| 00:16.38 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.209) |
| 00:43.41 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 00:43.41 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 00:45.06 | *** join/#asterisk-dev mjordan (~matt@75.76.55.191) |
| 00:45.06 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 00:45.51 | X-Rob | mjordan: what needs to happen before 12.0.1? |
| 00:46.03 | mjordan | 12.0.1 or 12.1.0? |
| 00:46.07 | X-Rob | or.. whatever |
| 00:46.07 | X-Rob | 8) |
| 00:46.50 | mjordan | X-Rob: can you see this - https://issues.asterisk.org/jira/secure/ReleaseNote.jspa?projectId=10376&version=12493 |
| 00:47.11 | mjordan | those would be the issues that I tagged for "we should probably do that" |
| 00:47.27 | X-Rob | mjordan: yes I can, and OK. |
| 00:47.32 | X-Rob | moar pjsip commands would be handy |
| 00:47.35 | mjordan | but that's missing one or two |
| 00:47.40 | X-Rob | also codecs |
| 00:47.44 | X-Rob | which I swear I'm testing today. |
| 00:47.49 | mjordan | yeah, that should be on the list. It was in my head... :-) |
| 00:48.00 | mjordan | multihomed |
| 00:49.12 | X-Rob | multihomed works |
| 00:49.18 | X-Rob | we've got that in our RPM, and it's good. |
| 00:49.29 | X-Rob | Did I comment on the review of that? |
| 00:49.32 | X-Rob | looks |
| 00:50.42 | mjordan | we haven't committed it yet, since we don't have any tests for it yet. Going to be hard to automate those tests, however, so I'm still mulling on what to do about that |
| 00:52.56 | X-Rob | also just applied the codec patch |
| 01:02.26 | mjordan | X-Rob: commented on ASTERISK-23173. I'm curious what of those options you guys use |
| 01:04.27 | X-Rob | mjordan: uh, URI_OPTIONS? That's for auto-answering/intercom |
| 01:04.39 | X-Rob | oh wait |
| 01:04.47 | X-Rob | 73 is PEER INFO |
| 01:05.20 | X-Rob | From that, we suck the User-Agent that's registered to it, and if we know about it we change the header going to it for paging |
| 01:16.37 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.184) |
| 01:17.26 | X-Rob | wups, sorry tm1000, that's my nat being fucked |
| 01:17.27 | X-Rob | HOWEVER |
| 01:17.30 | X-Rob | Codecs were correct! |
| 01:17.50 | X-Rob | ...and this is the wrong channel. |
| 01:23.28 | tm1000 | X-Rob: call me back then |
| 01:23.37 | tm1000 | lets g722 this joint in PJSIP yo |
| 01:24.11 | mjordan | lol |
| 01:24.19 | mjordan | X-Rob: so, mostly just useragent is what you need? |
| 01:27.48 | X-Rob | mjordan: yes |
| 01:27.54 | X-Rob | But I like the idea of PJSIP_REGISTRATION |
| 01:28.01 | tm1000 | mjordan: is that a free plugin or JIRA or paid? |
| 01:28.11 | X-Rob | that's an excellent idea.. except, how do I tell which endpont? |
| 01:28.45 | tm1000 | X-Rob: yes |
| 01:28.48 | tm1000 | and its HD |
| 01:28.51 | tm1000 | whooo whooo |
| 01:29.07 | X-Rob | m=audio 12410 RTP/AVP 9 0 8 101 |
| 01:29.07 | X-Rob | c=IN IP4 192.168.15.5 |
| 01:29.07 | X-Rob | a=rtpmap:9 G722/8000 |
| 01:29.07 | X-Rob | a=rtpmap:0 PCMU/8000 |
| 01:29.07 | X-Rob | a=rtpmap:8 PCMA/8000 |
| 01:29.07 | X-Rob | a=rtpmap:101 telephone-event/8000 |
| 01:29.15 | X-Rob | Aww yis. Except for the IP being broken. |
| 01:29.20 | X-Rob | Which I think actually is another issue. |
| 01:29.29 | X-Rob | because I've tried to be too smart in sipsettings. |
| 01:29.38 | mjordan | X-Rob: you'd specify the endpoint still - something like PJSIP_REGISTRATION(endpoint_name,option) |
| 01:30.48 | mjordan | it's a thought anyway. We'll move the issue on up - I just wanted to make sure it wasn't something like expire. No idea why that'd be useful. |
| 01:30.50 | tm1000 | rofl |
| 01:31.07 | tm1000 | X-Rob: "Oh for F*** sake, its still sending the wrong IP….ok this is a bug now" |
| 01:31.14 | X-Rob | heh |
| 01:31.21 | X-Rob | checks his endpoints |
| 01:32.33 | mjordan | not a grandstream, is it? |
| 01:33.03 | X-Rob | So. I have -one- endpoint defined, which has an external_media and external_signalling address |
| 01:33.13 | X-Rob | wait. one TRANSPORT |
| 01:33.17 | X-Rob | only one transport |
| 01:33.34 | X-Rob | I have my handset endpoints defined WITHOUT a transport, so they should pick up the first. |
