| 16:33.52 | *** join/#asterisk-dev infobot (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 16:33.52 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- IAX2/guest@misery.digium.com/asterisk-dev -=- Asterisk trunk: Now containing 235689464 uses of the letter c! -=- Tier 2 and Tier 3.14159 support is in #asterisk -=- ast_atomic_fetchadd_int(&developers_engaged, +1); -=- This is still not tier 2 support |
| 16:33.54 | _mihai_ | russellb: dude, are you on vacation? |
| 16:33.56 | TimRiker | infobot: save |
| 16:33.57 | infobot | saved user and chan files |
| 16:34.18 | russellb | ooh |
| 16:34.29 | russellb | _mihai_: i'm at a training course for the week .. |
| 16:34.46 | russellb | TimRiker: thanks :-D |
| 16:34.54 | TimRiker | russellb: sure. np. |
| 16:35.15 | Qwell[] | infobot is such a nub :p |
| 16:35.15 | _mihai_ | russellb: ah, ok... when you get back, you might want to take a look at #10289 |
| 16:35.17 | Qwell[] | jbot++ |
| 16:35.22 | Qwell[] | </troll> |
| 16:35.31 | TimRiker | ~lart Qwell |
| 16:35.31 | infobot | strangles Qwell with a 9-pole serial cable |
| 16:35.36 | Qwell[] | TimRiker: thanks :) |
| 16:35.41 | Qwell[] | (not for the lart) |
| 16:35.42 | russellb | _mihai_: will do ... i think i saw it, just haven't had time to think about it |
| 16:35.44 | TimRiker | hehehe |
| 16:35.47 | russellb | infobot: who's your daddy? |
| 16:35.48 | infobot | YOU are, Mr Sexy Pants! |
| 16:36.02 | Qwell[] | ~timriker |
| 16:36.03 | infobot | it has been said that timriker is my owner http://rikers.org/ mailto:Tim@Rikers.org mailto:TimR@Debian.org maintainer of BZFlag, member of a ton of open source projects http://www.advogato.com/person/timriker/ http://sourceforge.net/users/timriker/ the guy who GPL'd SCO's ABI files, giving every Linux user the right to use them ;-), or a very cool guy. |
| 16:36.28 | *** join/#asterisk-dev Corydon76-work (n=tilghman@pdpc/supporter/sustaining/Corydon76-home) |
| 16:36.28 | *** mode/#asterisk-dev [+o Corydon76-work] by ChanServ |
| 16:36.50 | putnopvut | M10199 |
| 16:36.51 | MuffinMan | [feedback] [Asterisk] Applications/app_queue 0010199: Random replacement of channel name with other text in queue log entries reported by jfitzgibbon http://bugs.digium.com/view.php?id=10199 |
| 16:36.58 | _mihai_ | russellb: we're getting close to figuring out the root cause of vnak storms.... |
| 16:37.21 | putnopvut | If someone doesn't mind having a look at the backtraces for 10199, it looks like the crash is happening in a place it definitely SHOULDN'T be. |
| 16:37.39 | putnopvut | Does anyone think this is attributable to the fact safe asterisk is being used? |
| 16:38.18 | russellb | putnopvut: it happens. no, that shouldn't change things |
| 16:38.54 | Corydon76-work | putnopvut: No, it's fairly obvious why it's crashing |
| 16:38.56 | Corydon76-work | NULL pointer deref |
| 16:39.08 | Corydon76-work | Could you update the patch to current 1.4? |
| 16:39.14 | putnopvut | Yeah, I'll do that. |
| 16:39.20 | Corydon76-work | I get 4 conflicts when attempting to apply the patch |
| 16:39.34 | putnopvut | But the NULL pointer deref doesn't make sense to me though. |
| 16:40.08 | Corydon76-work | I'll take a look at it after you get the patch updated |
| 16:41.12 | Juggie | Corydon76-work, whats your day job? |
| 16:41.42 | Corydon76-work | Well, the job I'm coming off right now is lead programmer for a Digium reseller |
| 16:42.28 | Juggie | ya i know your moving to digium |
| 16:42.32 | Juggie | i was just wondering what you did before |
| 16:42.40 | Corydon76-work | That's it. |
| 16:43.07 | Corydon76-work | Before this, I was the DBA for the 3rd largest online bookseller in the country |
| 16:43.55 | Juggie | fun, today my day is filled with db problems |
| 16:44.22 | Corydon76-work | It's why I tend to be a MySQL bigot... |
| 16:44.45 | Juggie | this is mssql :( |
| 16:45.10 | Corydon76-work | I've worked against mssql before |
| 16:45.25 | Juggie | yeah... 2 nights ago in the middle of the night my replication just hosed itself |
