| 00:04.50 | *** join/#asterisk-dev moy (n=moyhu@189.169.83.74) |
| 00:09.53 | gitguy | i'm using it already |
| 00:58.49 | *** join/#asterisk-dev mogorman (n=mog@c-68-62-219-86.hsd1.al.comcast.net) |
| 00:58.49 | *** mode/#asterisk-dev [+o mogorman] by ChanServ |
| 01:30.34 | *** join/#asterisk-dev IgorG (n=IgorG@195.162.49.23) |
| 02:15.36 | gitguy | it is true that asterisk suffers of "design mistakes" and things like that, like some peoples like to say? (i don't want to mention names) |
| 02:15.37 | *** join/#asterisk-dev DarkRift (i=dark@bas10-montreal02-1177582051.dsl.bell.ca) |
| 02:16.11 | gitguy | i ask this because i like asterisk and i care of it |
| 02:17.51 | Qwell | every software suffers from "design mistakes" |
| 02:19.33 | gitguy | yeah |
| 02:20.15 | gitguy | "evolution is better than intelligent design" <-- i like that more |
| 02:20.54 | gitguy | http://www.kroah.com/log/images/ols_2006_keynote_08.jpg |
| 02:30.46 | jeffspeff | hey devs, got a question for you... I'm wanting to stress test my system. It's my first * box ever, using VOIP and SIP. The machine has pretty decent specs and I've had 2 concurrent calls without problem; but I want to find out how many calls it can handle in this environment (environment being internet speed, current LAN topology, system specs, etc.) before you start getting echo and the system becomes unusable... |
| 02:31.31 | jeffspeff | i asked this earlier in the main * channel somebody recommended sipp... i installed it and tried it out, but i'm getting errors; and that person has since left the channel and nobody else is of any help... |
| 02:31.38 | jeffspeff | got any suggestions? thanks. :) |
| 02:35.58 | *** join/#asterisk-dev denon (n=denon@tooth.decay.org) |
| 02:42.31 | *** join/#asterisk-dev gmaruz1 (n=gmaruzz@213-140-6-122.ip.fastwebnet.it) |
| 02:46.59 | jeffspeff | Qwell, any suggestions or ideas? |
| 03:11.51 | Corydon76-dig | Qwell is probably spending time with his family |
| 03:30.34 | *** join/#asterisk-dev hd2 (n=blah22@adsl-76-195-128-35.dsl.crchtx.sbcglobal.net) |
| 03:31.19 | hd2 | I am using switchvox and the IVR menu is not accepting DTMF for some reason from external lines as well, any clue on what the issue might be? |
| 03:32.33 | *** part/#asterisk-dev hd2 (n=blah22@adsl-76-195-128-35.dsl.crchtx.sbcglobal.net) |
| 06:01.12 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 06:01.12 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 06:07.28 | *** join/#asterisk-dev mvanbaak_ (n=michiel@vanbaak.xs4all.nl) |
| 06:59.36 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.131) |
| 07:27.34 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:35.30 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:43.46 | oej | Morning |
| 07:58.31 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:59.21 | *** join/#asterisk-dev qdk_ (n=qdk@85.235.253.139) |
| 08:15.07 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 08:17.06 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 09:08.15 | *** join/#asterisk-dev RoyK (n=roy@fw.fortel.no) |
| 09:45.14 | *** join/#asterisk-dev lvl- (n=lvl@101.248.dhcp.lht.hhs.nl) |
| 09:55.46 | *** join/#asterisk-dev lvl- (n=lvl@101.248.dhcp.lht.hhs.nl) |
| 10:05.18 | *** join/#asterisk-dev friendly12345 (n=friendly@ppp59-167-137-15.lns3.mel6.internode.on.net) |
| 11:08.47 | *** join/#asterisk-dev DagMoller (n=aguirre@unaffiliated/dagmoller) |
| 11:30.45 | jmls_sleeps | anyone know when creslin may be around ? |
| 11:31.20 | jmls_sleeps | this libpri thing is *really* causing me some grief |
| 12:08.09 | snuff-home | mm wtf.. |
| 12:08.11 | snuff-home | M12634 |
| 12:08.12 | MuffinMan | [new] [Asterisk] Applications/General 0012634: CRBT using Asterisk PBX reported by Sukhbir (Karma: neutral) http://bugs.digium.com/view.php?id=12634 |
| 12:08.47 | jmls_sleeps | sure, mantis is used for user questions. what's wrong with that |
| 12:08.52 | jmls_sleeps | sniggers |
| 12:09.18 | snuff-home | ;) |