| 01:33.52 | X-Rob | (?Or is that wrong?) |
| 01:34.05 | mjordan | that jives with my understanding |
| 01:34.15 | X-Rob | Well. it's not working. |
| 01:34.22 | mjordan | hm. What IP address is wrong? |
| 01:34.31 | X-Rob | See that paste above, with the codecs? |
| 01:34.39 | mjordan | ah |
| 01:34.41 | mjordan | in the SDP |
| 01:34.47 | mjordan | rtp_symmetric? |
| 01:34.54 | X-Rob | Ooh. |
| 01:34.58 | X-Rob | I think that's on all the time |
| 01:35.00 | X-Rob | but I'll check |
| 01:35.22 | mjordan | local_net is set? |
| 01:37.03 | X-Rob | mjordan: It's the 'match the first transport' thing that's derping. |
| 01:37.10 | X-Rob | I have this as a transport |
| 01:37.28 | X-Rob | https://www.irccloud.com/pastebin/EbWCz6ld |
| 01:37.45 | mjordan | hm. |
| 01:37.53 | mjordan | which transport is it using? |
| 01:37.54 | X-Rob | that's the only one |
| 01:37.57 | X-Rob | is there a way for me to find out? |
| 01:38.49 | mjordan | "pjsip show endpoint" or bust |
| 01:39.15 | X-Rob | <PROTECTED> |
| 01:39.18 | X-Rob | ^^ nothing |
| 01:39.30 | X-Rob | because of file's point that it should auto-detect and figure it out |
| 01:39.31 | mjordan | hm. |
| 01:39.36 | mjordan | yes, that's wrong |
| 01:39.48 | mjordan | I wonder if it's just the name that's derping it. |
| 01:40.08 | mjordan | what happens if you explicitly set the transport option on the endpoint? transport=0.0.0.0-udp |
| 01:40.13 | X-Rob | works |
| 01:40.14 | X-Rob | <PROTECTED> |
| 01:40.24 | mjordan | huh |
| 01:40.26 | mjordan | BUG! |
| 01:40.35 | X-Rob | as I don't allow trunks to auto detect |
| 01:40.39 | mjordan | no idea why that would happen |
| 01:41.36 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 01:46.15 | X-Rob | I don't have the multihomed patch in this build |
| 01:46.30 | X-Rob | but this new machine isn't multihomed. One IP, one default gateway, NAT. |
| 01:46.35 | X-Rob | as boring and standard as it could be |
| 01:48.55 | tm1000 | "mjordan: no idea why that would happen" <--user error..... |
| 01:48.59 | tm1000 | watches and waits |
| 01:51.08 | X-Rob | mjordan: You know how that 'local_net' is 192.168.5.0/24? Guess what my local network is. (I'll give you a hint. It's not 192.168.5.0/24) |
| 01:51.21 | X-Rob | Works really well now. |
| 01:53.30 | X-Rob | HOWEVER, the 'don't define a transport' thing doesn't work. |
| 01:55.58 | *** join/#asterisk-dev serafie (~erin@24.96.64.240) |
| 01:58.05 | file | it doesn't auto-detect it at configuration time |
| 01:58.10 | file | it's on a per-message basis |
| 01:58.24 | file | because until then, PJSIP has no idea where the message is going to go |
| 02:04.39 | X-Rob | Um. I haz a question |
| 02:04.45 | X-Rob | 302/302 Not in use 0 of inf |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.46 | X-Rob | <PROTECTED> |
| 02:04.47 | X-Rob | <PROTECTED> |
| 02:04.47 | X-Rob | <PROTECTED> |
| 02:04.54 | X-Rob | Shit. Sorry. meant to pastebin that. |
| 02:05.00 | snuff-work | mm pastebin |
| 02:05.06 | X-Rob | the 'Contact:' has a transport tag on it. wat. |
| 02:05.30 | X-Rob | 303 = my android phone, using sipdroid |
| 02:05.36 | X-Rob | also, I don't HAVE a transport called 'udp' |
| 02:06.17 | X-Rob | doing a 'pjsip show endpoint 303' says that it's using transport 0.0.0.0-udp (which is correct) |
| 02:10.14 | file | it's a SIP URI |
| 02:10.30 | file | SIP URIs have a parameter called "transport" which specifies the type of transport they want to be contacted over |
| 02:10.37 | X-Rob | ok. Just stressing slightly. Don't mind me. |
| 02:10.48 | X-Rob | I have managed to break EVERYTHING |
| 02:10.59 | X-Rob | (to do with NAT) |
| 02:11.11 | X-Rob | so I'm starting from the beginning and seeing where I've stuffed up |
| 02:16.42 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.221) |
| 02:20.43 | X-Rob | https://www.irccloud.com/pastebin/EPrClgok |
| 02:20.57 | X-Rob | file: can I borrow your brain for a second? |
| 02:21.29 | file | this late I'm barely awake |
| 02:21.40 | X-Rob | No worries then. |
| 02:22.17 | X-Rob | I appear to be losing the external_media_address=139.130.56.101 somehow |