| 16:45.36 | Corydon76-work | Ouch... |
| 16:45.37 | Juggie | for no reason you know, just for fun |
| 16:45.43 | Qwell[] | it's mssql |
| 16:45.49 | Qwell[] | that's plenty enough reason |
| 16:45.51 | Juggie | of course, its a 200million row table |
| 16:46.06 | Corydon76-work | Yick |
| 16:46.24 | Juggie | id just like to state for the record, its not my design :) |
| 16:46.25 | russellb | my tablez are bigger than yourz |
| 16:47.06 | Corydon76-work | I preferred my tables for sales data... 700000 rows, 272 columns |
| 16:47.24 | putnopvut | 272 columns....yay!!! |
| 16:47.42 | Corydon76-work | putnopvut: isbn and one column for each day of the year |
| 16:48.05 | Corydon76-work | Then I could yank the 90day sales report with a single query |
| 16:48.11 | Corydon76-work | single row, even |
| 16:48.29 | *** join/#asterisk-dev jefforulez (n=jeffo@38.96.187.252) |
| 16:48.37 | Corydon76-work | even better, it was FAST |
| 16:48.59 | Corydon76-work | which, when you're doing sales predictions for 700,000 items was critical |
| 16:49.19 | putnopvut | Right. Maybe my math's off, but why only 271 days represented? |
| 16:49.23 | Juggie | Corydon76-home, i do agent predictions. |
| 16:49.27 | Juggie | similar. |
| 16:49.31 | Qwell[] | TimRiker: jbot factoids? :p |
| 16:49.33 | Corydon76-work | putnopvut: one for each day of the fiscal year |
| 16:49.45 | Corydon76-work | Err, sorry, 371 |
| 16:49.55 | putnopvut | Ah, makes more sense now. |
| 16:50.13 | Corydon76-work | 53 * 7 |
| 16:50.53 | Corydon76-work | Most of the time you'd have 52 week years, but every once in a while, you get a 53 week year |
| 16:51.27 | Corydon76-work | Ah, the things we did to optimize speed... |
| 16:53.11 | TimRiker | Qwell[]: apt/ibot/infobot/jbot/purl are all the same bot. same process multi-attached. |
| 16:53.32 | Qwell[] | I mean the 780 million row oracle db |
| 16:53.40 | TimRiker | oh, you mean the oracle? no, no. :) |
| 16:54.44 | russellb | lol .. |
| 16:55.05 | TimRiker | family history database actually. and the 780 million does not count "forwarded" old entries. If I did an actual row count it's likely about 1.4 trillion iirc. |
| 16:58.41 | TimRiker | ~status |
| 16:58.56 | infobot | Since Tue Jul 24 19:03:47 2007, there have been 9 modifications, 167 questions, 0 dunnos, 0 morons and 143 commands. I have been awake for 21h 54m 54s this session, and currently reference 113280 factoids. I'm using about 19000 kB of memory. With 0 active forks. Process time user/system 1584.88/118.69 child 0/0.01 |
| 16:58.59 | Qwell[] | TimRiker: surely the 780 million records isn't just your family? |
| 16:59.00 | TimRiker | well, that depends on your definition of family. :) it's global. |
| 17:01.51 | TimRiker | backend for the new familysearch.org |
| 17:01.51 | Qwell[] | neat |
| 17:34.38 | *** join/#asterisk-dev IPmonger (n=ipmonger@c-71-224-31-246.hsd1.pa.comcast.net) |
| 17:44.44 | *** part/#asterisk-dev IPmonger (n=ipmonger@c-71-224-31-246.hsd1.pa.comcast.net) |
| 17:57.33 | *** join/#asterisk-dev blitzrage (n=blitzrag@asterisk/documenteur-extraordinaire/blitzrage) |
| 18:06.06 | *** join/#asterisk-dev ctaloi (n=ctaloi@nat-66-218-1-47.usadatanet.com) |
| 18:30.58 | *** join/#asterisk-dev bkuhn (i=bkuhn@sflc/staff/conservancy.president.bkuhn) |
| 18:32.03 | Corydon76-work | Juggie: ping |
| 18:32.32 | Simon-- | _mihai_: there? |
| 18:32.33 | Simon-- | doh |
| 18:32.38 | Simon-- | damnit |
| 18:33.06 | Simon-- | russellb: one storm again this morning, even with chan_iax2 patches from 1.4.9... |
| 18:34.32 | jefforulez | Simon: do you have tcpdumps from your vnak storm? |
| 18:34.41 | jefforulez | specifically, from right before it happens? |
| 18:35.13 | jefforulez | do you know how many active channels you had when it happened? |
| 18:35.28 | jefforulez | i.e. was it load-driven? or (bad) luck of the draw? |
| 18:40.11 | *** join/#asterisk-dev CunningPike (n=arodgers@209.17.159.211) |