| 12:10.24 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:10.24 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 12:12.57 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 12:18.02 | snuff-home | jmls_home, u can close 12439 |
| 12:18.06 | snuff-home | user says :) |
| 12:18.51 | jmls_home | done ;) |
| 12:19.52 | *** join/#asterisk-dev Shazaum (n=shazaum@200.175.61.250.static.gvt.net.br) |
| 12:20.15 | snuff-home | jmls u use 1.4 svn or trunk? |
| 12:21.43 | jmls_home | currently 1.4 svn. Testing 1.6.0 .. |
| 12:23.19 | snuff-home | ah |
| 12:24.23 | jmls_home | ah ? |
| 12:25.39 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 12:25.39 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 12:26.28 | snuff-home | just wondered.. since 1.6.0 doesn't have the file i/o buffer for recordings |
| 12:26.47 | snuff-home | since i know u do chan recording a bit |
| 12:32.09 | jmls_home | I did do a large test recently with 20000 calls under 1.6, both legs MixMonitored and didn't notice anything. Had 29 channels inbound and 29 outbound |
| 12:33.24 | snuff-home | mm.. well if u ever feel like running that test again.. aka with/without i/o buffer.. |
| 12:33.44 | snuff-home | since 1.6.0 doesnt have the patch.. it can be found in #11962 |
| 12:34.32 | jmls_home | why the change ? Is it meant to work better ? |
| 12:35.33 | snuff-home | yes.. reduce writing to disk until have something little more substantial than 1kb ;) |
| 12:36.26 | jmls_home | cool. I'll try and crank up 1.6 tonight and let you know |
| 12:37.33 | snuff-home | np.. |
| 12:45.49 | atis_work | jmls - i remember you had queue member selection strategy based on curl, right? |
| 12:47.12 | atis_work | jmls_home: i just wonder how well does that works. don't you have queue reloads all the time? |
| 12:47.45 | atis_work | jmls_home: i have chosen to put small curl usage within app_queue - pass existing memberlist and read one/many from there |
| 12:52.29 | jmls_home | atis_work: yes, I do |
| 12:52.59 | jmls_home | I use curl for realtime queues and queuemembers, but I have a single queuemember (a local channel) |
| 12:53.12 | atis_work | oh |
| 12:53.14 | jmls_home | this local channel then uses the curl function to query the application for the member |
| 12:53.42 | jmls_home | works _great_ |
| 12:53.45 | atis_work | i just wonder - what's the selection criteria you use |
| 12:54.16 | jmls_home | whatever I want to code in my app. that's the beauty, you are not restricted to the app_queue criteria |
| 12:54.32 | atis_work | yes, i'm just trying to find more general way for this |
| 12:54.46 | jmls_home | an example: we can select a member based on their conversion success rate |
| 12:54.50 | atis_work | yes yes |
| 12:55.01 | jmls_home | or the amount of minutes they have spent on the phone |
| 12:55.19 | jmls_home | or the highest / lowest commission earned at the time |
| 12:55.45 | atis_work | well, i put call to curl within strategy selection, so queue has all members, their device states, etc.. then my script receives list of all queue members, process that and returns one or several |
| 12:56.17 | atis_work | don't you have troubles with device states? |
| 12:56.48 | jmls_home | no, the application keeps track if the member is ready or not |
| 12:58.00 | ctooley | atis_work, for instance if you're doing customer service you might do: select case when hair_color = "blonde" then 1 when hair_color = "brunette" then 2 end as weighting, phone_number from agents order by weighting desc limit 1; |
| 12:58.22 | ctooley | atis_work, on the other hand if it's sales you run: s/desc/asc/g |
| 12:58.38 | atis_work | :) |
| 12:58.47 | ctooley | I'm sure that strategy is not built into app_queue |
| 12:59.03 | atis_work | yes, well that's what also i'm doing now |
| 13:02.41 | ctooley | I prefer to put red heads in tier 2 of customer support. They're usually good at detecting the people that are A) never going to get it, B) don't have an actual problem, and C) are never buying from us again anyway. |