| 02:22.24 | wedhorn | Hey snuff-work, being doing some fixups on skinny if you're interested (12 and trunk) |
| 02:22.33 | X-Rob | I'm now going to start hitting it with a hammer. |
| 02:34.15 | snuff-work | mm i am.. i did see a few of ur commits the other day when i decided to read through the last 20 days =) |
| 02:34.47 | snuff-work | just rather busy with other stuff but hopefully can give it a go sooner than later |
| 02:36.25 | wedhorn | okay, most of them are straight fixes for issues I've seen |
| 02:36.38 | wedhorn | i've got a phone on a bad network connection, which is great for finding issues |
| 02:40.47 | snuff-work | i can't remember 1/2 my issues.. |
| 02:41.33 | snuff-work | think the other main feature i kinda wanted back was the 'conf' button to be able to work similar to ccm |
| 02:41.48 | wedhorn | yep, know what you mean about the issues. |
| 02:42.05 | snuff-work | we can disable any key from being shown atm right? |
| 02:42.10 | wedhorn | ah, the conf yes... Were waiting for the new bridging... |
| 02:42.15 | snuff-work | aka hiding cfwd all |
| 02:42.39 | snuff-work | i can't remember if our masking stuff went all the way to the config file |
| 02:42.47 | wedhorn | yes from the code, don't think we can use the conf file though, but could be wrong |
| 02:43.14 | snuff-work | i think we were looking 'into' doing via config |
| 02:46.19 | wedhorn | looks like we've just got a const keymask |
| 02:46.27 | wedhorn | so no from the conf |
| 02:48.30 | snuff-work | mm np wasn't sure if we got that far |
| 02:49.16 | X-Rob | mjordan: if you could update https://reviewboard.asterisk.org/r/3102 with a comment that it breaks NAT, as it should then check against localnets to see if it should or shouldn't mangle the address, that would be cool. |
| 02:50.00 | file | it should alter it first, and then the NAT code should modify it again |
| 02:51.26 | X-Rob | ah. Well. it doesn't. |
| 02:51.41 | X-Rob | After swearing at nat for a while, I started doing a bisect. |
| 02:51.56 | file | nods |
| 02:52.16 | file | note added |
| 02:52.24 | X-Rob | <3 thanks, file |
| 02:54.20 | file | can't lift up head... too tired... |
| 03:30.50 | file | fixed |
| 03:31.35 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.213) |
| 03:40.05 | file | and now sleep |
| 03:49.55 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.213) |
| 04:15.47 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.213) |
| 04:28.27 | *** join/#asterisk-dev Fwny (~potato@192-0-250-120.cpe.teksavvy.com) |
| 05:34.51 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 06:49.55 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 07:10.12 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 07:55.27 | *** join/#asterisk-dev bulkorok (~Benjamin@85.183.61.47) |
| 08:22.12 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:e8ad:d6d3:a44c:51c2) |
| 08:43.27 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 08:47.43 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 09:23.10 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 09:23.48 | skrusty | morning |
| 09:37.01 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 09:37.01 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 09:43.47 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 10:08.56 | *** join/#asterisk-dev danjenkins (~dan@cpc65687-folk2-2-0-cust207.1-2.cable.virginmedia.com) |
| 10:17.46 | *** join/#asterisk-dev coreyf1513 (~cfarrell@2001:4830:117f:3:718f:e9f4:9f7b:8f6f) |
| 10:25.57 | *** join/#asterisk-dev danjenkins_ (~dan@nat/digium/x-jvcnyemoiajfhmjc) |
| 10:32.35 | *** join/#asterisk-dev drmessano (~nonya@pdpc/supporter/active/drmessano) |
| 10:52.33 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 11:16.08 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 11:18.24 | *** join/#asterisk-dev vlad_starkov (~vlad_star@195.218.175.34) |
| 11:20.16 | *** join/#asterisk-dev w9sh (~sph@50-79-224-193-static.hfc.comcastbusiness.net) |
| 11:28.31 | skrusty | quiet in here today |
| 11:52.52 | file | 'tis morning |
