, the new hint gets discarded and a warning is displayed. |
| 20:35.13 | jrose_atDigium | Right now though, I'm thinking it would be better to just discard the label and go ahead and add the hint, while still displaying the warning. |
| 20:35.27 | jrose_atDigium | *Also, if the whole line is discarded, it should probably be an error rather than a warning. |
| 20:35.42 | jsmith | I wouldn't throw out the whole line -- simply ignore the second label in the extension |
| 20:35.45 | Qwell | I don't know if the patch checks for exten/label, but it probably needs to. Then yes, I agree the duplicate should be discarded. |
| 20:35.57 | Qwell | with a warning or something of course |
| 20:36.35 | Qwell | for instance, I believe 1111,1(bacon),NoOp and 1112,1(bacon),NoOp would be valid. |
| 20:36.49 | jrose_atDigium | Clearly |
| 20:36.50 | Qwell | Though, it would get a bit funky with pattern matches.. |
| 20:36.53 | jrose_atDigium | This is more like... |
| 20:36.54 | Qwell | example: |
| 20:37.08 | jrose_atDigium | 1111,1(bacon),NoOp |
| 20:37.10 | Qwell | _XXXX,1(bacon),NoOp and 1111,1(bacon),NoOp |
| 20:37.17 | jrose_atDigium | 1111,2(bacon),NoOp |
| 20:38.07 | jrose_atDigium | Yeah, in that case it sorta seems like the more specific pattern should have priority. |
| 20:38.36 | Qwell | right, and it would have priority, but the label checking would be weird |
| 20:38.57 | Qwell | I think mine are more edge cases than anything. With the report as written, I agree with your assessment. |
| 20:56.04 | *** join/#asterisk-dev irroot (~gregory@197.174.44.193) |
| 21:04.58 | Corydon76-home | Qwell: just thought of another, perhaps strange label lookup. Let's suppose there is _XXXX,1(bacon),Foo() and 1111,1,Bar(), and you Goto(1111,bacon). What do you suppose will happen? |
| 21:05.28 | kpfleming | indeed, that is probably 'undefined behavior' at the moment |
| 21:06.11 | Corydon76-home | I suspect bacon will be found, yet Bar() will execute. |
| 21:07.12 | kpfleming | right... bacon will be found and turned into 'priority 1', then a lookup for '1111,1' will be done |
| 21:07.53 | Corydon76-home | The real lesson here is "Don't do that" |
| 21:10.04 | *** join/#asterisk-dev jrose_atDigium (~jon@nat/digium/x-harrvcndzctkhfcn) |
| 21:10.17 | Qwell | that hurts my brain |
| 21:10.32 | kpfleming | could smell the smoke from his office |
| 21:11.00 | Corydon76-home | Did it smell like bacon frying? |
| 21:11.07 | kpfleming | no, unfortunately |
| 21:11.11 | WIMPy | I guess I'm lucky that I didn't dare to use labels on mor interesting dialplans. |
| 21:11.28 | WIMPy | likes to split and recombine between priorities. |
| 21:11.51 | Corydon76-home | WIMPy: well, no, labels are perfectly safe, if you use them correctly |
| 21:12.37 | Corydon76-home | When you're doing branch logic, they help enormously, especially when you need to go back and add additional priorities somewhere in the middle |
| 21:12.52 | WIMPy | Sure. |
| 21:12.54 | kpfleming | labels, with 'n' and 's' priorities, are immensely useful |
| 21:14.02 | WIMPy | But then when you do some pattern magic, n doesn;t work anyway. |
| 21:14.04 | Corydon76-home | Even if you didn't have 'n' and 's', they would still be extremely helpful, because they act as "bookmarks" |
| 21:14.12 | Qwell | WIMPy: sure it does |
| 21:15.10 | Corydon76-home | WIMPy: it can, as long as you either intersperse (probably unreadable logic) or set a starting priority |
| 21:15.13 | WIMPy | Not if you do things like _[12],1,noop ; 1,2,noop ; 2,2,noop ; _[12],3,noop |
| 21:15.34 | Corydon76-home | WIMPy: that's what the 's' priority is for |
| 21:15.47 | Corydon76-home | 1, n, s, n, in your example |
| 21:16.06 | WIMPy | I guess I missed that one. |
| 21:16.07 | *** join/#asterisk-dev otherwiseguy (~otherwise@rrcs-173-197-0-67.west.biz.rr.com) |
| 21:17.13 | WIMPy | But it might make smoke signals (even more) likely :-) |
| 21:17.20 | Corydon76-home | _[12],1,... 1,n,... 2,s,... _[12],n,... |
| 21:17.52 | WIMPy | It's not the most readable construction anyway. |
| 21:18.13 | WIMPy | But if I can get around a set of ExecIfs or the like, I will. |
| 21:18.19 | Corydon76-home | I wouldn't say that 's' is as-often documented, because it's a really advanced user feature. It's quite easy to get wrong. |
| 21:18.38 | kpfleming | especially when users confuse it with extension 's' |
| 21:18.56 | jrose_atDigium | Really advanced crazy user. |
| 21:18.58 | Corydon76-home | I tend to use Exec with ${IF()} instead, with an else of NoOp |
| 21:19.12 | jrose_atDigium | I would think that it would generally be better and more understandable to just use macros. |
| 21:19.26 | Corydon76-home | Or Gosubs |
| 21:19.57 | WIMPy | I'd see that as something totally different. |
| 21:20.21 | WIMPy | But the good thing is that ther's enough possibilities for everyone. |
| 21:20.43 | Corydon76-home | jrose_atDigium: Not always. For example, I used _8[0-9] to set location state on a particular phone system, and then used a Playback at each of the various individual extensions to say different things as to what the user had just set their extension to. |
| 21:21.02 | Corydon76-home | But the logic for setting the location was identical |
| 21:21.26 | kpfleming | i've used that before many times too |