| 18:40.58 | Simon-- | jefforulez: I do, yes, but it has been difficult to work with them because our normal IAX load is about 1 GB/hour |
| 18:41.12 | jefforulez | right. :) |
| 18:41.31 | jefforulez | i've been splitting ours up into chunks by re-running them through tcpdump |
| 18:41.33 | Simon-- | luckily a recompiled tcpslice seems to work |
| 18:41.36 | Simon-- | ahh |
| 18:41.52 | jefforulez | tcpdump -s 2048 -C 5 -r in.pcap -w out.pcap |
| 18:42.05 | jefforulez | gives 5MB chunks. |
| 18:42.06 | Simon-- | tcpslice at least lets you specify bytime.. |
| 18:42.15 | jefforulez | ah. cool. good to know! |
| 18:42.17 | Simon-- | and tshark -b lets you record by time, too, which is nice |
| 18:42.37 | Simon-- | (auto file split during capture) |
| 18:42.52 | jefforulez | are you seeing the same pattern at the beginning of the storm from v1.4.8? |
| 18:42.56 | Simon-- | are you seeing the VNAK storming as well? |
| 18:43.16 | jefforulez | yeah. i work with mihai. i think you chat'd with him, right? |
| 18:43.36 | Simon-- | went from 1.4.7.1 to 1.4.8+changes from trunk, haven't tried 1.4.8, but have seen this happening since moving from 1.0 to 1.4.3ish |
| 18:43.45 | Simon-- | tried to message him but I keep missing him :) |
| 18:44.02 | Simon-- | just appended a note to 10289 ... |
| 18:44.25 | jefforulez | k. i'll take a look. |
| 18:44.33 | Simon-- | anyway, there was one storm this morning running 1.4.8+plus his patches as of yesterday, but I wasn't capturing at the time |
| 18:44.53 | Simon-- | it seems to come in weird waves where one day will be fine, the next will have 15 storms.. |
| 18:45.20 | jefforulez | yeah. i keep thinking this is a lunar problem. :) |
| 18:45.25 | Juggie | Corydon76-home, you rang? |
| 18:45.44 | Juggie | er, Corydon76-work |
| 18:45.49 | Simon-- | it actually started storming on a call I was on the other day. voice continued, just choppy as it saturated an upsteam openbsd vpn link |
| 18:46.07 | jefforulez | yup. seen that behavior here too. |
| 18:46.25 | jefforulez | server impact seems to be dependent on the vnak'ing client. |
| 18:46.26 | Corydon76-work | We need to talk about the API you exposed |
| 18:46.53 | jefforulez | unfortunately we have a client which runs locally on the server. when it vnaks, all hell breaks loose! |
| 18:47.07 | Juggie | Corydon76-work, ok, whats up? |
| 18:47.28 | Corydon76-work | If you're going to expose an API, it needs to start with the prefix ast_agi_ |
| 18:47.29 | jefforulez | so we're adding some more patches which clear out the retransmission queue when bad frames are detected. |
| 18:47.39 | Juggie | o, the, cli function |
| 18:47.40 | Juggie | ok, np.. |
| 18:47.51 | Corydon76-work | I'd go with ast_agi_fdprintf |
| 18:48.01 | jefforulez | and we're still looking into figuring out why bad frames get left around in the transmission queue in the first place. |
| 18:48.02 | Juggie | Corydon76-work, ok, anything else? |
| 18:48.05 | Corydon76-work | and change everything to match that API name |
| 18:48.09 | Simon-- | jefforulez: did you find a way to reproduce it in any reliable fashion? I have not yet been able to reproduce it between two test boxes. I was simulating lag, loss, corruption, and out of order packets with Netem |
| 18:48.17 | Corydon76-work | Nope, just a policy problem |
| 18:48.22 | Corydon76-work | Have you tested yet? |
| 18:48.24 | jefforulez | we've also talked about changing iaxclient to allow a limit on the number of times it'll vnak a frame. |
| 18:48.39 | Juggie | Corydon76-work, yep. not alot but some. |
| 18:48.52 | Juggie | i modded app_record to register 'record' |
| 18:49.00 | Corydon76-work | Okay |
| 18:49.07 | jefforulez | not yet. mihai seems to have reproduced it on this box, but i've not been able to myself. |
| 18:49.22 | Corydon76-work | Juggie: can you invoke that command then? |
| 18:49.41 | jefforulez | if you can filter your tcpdump and get the beginning of the call that caused the storm, that'd be great. |