| 12:20.18 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 12:28.55 | *** join/#asterisk-dev mjordan (~matt@75.76.55.191) |
| 12:28.55 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 12:58.32 | *** part/#asterisk-dev mjordan (~matt@75.76.55.191) |
| 13:28.57 | *** join/#asterisk-dev protocoldoug (~doug@unaffiliated/protocoldoug) |
| 13:40.03 | *** join/#asterisk-dev ehsjoar (~jonas@c-24-9-87-13.hsd1.co.comcast.net) |
| 13:46.36 | skrusty | anyone about? |
| 13:47.38 | skrusty | just playing with ARI and wondering if i've found a bug, or something odd at my end, but calling PLay on a channel seems to have channelid and media the wrong way round. just checking my code, but so far seems to be sending the ARI uri correctly |
| 13:51.44 | opticron | looks |
| 13:52.32 | skrusty | hmm... |
| 13:52.48 | opticron | the code looks right to me |
| 13:52.53 | skrusty | well, two things, firstly maybe my fault, didn't realise i needed sound:<filename> |
| 13:53.01 | opticron | yes |
| 13:53.02 | skrusty | so my bad there |
| 13:53.10 | opticron | the sound: is very important |
| 13:53.15 | skrusty | but the output was what confused me (console): res_stasis_playback.c:350 play_on_channel: Attempted to play URI 'SIP/100-00000009' on channel 'demo-congrats' but scheme is unsupported |
| 13:53.24 | skrusty | implies that the filename is the channel |
| 13:53.27 | opticron | ah, the error message is backwards |
| 13:53.35 | skrusty | yeah |
| 13:53.44 | skrusty | i'll fix my code to add the schema now |
| 13:53.51 | opticron | I'll fix the error message |
| 13:53.58 | skrusty | nice one :) |
| 13:54.35 | skrusty | yay! :D |
| 13:54.47 | skrusty | first ARI demo working using AsterNET.ARI! |
| 13:54.55 | opticron | several appear to be backward... |
| 13:55.00 | skrusty | ah |
| 13:55.22 | skrusty | would never have noticed myself had i known about the schema |
| 13:55.22 | file | opticron, I wonder when that happened... |
| 13:55.58 | skrusty | does that mean there's support for remote media? like http://..../file.mp3 in ARI Play? |
| 13:56.10 | file | there can be in the future, but not yet |
| 13:56.14 | skrusty | got ya |
| 13:56.16 | skrusty | nice |
| 14:01.14 | opticron | skrusty, fixed |
| 14:01.20 | skrusty | nice :) |
| 14:05.04 | *** join/#asterisk-dev serafie (~erin@nat/digium/x-hhgifxmjqhsvzknz) |
| 14:12.33 | *** join/#asterisk-dev roderickm (~roderickm@67.63.143.254) |
| 14:33.37 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 14:33.37 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:44.22 | *** join/#asterisk-dev mjordan (~matt@nat/digium/x-uxcsrqcgtdrxdxwe) |
| 14:44.22 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:51.50 | skrusty | do sip channels support dtmf duration? just done a quick test of the dtmf event in ari and notice that duration is blank |
| 14:52.49 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:52.49 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:53.07 | mjordan | DTMF can be generated by a SIP channel in three different ways, so you'll have to be a bit more specific. In general, however, regardless of how a DTMF tone is generated, there should be a duration. |
| 14:54.40 | file | I just realized I could make a small module that reads in a config and makes extra codecs available to the Asterisk core |
| 14:55.18 | skrusty | mjordan: sorry for lack of specifics - not sure what method x-lite uses |
| 14:57.26 | mjordan | skrusty: it'd be determined by your sip.conf |
| 15:01.15 | skrusty | cheers, i think i found the problem anyway, i'll stop asking silly questions now ;) |
| 15:01.51 | file | if nobody asked silly questions nobody would ever learn |
| 15:03.00 | skrusty | ;) |
| 15:03.41 | skrusty | well, i think i have a working ARI framework up and running in c#! just need to write something now that uses it :] |
| 15:04.34 | mjordan | cool! |
| 15:04.44 | skrusty | https://asternetari.codeplex.com/ |
| 15:05.04 | skrusty | very early days, but can answer a call, play audio fiels, get dtmf events etc |
| 15:05.17 | skrusty | going to write a little wee mini conf bridge for fun i think :D |
| 15:05.51 | mjordan | that's a good place to start |