| 13:02.48 | ctooley | It gives them a place to vent. |
| 13:04.37 | ctooley | jmls_home, you've got 2 E1 PRIs into the call center? (one in, one out)? |
| 13:05.26 | jmls_home | ctooley: got 4 actually. 2 out, 1 in+out, 1 test |
| 13:06.06 | ctooley | ah, well, you said 29 channels in, 29 channels out earlier, I made a bad guess. |
| 13:06.21 | ctooley | So, mostly outbound then. |
| 13:06.30 | jmls_home | Ah, that was my test server - I've got a loopback cable on spans 2+3 |
| 13:06.54 | jmls_home | so I can test outbound with the dialler with span 3, coming into queues on span 2 |
| 13:07.01 | ctooley | ah |
| 13:07.36 | jmls_home | it's great ;) |
| 13:08.27 | ctooley | I was mostly wondering because as inbound calls becomes a bigger ration compared to queue agents, you start to run the risk of a race condition that sends two calls to the same agent. |
| 13:09.01 | ctooley | Unless you're making outbound calls that get transferred into the queue, that shouldn't happen as much. |
| 13:09.13 | jmls_home | I was getting that until i started using curl - now my application decides which agent to send the connected call to, not asterisk |
| 13:09.34 | mvanbaak | hhmm, asterisk uses openssl for some crypto stuff right ? |
| 13:10.36 | ctooley | jmls_home, your application still has to depend on some sort of device state (whether it maintains the state itself or not) and that leads to a possibility of conflict. Unless you mark the agent busy as soon as selection is made and assume that the call completes, your agent could get another call. |
| 13:11.10 | ctooley | And if you do that, you have to handle failed attempts to deliver to an agent. |
| 13:12.06 | ctooley | is just discussing because we're supposed to implement something similar and there's not anyone here to discuss it with. :) |
| 13:12.19 | ctooley | here == my job |
| 13:12.50 | jmls_home | my agents are in a table, with the "ready" flag set. I find the first "ready" agent (which locks the record), set ready to no, release the record. Whilst that is ongoing, another process cannot get the record, so it moves onto another read agent |
| 13:12.54 | jmls_home | (ready agent) |
| 13:16.15 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:16.15 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 13:25.25 | *** join/#asterisk-dev moy (n=moyhu@nat/ibm/x-a74478956108437d) |
| 13:33.38 | *** part/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 13:37.13 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 13:38.01 | *** join/#asterisk-dev wwalker (n=wwalker@pdpc/supporter/sustaining/wwalker) |
| 13:53.32 | *** join/#asterisk-dev PTorres (n=PTorres@200.68.87.146) |
| 13:58.03 | *** join/#asterisk-dev sruffell (n=sruffell@216.207.245.1) |
| 13:58.03 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:01.54 | atis_work | jmls_home: but you still have to keep track of agents doing some calls, etc |
| 14:03.49 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.80.140) |
| 14:04.22 | anonymouz666 | file: how hard is to backport func_volume to asterisk 1.4.20-rc2? |
| 14:04.36 | file | rather easy... |
| 14:05.00 | anonymouz666 | I wanna test it here |
| 14:05.08 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.12.183) |
| 14:25.26 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.136) |
| 14:29.27 | snuff-home | mm file.. since u took over the iax2 bridging stuff.. did u merge in the code i put up or is it more part of ur ongoing 'bridge' work? |
| 14:30.12 | file | the bridge enhancements thingy? it's got nothing to do with the bridging branch |
| 14:30.23 | file | I created a branch yesterday, doing some code review and testing |
| 14:30.34 | snuff-home | ah.. |
| 14:34.33 | snuff-home | mm i wonder if the last fix i did on that patch still patches correctly |
| 14:43.41 | *** join/#asterisk-dev Cresl1n (n=matt@216.207.245.1) |