| 18:49.57 | jefforulez | we want to look at the setup frames and then what's getting vnak'd |
| 18:50.11 | jefforulez | see if it's the same bad frame data we're seeing here. |
| 18:50.24 | Juggie | i believe so yes.. it shows up in the 'agi show' i dont remember if i tested a actual record on the final code i gave you but it did work yeah |
| 18:50.56 | Juggie | i'll check it all again when i upload a patch with the changed function name |
| 18:51.24 | Juggie | but code wise theres no reason it should make a difference if the function is in res_agi or somewhere else |
| 18:51.28 | Juggie | so it should be no problem |
| 18:51.42 | Simon-- | jefforulez: I have a suspicion it relates somehow to the very basic issue I reported here: http://bugs.digium.com/view.php?id=10232 ... that bug would indicate very serious issues with the transmission stage |
| 18:54.03 | Simon-- | the ->POKE <-PONG ... <-PONG ->INVAL shows to me that something is broken at the transmit stage, because the ACK never goes out but it did process the original PONG |
| 18:54.40 | Juggie | Corydon76-work, thanks for reviewing it, i'll do some work on it tomorow, right now i have to run and grab a shower and catch a flight in 2hours. |
| 18:55.11 | jefforulez | reading 10232 now... |
| 18:57.48 | jefforulez | Simon: interesting. |
| 18:58.16 | jefforulez | Simon: do you see the same problem in v1.2.x? |
| 18:58.56 | Simon-- | jefforulez: went from 1.0 to 1.4, so not tried... |
| 19:02.39 | jefforulez | Simon: do you think you could put together a tcpdump with the call which caused the storm this morning? |
| 19:03.04 | Simon-- | jefforulez: not without a time machine :) I'll start another trace and see if it happens again.. |
| 19:03.32 | jefforulez | oh. your dump starts after the storm started? |
| 19:03.35 | Simon-- | ah damnit, just missed another one |
| 19:03.50 | Simon-- | yes |
| 19:04.00 | jefforulez | sorry, i misunderstood previously. |
| 19:04.19 | Simon-- | well, I do have traces before but that was without the signed wrap corrections.. |
| 19:08.05 | jefforulez | k. i'm working on putting our latest v1.4.9 on a used machine and then setting up a rotating tcpdump so we can get more data. |
| 19:08.09 | jefforulez | i'll ping you when i have more. |
| 19:08.20 | jefforulez | and please ping back if you get a new dump./ |
| 19:08.36 | jefforulez | i'd be very curious to see the frame data that's causing the storm here. |
| 19:09.24 | jefforulez | Simon: btw, did you report before that v1.4.9 fixed the problem? mihai told me someone had said there here... |
| 19:12.51 | Simon-- | jefforulez: will do. btw, I am using tshark -i eth0 -s 8192 -b filesize:10240 -w iax -n udp port 4569 |
| 19:13.01 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-82-81-220-132.cablep.bezeqint.net) |
| 19:13.26 | *** join/#asterisk-dev JerJer (n=PhatJ@pdpc/supporter/bronze/jerjer) |
| 19:13.26 | *** mode/#asterisk-dev [+o JerJer] by ChanServ |
| 19:14.25 | Simon-- | jefforulez: that bug I reported? that was right when 1.4.8 was released. but the code changes since only relate to VNAKs, so it hasn't changed.. |
| 19:14.44 | JerJer | inbound sip call from Level3... their lovely little document says they expect a 100 and a 183 |
| 19:14.52 | JerJer | but asterisk sends a 180 |
| 19:15.09 | JerJer | isn't 183 supposed to be for the other direction - sip call going out ? |
| 19:15.35 | jsmith | JerJer: Nope... 183 is usually just like a 180, only it has SDP with it (usually) |
| 19:15.46 | jsmith | JerJer: It's a "here, I'm ringing, and here's some early media" |
| 19:16.07 | jsmith | JerJer: A 180 without SDP says "Here, I'm ringing... you generate ringback tone" |
| 19:16.16 | JerJer | i don't see any way to tell asterisk we want a 183 versus 180 then |
| 19:17.06 | jsmith | The m option to Dial() is one way to do it. |
| 19:19.58 | jcmoore | otherwise you can hack chan_sip to send both |
| 19:25.29 | oej | 183 doesn't mean "ringing" |
| 19:25.32 | oej | 180 is ringing |
| 19:25.37 | oej | 183 is early media on the way |