| 15:06.04 | file | yeah... feedback on what's missing for something like that would be nice |
| 15:06.10 | skrusty | sure |
| 15:06.28 | file | knows speaker detection is |
| 15:06.32 | skrusty | yeah |
| 15:06.38 | skrusty | we were talking about that the other day |
| 15:06.55 | skrusty | is that an "in progress" thing, or just on the cards? :) |
| 15:07.14 | mjordan | it's certainly doable. It hasn't been discussed much yet though |
| 15:07.23 | mjordan | one thing that I'm not sure of is if we need it outside of a multi-party bridge |
| 15:07.39 | mjordan | that is, is there value in speaker detection in only a two-party bridge? |
| 15:07.47 | skrusty | dunno... |
| 15:08.03 | skrusty | maybe, if you wanted to track who spoke in a call |
| 15:08.04 | skrusty | maybe? |
| 15:08.36 | skrusty | might be handy for automated asr or something? |
| 15:09.48 | mjordan | well, so that's why I'm not sure. Right now, bridge_softmix has the capabilities to do all that, but bridge_simple does not. That's partially on purpose, as the more you do with an audio frame in the bridging core, the less performance you're going to get. |
| 15:10.10 | mjordan | In a multi-party bridge, the usefulness is more obvious; in a two-party, I think we'd want to know what use case is driving it |
| 15:10.22 | skrusty | yeah |
| 15:10.52 | skrusty | i think it's going to be pretty niche |
| 15:11.32 | mjordan | luckily, if we decide we want to add it, it isn't hard. Talk detection is built into the bridging core, so it's more of a matter of just raising it in the mixing technologies |
| 15:12.32 | mjordan | and ARI would just have to add a talk detection hook onto the bridges it creates |
| 15:12.33 | skrusty | so how do you think it will be implemented for multi-party bridges? |
| 15:12.48 | skrusty | ah |
| 15:12.48 | mjordan | skrusty: you'll just get an event when someone starts/stops talking |
| 15:12.58 | mjordan | from the perspective of an ARI consumer. |
| 15:13.00 | skrusty | yeah sorry, i get that |
| 15:13.16 | mjordan | if you're curious, ARI would call ast_bridge_talk_detector_hook |
| 15:13.24 | skrusty | yes |
| 15:13.27 | skrusty | that's what i meant :) |
| 15:13.33 | mjordan | ConfBridge uses it |
| 15:13.38 | skrusty | got ya |
| 15:15.06 | *** join/#asterisk-dev kharwell (~kharwell@nat/digium/x-rnqhwnltfltivluu) |
| 15:19.45 | *** join/#asterisk-dev otherwiseguy (~otherwise@99-46-2-71.lightspeed.mssnks.sbcglobal.net) |
| 15:21.33 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.124.160) |
| 15:21.42 | *** part/#asterisk-dev mjordan (~matt@nat/digium/x-uxcsrqcgtdrxdxwe) |
| 15:38.04 | *** join/#asterisk-dev tgu (~tgu@host-212-68-194-46.brutele.be) |
| 15:38.30 | wdoekes | is this red stuff why I get spammed for "breaking the tests?" https://bamboo.asterisk.org/bamboo/browse/AST-DOC2-JOB1-104/log "No failed tests found, a possible compilation error. |
| 15:38.44 | *** join/#asterisk-dev mjordan2 (~matt@nat/digium/x-cryandavisqecqph) |
| 15:38.44 | *** mode/#asterisk-dev [+o mjordan2] by ChanServ |
| 15:39.09 | mjordan2 | jsmith: mind if I PM you? |
| 15:53.48 | *** part/#asterisk-dev bulkorok (~Benjamin@85.183.61.47) |
| 15:59.45 | *** join/#asterisk-dev infobot (~infobot@rikers.org) |
| 15:59.45 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Asterisk 12.0.0 now released. Do the release dance! -=- Follow on Twitter at @AsteriskDev |
| 16:14.08 | tzafrir | testing again the configuration from my demo: asterisk does not like span numbers >= 32 |
| 16:14.21 | tzafrir | (if they are pri) |
| 16:23.12 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 16:30.24 | *** join/#asterisk-dev gtjoseph (~gtjoseph@unaffiliated/gtj) |
| 17:14.06 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.124.173) |
| 17:24.27 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.124.173) |
| 17:32.07 | *** join/#asterisk-dev Pullphinger (~Pullphing@12.40.23.68) |
| 17:38.08 | *** join/#asterisk-dev seaux (~ocdcoder_@173-21-149-139.client.mchsi.com) |
| 17:38.59 | seaux | hey all, rbrindley/Ryan/awk/whatever-else-you-knew-me-as :) |