| 14:43.41 | *** mode/#asterisk-dev [+o Cresl1n] by ChanServ |
| 14:44.31 | anonymouz666 | file: working, nice. |
| 14:44.57 | *** join/#asterisk-dev putnopvut (n=putnopvu@216.207.245.1) |
| 14:44.57 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:45.43 | ctooley | I would just like to thank you guys for the hard work on CDRs. After working with the different CDR formats here I have to say the CDR handling in Asterisk is relatively clean. |
| 14:48.29 | *** join/#asterisk-dev jcollie (n=jcollie@fedora/jcollie) |
| 14:49.08 | jsmith | ctooley: Are you serious? |
| 14:50.01 | ctooley | let's see.... 118 columns, 3 INTs, 4 VARCHARSs, 111 CHAR columns |
| 14:50.01 | codefreeze-lap | jsmith: sssshhhhhhh! you spoiled a "special moment" for me! ;) |
| 14:50.55 | ctooley | some CHAR columns left padded with spaces, some right padded with CHAR spaces, ALL indexes are based on CHAR columns. There's not even an id column that's an auto increment INT. |
| 14:51.45 | ctooley | There are 17 boolean flags, all of which are CHAR(4) and whitespace padded, and in an inconsistent manner. |
| 14:52.11 | codefreeze-lap | ctooley: The end user is **always** in charge of the db tables. We just populate the ones we need to. you can handle the rest. |
| 14:52.14 | ctooley | booleans, not even descriptive flags. |
| 14:53.35 | jsmith | ctooley: Ouch.... I see what you mean |
| 14:53.37 | ctooley | codefreeze-lap, I wish I could say that's even legacy CDR format, but it was "designed" in 2006. |
| 14:56.08 | ctooley | In another one there are 143 columns. The primary key is a 3 column index that includes (CUSTOMER_NUMBER CHAR(64), DNIS_SET CHAR(64) and CALL_ID CHAR(128)) |
| 14:58.07 | *** join/#asterisk-dev jpeeler (n=jpeeler@216.207.245.1) |
| 14:59.11 | codefreeze-lap | ctooley: what exactly are you looking at? |
| 15:00.35 | ctooley | codefreeze-lap, I work at a phone company. We have quite a few different "systems" that generate CDRs. That last one is from a legacy Prepaid system. |
| 15:00.48 | codefreeze-lap | wow! |
| 15:00.58 | ctooley | The first one, sadly, is our current universal CDR format. |
| 15:01.07 | codefreeze-lap | Disk space to **burn**, evidently |
| 15:01.25 | ctooley | I'm in the process of working with a team of (sadly) 12 people to rework the CDRs |
| 15:01.49 | codefreeze-lap | generation, or post-processing? |
| 15:01.52 | ctooley | We expect to be able to reduce our UCDR format by more than half, probably down to 1/3 |
| 15:02.00 | ctooley | The UCDR format is post processing. |
| 15:02.06 | codefreeze-lap | wise |
| 15:02.34 | ctooley | Actually of the 6 different formats I've reviewed, all of them are the CDRs we bill off off. |
| 15:02.44 | ctooley | s/off off/off of/g |
| 15:03.09 | ctooley | That's what you get when you have 4 different phone companies merged into one. |
| 15:04.19 | ctooley | actually my potd is getting a calling card platform billable so we can converge from 7 different platforms (4 outsourced and 3 inhouse) to only 2. |
| 15:04.42 | *** join/#asterisk-dev russellb (n=russell@asterisk/developer-and-stable-maintainer/drumkilla) |
| 15:04.42 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 15:05.37 | ctooley | Actually wwalker (you here?) wrote an Audio Conferencing product for us that has to take the Asterisk CDRs and convert them to UCDRs and the column size literally grows more than 10X |
| 15:05.52 | ctooley | with just the data that's in the existing Asterisk CDR. |
| 15:07.41 | wwalker | easily more than 10X, but I'm not using the asterisk CDRs, I'm using records the adhearsion app creates because of the multileg issue and the "who to bill is not related to the ANI or DNIS" issue |
| 15:08.02 | *** join/#asterisk-dev codefreeze-lap (n=murf@216.166.159.235) |
| 15:09.13 | ctooley | wwalker, Ah, my bad |
| 15:09.13 | ctooley | codefreeze-lap was missing... for his benefit... |