| 19:25.46 | oej | might be a prelude to a cancel or a bye |
| 19:25.55 | oej | or just inband indications |
| 19:25.59 | oej | 180 may have an sdp too |
| 19:26.04 | oej | confusing enough |
| 19:27.41 | jcmoore | huh? sip isn't confusing. :-) |
| 19:27.56 | jsmith | jcmoore: Nope... it's just the *implementations* that are. |
| 19:28.10 | jsmith | oej: Thanks for that clarification... that's what I was trying to say, I just didn't do a very good job |
| 19:29.37 | kpfleming | 180 with SDP? that hurts my brain |
| 19:29.54 | oej | yes, it's bad. But we do support it. |
| 19:30.23 | jsmith | kpfleming: You'd be surprised how many times I find it. |
| 19:30.26 | Juggie | oej, i'm going on vacation but i might play with enabling minivm w/ the event system while i am there if i get bored. |
| 19:30.36 | oej | There was a case with an asian gw a time ago where we had to implement it |
| 19:30.40 | oej | juggie: thanks. |
| 19:30.54 | jsmith | oej: Speaking of minivm, how does VM notification work in minivm? Just curious... |
| 19:31.03 | oej | Doesn't work |
| 19:31.17 | Juggie | oej, me and Corydon76-home have also commited a ton of updates to res_agi, some which might intreast you for minivm |
| 19:31.22 | oej | Minivm was designed only for sending by e-mail at start |
| 19:31.22 | jsmith | OK, that's why I couldn't find anything :-) |
| 19:31.31 | jsmith | oej: I see |
| 19:31.32 | oej | but you can easily add notification as a separate app |
| 19:31.48 | Juggie | if you wanted to expose a proper interface to minivm over agi |
| 19:31.52 | oej | everything in minivm needs to be small buildingblocks so people can decided themselves about notification |
| 19:32.39 | Juggie | oej, its hard to make mwi totally seperate because the channel is going to ask for it if say a sip phone registers |
| 19:33.09 | oej | juggie: Yes, but you can notify the vent system with an app and thus enable the channel notification |
| 19:33.12 | Juggie | well, not necessairly ask for it, but it will want to see it via events |
| 19:33.30 | Juggie | right, but you woudnt want to send the notify ONLY via the dialplan |
| 19:33.51 | oej | as Mark says: It's only software, Anything is possible. |
| 19:33.53 | oej | :-) |
| 19:33.54 | Juggie | in that case if you ever restarted a phone it woudnt know its status until someone left him another vm and you called the notify app via the dialplan |
| 19:34.22 | Juggie | i entirely agree with what you are going for, but some things need to be a little bit hooked in |
| 19:34.30 | oej | juggie: the channel can query and we'll check the repository. I don't know the details on mr Bryants new system, but I'm sure you can solve that case too. |
| 19:35.05 | Juggie | oej, exactally thats all i am saying, minivm has to be able to put events queue without being told via the dialplan |
| 19:35.20 | Juggie | anyways i have a plane to catch so i must go, have a great evening. |
| 19:35.33 | oej | have a nice flight |
| 19:35.49 | Juggie | thanks, i have to attend 3 weddings in 1.5 weeks :( |
| 19:36.10 | oej | juggie: You can reuse 90% of your speech at the dinners. |
| 19:36.32 | oej | file: What happened with the muffin man? Turned gorilla? |
| 19:37.57 | Corydon76-work | oej: have you seen the muffin man? |
| 19:38.15 | oej | Corydon: "THE Muffin Man" ? |
| 19:38.25 | oej | Corydon: I need some ginger candy |
| 19:39.20 | Corydon76-work | oej: Sorry, you probably don't have the tune circulating in your head, do you? |
| 19:40.47 | *** part/#asterisk-dev TimRiker (n=timr@pdpc/supporter/bronze/TimRiker) |
| 19:41.13 | Corydon76-work | oej: http://blackdog4kids.com/holiday/summer/do/songs/muffin-p.html |
| 19:53.25 | file | phone call dragged me away! |
| 19:53.39 | file | oej: the muffin man is MIA |
| 19:53.43 | blitzrage | unavoidable? |
| 19:53.44 | file | suspected alien abduction |
| 19:54.22 | oej | I guess MuffinMan had a Canadian passport and was refused entrance to the US |
| 19:54.43 | blitzrage | sounds about right |