| 17:42.25 | seaux | anyway, I just added a CEL event for RINGING in channel.h. is there any interest in this getting back into Asterisk? (we use this CEL RINGING event to help detect PDD which is pretty useful for us to know) |
| 17:45.28 | pabelanger | seaux, you could put it up on reviewboard.asterisk.org |
| 17:46.21 | seaux | okay, i'll throw it up. it's only a 3 line patch so i was wondering if there was some reasoning why it wasn't an event yet. kinda figured it might have been intentionally left out |
| 17:47.42 | pabelanger | not sure, CELGenUserEvent(RINGING) is likely a workaround. Depending on how you did dialplan |
| 17:48.45 | seaux | GUE wouldn't work for me. I'm dialing two SIP channels in Dial and want to know when they get set to RINGING |
| 17:49.11 | seaux | i basically just added a cel event after the ami Newstate event when the state is set to RINGING |
| 17:49.22 | seaux | also, my reviewboards account is inactive ;( |
| 17:50.40 | pabelanger | nods |
| 17:50.53 | pabelanger | Toss it up, and people with review. Worst case, they say no |
| 17:51.24 | seaux | yea |
| 17:53.41 | seaux | who manages reviewboard? I need someone to change my email address |
| 17:54.03 | seaux | lol nvm *facepalm* |
| 18:07.49 | *** join/#asterisk-dev danjenkins (~dan@cpc65687-folk2-2-0-cust207.1-2.cable.virginmedia.com) |
| 18:09.42 | Qwell | throws a taco at seaux |
| 18:11.08 | seaux | yumm |
| 18:11.32 | seaux | who did the new event/state code? i like. |
| 18:15.00 | file | stasis? leedm777 |
| 18:16.00 | seaux | idk, i'm out of the loop. but in 11 I had to add the event in channel.c, not it seems there are monitors/handlers and it just has to be added to cel.c which is really nice. |
| 18:16.11 | seaux | s/not it/now it/ |
| 18:40.40 | mjordan2 | seaux: is your e-mail address still at 256labs/ |
| 18:40.51 | X-Rob | I have a crazy arse codec issue. |
| 18:40.53 | X-Rob | http://pastebin.com/0k6uHDjj |
| 18:41.12 | X-Rob | about line 135 onwards |
| 18:41.14 | seaux | mjordan2: i just changed it. it was my digium.com email and i couldn't find the link to edit my profile |
| 18:41.58 | datachomper | I havn't played ping ping since seaux was at digium :-( |
| 18:43.05 | opticron | whoa...I didn't realize that seaux was RB |
| 18:43.21 | seaux | yea :) i have identity crisis |
| 18:43.36 | file | why the heck is codec_gsm in there... |
| 18:43.57 | seaux | datachomper: man, that was in 2010 O.o |
| 18:44.21 | seaux | stares at his dusty ping ping paddle. |
| 18:44.31 | *** join/#asterisk-dev vlad_starkov (~vlad_star@109.188.126.230) |
| 18:44.35 | opticron | yeah, it's been a while |
| 18:44.53 | file | X-Rob, the SIP signaling and core debug would potentially be useful there... |
| 18:45.23 | seaux | file: i read u media blog post. |
| 18:45.40 | file | seaux, people read those things? |
| 18:46.58 | seaux | file: apparently. i thought it was a marketing blog about Asterisk in the media. you tricked me. |
| 18:47.00 | X-Rob | file: I ... THINK it's something to do with the call coming in via a local channel, but I'm lost. I'll try and nail it down to a small, repeatable, case. SIP Signalling was all G722, it was working fine. And if I join the conf normally,it works fine too. |
| 18:47.19 | file | waves fist at Local channels |
| 18:48.28 | mjordan2 | seaux: did you get logged in? Or do you need it reset? |
| 18:48.36 | X-Rob | also, issue does not appear when using alaw/ulaw. |
| 18:51.23 | *** join/#asterisk-dev vlad_starkov (~vlad_star@91.206.59.129) |
| 18:52.03 | seaux | mjordan2: reviewboard infiltration was success :) |
| 18:52.10 | mjordan2 | excellent |
| 18:52.34 | mjordan2 | seaux: do you have a new license contributor agreement signed? |
| 18:52.52 | seaux | prob not, haven't signed anything since 09 |
| 18:57.54 | mjordan2 | seaux: mind doing that? It'll be necessary before anything could be accepted |
| 19:00.20 | seaux | ofc |
| 19:07.08 | file | opticron, hi. |
| 19:08.10 | opticron | file, hello |
| 19:08.31 | file | opticron, any progress on le Grandstream insanity? |
| 19:08.47 | file | (just curious) |
| 19:09.02 | file | ((also avoiding rewriting ast_format_cap)) |