| 21:21.47 | WIMPy | Like feature codes for CF. |
| 21:21.48 | Corydon76-home | i.e. 0=out of office, 1=in office, 3=forward to cell, then voicemail, 9=always send to voicemail |
| 21:21.55 | kpfleming | in fact, when murf was here and talking about changing the guts of extension matching, he had a proposal that wouldn't have supported that usage, and i had to show him |
| 21:23.47 | WIMPy | I'd still love to have some sort of pre-dialplan extension that will always be executed before matching is done. |
| 21:24.00 | Corydon76-home | jrose_atDigium: Just curious, do you run Asterisk at home in a production setting (i.e. answering your home phone line)? |
| 21:24.03 | kpfleming | yep, that's been discussed a number of times too |
| 21:24.09 | jrose_atDigium | No, I don't. |
| 21:24.19 | jrose_atDigium | My Asterisk usage is strictly development. |
| 21:24.39 | kpfleming | has not had a home phone line for more than five years |
| 21:24.40 | Corydon76-home | jrose_atDigium: there's a lot of things you can learn by running such a hobbyist deployment that you might not otherwise experiment with. |
| 21:25.00 | Corydon76-home | kpfleming: not even with VoIP? |
| 21:25.04 | kpfleming | nope |
| 21:25.30 | Qwell | kpfleming: +1 |
| 21:25.38 | Qwell | < also 5 years |
| 21:25.50 | *** join/#asterisk-dev joshaidan (~brianj@S010600095bfe15df.tb.shawcable.net) |
| 21:26.21 | Corydon76-home | I think the big thing that you learn by running a home deployment is figuring out what annoys you the most that could be fixed that users don't think is important enough to either complain about or file an issue for |
| 21:26.49 | WIMPy | Or you get used to it. |
| 21:27.00 | Corydon76-home | WIMPy: that's also true |
| 21:28.31 | Corydon76-home | jrose_atDigium: In any case, I strongly encourage you to do so, even if your line is just VoIP. Asterisk evolved a great deal just from the putterings of hobbyists. |
| 21:29.31 | jrose_atDigium | I'll think about it, but of my total phone use, I'd say about 99.9% of it is just me calling myself to debug Asterisk stuff :P |
| 21:29.37 | jrose_atDigium | I'm not what you'd call a prolific phone user. |
| 21:30.01 | WIMPy | used to be |
| 21:30.10 | Corydon76-home | Nor am I. I do, however, host 2 toll-free numbers for nonprofits from a server sitting next to me |
| 21:30.47 | jrose_atDigium | Is that tax deductible? |
| 21:30.51 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 21:30.51 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 21:30.55 | Corydon76-home | used to host 888-DIGIUM-1 and 888-DIGIUM-0 from the same machine |
| 21:31.00 | Corydon76-home | Yes, it is |
| 21:31.16 | jrose_atDigium | takes interest |
| 21:32.13 | Corydon76-home | Anything you spend on behalf of a charitable nonprofit is tax-dedeductible to the fullest extent possible under the law. |
| 21:32.25 | Corydon76-home | s/dede/de/ |
| 21:33.29 | jrose_atDigium | I wonder how far that could be pushed... |
| 21:33.34 | Corydon76-home | Anything you spend, miles on your vehicle, supplies you donate... pretty much anything done on behalf of a charity EXCEPT for your time. |
| 21:34.12 | Corydon76-home | jrose_atDigium: if you don't run your own business, pretty far, actually |
| 21:35.19 | Corydon76-home | If you do run your own business, you need to keep vehicle mileage logs to take that deduction, and you need your receipts... Businesses get audited far more often than individuals |
| 21:37.25 | Corydon76-home | An interesting one is: if you teach Sunday school, you can deduct mileage driven to church. |
| 21:37.49 | Corydon76-home | (both to and from) |
| 21:42.58 | Corydon76-home | Caveat: the organization needs to be a 501(c)(3) as indicated in IRS Publication 78. |
| 21:45.54 | Corydon76-home | Oh, and for cash donations over a certain amount, you need a signed letter from the organization for CYA. |
| 21:46.49 | Qwell | Corydon76-home: Will you help me register PayQwellsMortgage as a 501(c)(3)? Then I can donate all my money to it. |
| 21:47.18 | Corydon76-home | Good luck on that one |
| 21:47.55 | Corydon76-home | I can help you with the paperwork, but you get to pay the (nonrefundable) $850 application fee. |
| 21:48.35 | jrose_atDigium | Qwell: If you do that, I can set up an Asterisk box within the office that does nothing but routes SIP calls to one of your phones and then I'll be able to deduct my back and forth to work gasoline so that I can 'administer' said box. |
| 21:49.03 | Qwell | jrose_atDigium: Deal! But you also have to do dishes. |
| 21:50.15 | Corydon76-home | jrose_atDigium: Just FYI, I have done this before. I filed for (and received) nonprofit status on Nashville 2600 Organization. |
| 21:51.29 | Corydon76-home | and in Tennessee, that means sales tax exemption on everything the organization buys... 9.25% the organization doesn't have to pay for goods |
| 21:53.31 | jrose_atDigium | I guess I could set up an Asterisk server at home for the purpose of making pitch shifted calls to my friends... |
| 22:13.13 | *** join/#asterisk-dev irroot (~gregory@197.108.118.96) |
| 22:32.18 | *** join/#asterisk-dev joshaidan (~brianj@24.109.210.41) |
| 23:31.54 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-bohxxizvymjlpomd) |