| 19:56.45 | oej | Going on holiday again for two weeks, so I won't be online or read much e-mail... |
| 19:57.01 | blitzrage | holiday again?! |
| 19:57.04 | file | oej: where are you going?!? |
| 19:57.12 | oej | file: away |
| 19:57.12 | blitzrage | I've only had 1.5 weeks of holiday in the last 3 years |
| 19:57.17 | blitzrage | I don't take nearly enough time off :) |
| 19:57.20 | file | how... mysterious |
| 19:57.21 | oej | blitzrage: we have five weeks a year by law |
| 19:57.48 | oej | blitzrage: I only use three... |
| 19:57.48 | blitzrage | oej: I know... lazy europeans :) |
| 19:57.48 | oej | blitzrage: Lazy by law |
| 19:57.48 | file | lazy canadian. |
| 19:57.54 | blitzrage | ahhh... in the summer I take a lot of long weekends |
| 19:57.57 | blitzrage | I guess it probably works out |
| 19:58.25 | oej | I am going to discover some of Slartibartfast's work |
| 19:58.37 | oej | And now the question: What did he do and get an award for? |
| 19:59.13 | file | Norway. |
| 19:59.19 | oej | file: Muffin to you! |
| 19:59.23 | oej | Norwegian fjords |
| 19:59.34 | blitzrage | the fiddly bits! |
| 19:59.36 | putnopvut | HH2g!! |
| 19:59.43 | file | :D |
| 19:59.52 | blitzrage | never read it :) |
| 19:59.54 | blitzrage | saw the movie though |
| 20:00.00 | putnopvut | blitzrage: !!! |
| 20:00.05 | blitzrage | putnopvut: !!! |
| 20:00.08 | putnopvut | bad boy! |
| 20:00.16 | blitzrage | I avoid nerd books :) |
| 20:00.36 | oej | blitzrage: guess you never read the asterisk book, then... Just wrote it. |
| 20:00.47 | blitzrage | exactly |
| 20:01.13 | jsmith | oej: Why would he read it? It's full of lies! |
| 20:01.36 | oej | jsmith: Right, it's useless. Doesn't cover the next version of Asterisk at all. |
| 20:01.51 | blitzrage | it's true |
| 20:02.01 | jsmith | oej: And even the new edition won't cover 1.6 |
| 20:02.05 | Corydon76-work | Speaking of, any news? |
| 20:02.31 | jsmith | Corydon76-work: I've passed all my stuff off to them, so I'm assuming it's going through final copyediting/preparing-to-print stuff |
| 20:02.33 | blitzrage | I'm working on some of it right now actually |
| 20:02.37 | Corydon76-work | like when the print run commences? |
| 20:02.48 | jsmith | Corydon76-work: Our editor had surgery yesterday, so we're not getting a lot of feedback |
| 20:02.48 | blitzrage | Laurel sent me some code that I'm going to break probably by trying to make it fit in a box |
| 20:02.56 | jsmith | blitzrage: Oh, I thought you were done. |
| 20:02.59 | Corydon76-work | Ah... |
| 20:03.16 | Corydon76-work | So it's going to miss the August 1st print date? |
| 20:03.16 | blitzrage | jsmith: I was.... but this is in an appendix, and no one replied first time, so I'm going to take a shot at it |
| 20:03.29 | jsmith | Oh? |
| 20:03.34 | jsmith | I must have missed that. |
| 20:03.44 | jsmith | Corydon76-work: Not sure... maybe by a few days |
| 20:10.22 | oej | putnopvut should not be surprised... |
| 20:13.24 | *** join/#asterisk-dev jefforulez (n=jefforul@38.96.187.252) |
| 20:14.37 | tzafrir_laptop | BTW: anybody building h323 on 1.2? |
| 20:15.47 | tzafrir_laptop | It seems to be broken for quite some time. Unless I really fail to understand things there |
| 20:47.35 | *** join/#asterisk-dev bkuhn (i=bkuhn@sflc/staff/conservancy.president.bkuhn) |
| 20:49.47 | russellb | putnopvut: if you do it on viewvc it is likely to be a little faster |
| 20:50.06 | russellb | i think chan_sip has gone through more changes than any other file ... |
| 20:50.35 | Qwell[] | svn annotate is faster than svn blame |
| 20:51.55 | Qwell[] | M10304 |
| 20:51.56 | MuffinMan | [new] [Asterisk] Channels/chan_sip/Subscriptions 0010304: Hint subscriptions do not update without peer call-limit entry reported by Brett Matthews http://bugs.digium.com/view.php?id=10304 |
| 20:51.58 | Qwell[] | That's by design, isn't it? |
| 20:52.20 | Qwell[] | closing.. |
| 20:52.31 | mvanbaak | gheh, in 3 days they dropped this thinkpad T61 price with 500 euro |