| 19:09.07 | opticron | still waiting on spitts to get the net setup up and running to get a debug trace from functional phones |
| 19:09.10 | file | ah |
| 19:09.15 | opticron | s/net/nat/ |
| 19:09.56 | file | carry on |
| 19:10.04 | opticron | do I have to? |
| 19:10.18 | file | yes |
| 19:10.23 | opticron | D: |
| 19:25.14 | *** join/#asterisk-dev sgriepentrog (~sgriepent@nat/digium/x-hrxgriphxpoylufe) |
| 19:46.44 | *** join/#asterisk-dev otherwiseguy (~otherwise@2605:a601:437:ce01:2677:3ff:fe8f:2730) |
| 19:48.54 | seanbright | file: will your media framework changes clean up some of the special-casing of various formats in the core? lines like "return (format->id == AST_FORMAT_G722) ? 8000 : ast_format_rate(format);" |
| 19:50.13 | file | there will be built-in codecs, but all of that logic will be outside the media core itself |
| 19:50.59 | seanbright | cool |
| 19:51.03 | seanbright | that's good news |
| 19:51.23 | seanbright | the opus and vp8 passthrough patches were fairly intrusive |
| 19:51.56 | file | when I'm done adding a basic codec and format will involve writing a module and not touching the core, OR, if I/someone else gets around to it... defining it in a .conf file |
| 19:52.04 | seanbright | it would be nice to just drop in a .c file and compile rather than to have to patch a slew of other files |
| 19:52.15 | seanbright | ohh |
| 19:52.17 | file | if you have to do attribute negotiation then a module will be required |
| 19:52.18 | seanbright | that'd be fancy |
| 19:54.34 | file | seanbright, I expect a patch by Friday |
| 19:54.41 | seanbright | heh |
| 19:54.51 | coreyf1513 | what do you guys recommend for profiling asterisk? I seem to remember some kind of problem with asterisk in gprof. |
| 19:55.02 | file | actually I think my branch is to the point where you could... do... that... |
| 19:55.05 | seanbright | oprofile? |
| 19:55.23 | file | coreyf1513, I've used oprofile, gprof, and valgrind |
| 19:56.08 | mjordan2 | same here as what file said. It's usually a mixed bag, depending on what you want to look at |
| 19:56.44 | mjordan2 | once we had knocked off all the low hanging fruit, the format stuff showed up at the top of just about everything we profiled in just about any scenario |
| 19:56.56 | coreyf1513 | so can the valgrind results be trusted? I mean since valgrind has harsh effects on performance |
| 19:57.03 | mjordan2 | and that's always the kicker too - profiling a SIP call with little to no media looks a lot different |
| 19:57.26 | mjordan2 | coreyf1513: not from the perspective of 'how many calls can I pump through'. But valgrind is accurate with respect to what in your code is slowing you down |
| 20:00.20 | seanbright | file: will passthrough stuff still need to be handled specially? |
| 20:00.28 | seanbright | or will that be configured via .conf? |
| 20:00.45 | file | seanbright, I'm not saying it will be configured... I'm saying what I'm writing could do it :P |
| 20:00.57 | seanbright | gotcha |
| 20:01.38 | file | seanbright, *something* tells the core about a codec, and that *something* can also create a cached format for configuration purposes, and can also improve the RTP stack how to know about it |
| 20:01.39 | seanbright | but when you're done, i could drop in a codec_patented_codec_that_i_want_only_passthrough_for.so and be good to go? |
| 20:01.48 | file | yes. |
| 20:01.51 | seanbright | ok |
| 20:01.54 | seanbright | then i am contented |
| 20:02.15 | file | we could also write a module that is the *something* and constructs the various things based on values in a .conf file |
| 20:02.44 | seanbright | ah |
| 20:02.48 | seanbright | very meta |
| 20:04.13 | file | seanbright, like I said... I expect a patch by Friday |
| 20:04.38 | file | (seriously, it would be good because it would require me to document out things further for someone else :P) |
| 20:05.29 | seanbright | once i am done all of my work that involves _using_ asterisk, i'll be happy to add some work _on_ asterisk |
| 20:05.36 | seanbright | until then, i thank you for everything you're doing |
| 20:05.37 | seanbright | :-) |
| 20:05.53 | file | useless! |
| 20:06.42 | file | seanbright, but fine :P |
| 20:07.10 | coreyf1513 | I have a patch that allows a workaround for asterisk 11 ast_format_cmp performance. the patch allows #define WITHOUT_FORMAT_ATTR_INTERFACE for people who don't use res_format_attr_*.so. should I open my own issue for this? |
| 20:11.47 | pabelanger | coreyf1513, opening an issue on the tracker or getting into a SVN team branch never hurts |
| 20:11.56 | pabelanger | even if the issue gets closed, the patch is up |
| 20:12.41 | pabelanger | I havent' really been following the 11 performance stuff, what sort of hit are we talking about? |
| 20:13.52 | coreyf1513 | ast_format_cmp does an ao2_find to find if an attribute interface exists (same as 12 I think). this lookup is very expensive and ast_format_cmp is run a lot |
| 20:14.28 | pabelanger | coreyf1513, what sort of load difference do you get after you apply your patch? |
| 20:14.45 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-218-28-58.cablep.bezeqint.net) |
| 20:14.45 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 20:16.10 | file | that's only part of the performance penalty |
| 20:16.16 | file | copying formats is also expensive |
| 20:16.17 | coreyf1513 | pabelanger: I tested with ConfBridge and 2 SIP channels from a separate PBX (60 second calls). the bridge thread spent 10% of the time in ast_format_cmp, each channel thread spent 25% of the time (no transcoding in this test) according to callgrind |
| 20:17.55 | coreyf1513 | pabelanger: with the patch and the define ast_format_cmp is less than 1% of those threads |
| 20:18.15 | file | yup |
| 20:18.49 | pabelanger | okay, significant then |
| 20:18.52 | coreyf1513 | file: I'm aware the format copying is expensive, but nothing can be done about that in asterisk 11 (reducing the struct size would break ABI). also comparing was way more expensive |
| 20:32.19 | *** join/#asterisk-dev vlad_starkov (~vlad_star@91.206.59.131) |
| 20:40.23 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@216-191-106-163.dedicated.allstream.net) |
| 20:55.28 | *** join/#asterisk-dev vlad_sta_ (~vlad_star@109.188.126.99) |
| 21:22.55 | *** join/#asterisk-dev vincentroy (~vincentro@vroy.ca) |
| 21:25.43 | mjordan2 | coreyf1513: I think having an issue/patch is fine, if for no other reason than it makes the issue known and gives people a solution |
| 21:26.00 | mjordan2 | coreyf1513: you could also have a subversion branch that would get auto-merged from 11 |
| 21:26.06 | mjordan2 | if you wanted, that is |
| 21:29.26 | coreyf1513 | mjordan2: I opened ASTERISK-23179 for the issue that way other people can see the patch |
| 21:30.31 | mjordan2 | thanks |
| 21:33.18 | coreyf1513 | mjordan2: I'm somewhat interested in an automerge branch, but I've never had good luck with svn merge and I'm afraid I'll just break something |
| 21:34.03 | mjordan2 | k - if you change your mind, let me know. I don't mind setting you up with a cert if you feel like it's worth it some day |
| 21:34.22 | mjordan2 | lord knows you've been pushing lots of patches up :-) |
| 21:34.35 | Qwell | coreyf1513: worst case scenario, you break one of your own branches |
| 21:34.52 | Qwell | (assuming you'd prefer to just not have branch/trunk commit access) |
| 21:38.12 | coreyf1513 | if I can be setup without commit access to trunk/branches that would probably be helpful |
| 21:38.44 | Qwell | Speaking for mjordan2, with absolutely zero authority: It's easy. :) |
| 21:39.34 | Qwell | Setting up branch access is actually one step beyond a team branch. |
| 21:43.37 | coreyf1513 | ok so you've changed my mind :) what do you guys need from me to set that up? |
| 21:45.04 | mjordan2 | coreyf1513: nothing, just shoot me an e-mail with the e-mail address you want |
| 21:45.13 | mjordan2 | I can get it set up tomorrow |
| 21:59.41 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 21:59.42 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 22:36.13 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 22:59.10 | *** join/#asterisk-dev roderickm (~roderickm@99-124-89-102.lightspeed.dctral.sbcglobal.net) |