| 20:53.29 | russellb | Qwell[]: though I suppose we could improve it ... i'm not sure if it goes to IN_USE after 1 call regardless of the call limit |
| 20:53.32 | russellb | it probably should |
| 20:53.39 | russellb | and then only to BUSY at the limit |
| 20:55.18 | oej | We have two call limits if you check. One hard limit (call-limit) and a new one for busy |
| 20:55.33 | russellb | oej: right ... that's just trunk though, right? |
| 20:55.39 | oej | Yes |
| 20:55.42 | russellb | the busy limit, that is |
| 20:55.43 | russellb | cool |
| 20:56.07 | oej | It does improve things. You can still do call transfers even if you hit the limit. |
| 20:56.14 | russellb | yeah, that's awesome |
| 20:56.21 | russellb | i remember that now |
| 20:56.39 | russellb | i have another set of changes that improves performance |
| 20:56.45 | russellb | well, *can* improve it |
| 20:56.57 | russellb | i'll have to document it at some point :) |
| 20:58.00 | oej | I'm working on subscription redesign. How we handle subscriptions and such. |
| 21:08.55 | russellb | oej: neat |
| 21:14.41 | *** join/#asterisk-dev Jabroni (n=Jabroni@red-corp-200.76.249.142.telnor.net) |
| 21:15.37 | *** part/#asterisk-dev Jabroni (n=Jabroni@red-corp-200.76.249.142.telnor.net) |
| 21:30.28 | *** join/#asterisk-dev MrTelephone (n=MrTeleph@h697179-171.picriverisp.net) |
| 21:34.05 | *** join/#asterisk-dev Simon-- (n=sim@staff-nat.netnation.com) |
| 21:36.43 | Qwell[] | haha, chan_oss + aalib |
| 21:40.59 | russellb | :) |
| 21:41.14 | Qwell[] | aalib is awesome. I once watched a movie with aalib. |
| 21:41.29 | putnopvut | Is aa ascii art? |
| 21:41.32 | Qwell[] | yeah |
| 21:41.40 | Qwell[] | it converts the video into ascii art...it's really slick |
| 21:41.41 | putnopvut | You watched an entire movie with aalib? |
| 21:41.43 | Qwell[] | yes |
| 21:41.49 | MrTelephone | thats bs |
| 21:41.52 | MrTelephone | haha |
| 21:42.06 | Qwell[] | mplayer -vo aa I think it is |
| 21:42.12 | Qwell[] | make it really easy |
| 21:42.17 | MrTelephone | MGCP -> SIP, thousand per copy, I think im going to have a porsche by next year |
| 21:42.45 | MrTelephone | if it doesn't fly then I'll buy qwells minivan |
| 21:43.25 | putnopvut | M10302 |
| 21:43.26 | MuffinMan | [new] [Asterisk] Channels/chan_sip/General 0010302: Wrong duration value in SIP INFO messages when sending DTMF reported by litnialex http://bugs.digium.com/view.php?id=10302 |
| 21:43.50 | MrTelephone | what is the SIP RTO like? |
| 21:43.52 | putnopvut | I decided it was time to grow up and work on a sip issue. Do any of you think the claims about the bogus duration value are the cause of the problem? |
| 21:44.20 | *** join/#asterisk-dev lrizzo (n=luigi@host40-238.pool80117.interbusiness.it) |
| 21:44.38 | MrTelephone | i think its bogus |
| 21:44.48 | MrTelephone | does your dtmf work at all? |
| 21:45.19 | MrTelephone | if you send asterisk too many dtmf frames at once will it screw it up? |
| 21:45.31 | MrTelephone | I wanted to make a loop from an array[]="03204023404324 |
| 21:45.43 | MrTelephone | and have dtmf tones generated from the string |
| 21:46.28 | file | putnopvut: they didn't give all the needed information... we have to see the box that is sending the INFO dtmf, not the one that is receiving it |
| 21:47.11 | bbryant | MrTelephone, there's an ast_senddigit that takes a channel and a single char |
| 21:47.53 | putnopvut | When you say "i think its bogus" what is "its" in that sentence? |
| 21:49.19 | putnopvut | file: thanks. I'll make a note of it on the bug. |
| 21:50.12 | MrTelephone | bbryant, that may be friggin awesome |
| 21:50.48 | MrTelephone | whats this all about then? |
| 21:50.49 | MrTelephone | else if ((strlen(ev) == 1) && |
| 21:50.49 | MrTelephone | <PROTECTED> |
| 21:50.49 | MrTelephone | <PROTECTED> |
| 21:50.49 | MrTelephone | <PROTECTED> |
| 21:50.49 | MrTelephone | <PROTECTED> |
| 21:50.51 | MrTelephone | <PROTECTED> |
| 21:50.53 | MrTelephone | <PROTECTED> |
| 21:50.58 | MrTelephone | ev = a single digit pressed on the phone |
| 21:52.10 | file | that is different |
| 22:00.14 | MrTelephone | adding to ast_channel_tech tells asterisk what subroutines to call when a certain event occurs such as answer |
| 22:04.32 | file | putnopvut: :D |
| 22:05.25 | putnopvut | Yeah. |
| 22:08.22 | MrTelephone | im a foo |
| 22:14.34 | *** join/#asterisk-dev janinge (n=janinge@211.80-202-239.nextgentel.com) |
| 22:15.49 | MrTelephone | is iabuf an asterisk function? |
| 22:16.04 | MrTelephone | and how do u determine the max size |
| 22:20.10 | bbryant | MrTelephone, all asterisk functions that aren't static start with ast_ |
| 22:23.54 | MrTelephone | one more silly question before i head home |
| 22:24.32 | MrTelephone | in the channel code it tells asterisk to open a socket on IP:PORT.. where there is traffic on that port where does it go? |
| 22:25.29 | bbryant | usually there's a thread that's started to read from the socket |
| 22:27.56 | MrTelephone | in the channel code? |
| 22:29.29 | MrTelephone | thanks for the info |
| 22:29.31 | MrTelephone | gotta go |
| 22:33.27 | *** join/#asterisk-dev smurf (n=smurf@debian/developer/smurf) [NETSPLIT VICTIM] |
| 22:33.27 | *** join/#asterisk-dev festr__ (n=festr@ns.regnet.cz) [NETSPLIT VICTIM] |
| 22:33.27 | *** join/#asterisk-dev corramor (i=corramor@nat/digium/x-c727d05dcfa1952a) [NETSPLIT VICTIM] |
| 22:33.27 | *** join/#asterisk-dev Juggie (n=Juggie@CPE00c049d9f271-CM000a73a18a20.cpe.net.cable.rogers.com) [NETSPLIT VICTIM] |
| 22:33.27 | *** join/#asterisk-dev j4m3s (i=james@nat/digium/x-5e93b19f992e63ec) [NETSPLIT VICTIM] |
| 22:39.32 | Qwell[] | /path/to/somefile.so: /usr/lib/libstdc++.so.6: version `GLIBCXX_3.4.6' not found |
| 22:39.54 | Qwell[] | That means that they built it with a version of something (gcc?) that other stuff wasn't built with, right? |
| 22:40.07 | Qwell[] | s/with/against/g |
| 22:43.48 | *** join/#asterisk-dev kpfleming (i=kpflemin@nat/digium/x-64b068bb5d6d80b2) |
| 22:43.48 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 22:45.16 | blitzrage | anyone here from Minnesota? |
| 22:45.28 | Qwell[] | blitzrage: I've been to Minnesota. Does that count? |
| 22:45.49 | blitzrage | not really... but the Toronto Blue Jays spanked the Twins today :) |
| 22:48.25 | *** join/#asterisk-dev jefforulez (n=jeffo@pool-71-187-66-159.nwrknj.fios.verizon.net) |
| 23:01.32 | *** join/#asterisk-dev kpfleming (i=kpflemin@nat/digium/x-f6201a7f0dd63aa7) |
| 23:01.32 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 23:06.23 | *** join/#asterisk-dev ManxPower (n=manxpowe@209.16.72.142) |
| 23:14.42 | Simon-- | yay, reproducible vnak storm |
| 23:14.53 | snuff-work | :) |
| 23:15.57 | jefforulez | hey simon. :) |
| 23:16.20 | Simon-- | hey! |
| 23:16.26 | Simon-- | I got a VNAK/HANGUP fight just then |
| 23:16.52 | jefforulez | nice. can you reproduce it on demand? |
| 23:17.30 | Simon-- | yeah, trying to narrow it down |
| 23:17.50 | jgoddess | heh |
| 23:17.58 | jgoddess | maybe I should call youa gain |
| 23:18.00 | jgoddess | =P |
| 23:18.42 | jgoddess | I can do it from home and have it dial through the extension ;) |
| 23:19.35 | *** mode/#asterisk-dev [+o codefreeze] by ChanServ |
| 23:19.36 | Simon-- | jgoddess: found my bug, it will null pointer in a bit..will build you fixed packages |
| 23:28.34 | Qwell[] | codefreeze: does "make: expand.c:489: allocated_variable_append: Assertion `current_variable_set_list->next != 0' failed." during a build (on aelparse, it seems) mean anything to you? |
| 23:33.16 | Simon-- | jefforulez: calling an Echo() extension from an asterisk with a broken console /dev/dsp is nice.. it sends no voice frames at all :) |
| 23:37.56 | codefreeze | Qwell: huh? let me look at that. brb |
| 23:40.12 | codefreeze | Qwell: can't find current_variable_set_list in trunk....? |
| 23:41.13 | codefreeze | nor expand.c |
| 23:48.05 | file | Qwell[]: what version of make? |
| 23:49.59 | codefreeze | Ah, yes, file: it's not AEL code, it's internal to make! |