| 15:09.25 | ctooley | <wwalker> easily more than 10X, but I'm not using the asterisk CDRs, I'm using records the adhearsion app creates because of the multileg issue and the "who to bill is not related to the ANI or DNIS" issue |
| 15:11.00 | wwalker | I should have pushed a billing code into accountcode or userfield but I didn't |
| 15:12.42 | ctooley | is a big fan of that for troubleshooting purposes. |
| 15:13.32 | ctooley | I usually like having the data in file format when I'm using cdr_odbc as well, "just in case" |
| 15:16.32 | *** join/#asterisk-dev fenlander (n=fenlande@82.152.81.57) |
| 15:17.57 | snuff-home | mm.. all the cdr'ing u could ever want will be availible once the 'CEL' is finished |
| 15:18.18 | snuff-home | call event log = CEL for short |
| 15:18.50 | snuff-home | anyway time for snoozin.. |
| 15:23.36 | *** join/#asterisk-dev jayrod422 (n=jayrod42@node2.164.136.64.1dial.com) |
| 15:36.17 | *** join/#asterisk-dev joetester2 (n=joeteste@216.191.34.13) |
| 15:38.05 | ctooley | Those things require 1.6 :) |
| 15:39.35 | *** join/#asterisk-dev qdk_ (n=qdk@0x573fe57e.bynqu2.broadband.tele.dk) |
| 15:46.25 | *** join/#asterisk-dev mog (n=mog@c-68-62-219-86.hsd1.al.comcast.net) |
| 15:46.25 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 15:48.58 | *** join/#asterisk-dev gitguy (n=diego@adsl-134-171.click.com.py) |
| 15:51.42 | *** join/#asterisk-dev shido6 (n=shido6@74-130-51-159.dhcp.insightbb.com) |
| 15:52.43 | shido6 | hwo would i go about adding multiple recipients to a forwarded message in "vm_forwardoptions" |
| 15:52.56 | shido6 | and forward_message |
| 15:55.51 | *** join/#asterisk-dev jcollie (n=jcollie@fedora/jcollie) |
| 15:57.16 | putnopvut | shido6: you can forward to multiple recipients already, but it's not exactly obvious how it's done. |
| 15:57.35 | shido6 | i push "8" |
| 15:57.36 | putnopvut | You can enter multiple extensions and separate them with the '*' key |
| 15:57.48 | shido6 | WOW |
| 15:57.50 | shido6 | really?!?! |
| 15:57.53 | putnopvut | yes. |
| 15:58.47 | putnopvut | I seem to recall something on the bugtracker which was going to make the process more obvious by prompting for more extensions to dial, but I'm not sure of the status. |
| 16:00.48 | shido6 | um |
| 16:00.54 | shido6 | * takes u back to the main menu |
| 16:01.15 | Qwell | inside of the extension |
| 16:01.19 | Qwell | 5555*5556*5557 |
| 16:01.37 | shido6 | i am sorry thats not a valid extension |
| 16:01.45 | shido6 | i tried valid extensions seperated by a * |
| 16:03.04 | shido6 | s = strsep(&stringp, "*"); |
| 16:03.05 | shido6 | ? |
| 16:03.53 | putnopvut | shido6: it worked fine for me just now. |
| 16:04.10 | shido6 | what version |
| 16:04.16 | shido6 | and can i have a copy of your app_voicemail.c |
| 16:04.17 | shido6 | ? |
| 16:04.35 | shido6 | you listened to an old vmail and press "8" to forward , right? |
| 16:04.41 | shido6 | then pressed 1 to enter an extension |
| 16:04.41 | putnopvut | yes. |
| 16:04.50 | shido6 | then u entered 123*1234*12345 |
| 16:04.58 | shido6 | and it sent the message to all 3 ppl |
| 16:04.59 | shido6 | ? |
| 16:05.03 | putnopvut | No, it actually immediately asked for an extension, I didn't have to press 1 to enter an extension. |
| 16:05.08 | shido6 | ok |
| 16:05.12 | putnopvut | I'm using the 1.4 branch, latest svn checkout. |
| 16:05.21 | shido6 | latest svn.... |
| 16:05.22 | shido6 | hrmmmm |
| 16:14.13 | shido6 | hrmmm |
| 16:18.04 | *** join/#asterisk-dev codefreeze-lap (n=murf@71-36-6-234.chyn.qwest.net) |
| 16:19.30 | tzanger | $ cvs tag -b POINTLESS_TDM_TEST_BULLSHIT_BUSYWORK |
| 16:19.35 | tzanger | just so y'all know. :-( |
| 16:19.38 | shido6 | ... |
| 16:19.58 | russellb | o.O |
| 16:20.07 | shido6 | where in app_voicemail.c does it allow you to do that, putnopvut? |
| 16:20.45 | putnopvut | shido6: it's what you specified before, the strsep(&stringp, "*"); in forward_message |
| 16:20.46 | *** join/#asterisk-dev mort___ (n=mort@user-3e8886cc.tcl115.dsl.pol.co.uk) |
| 16:21.02 | shido6 | hrmm |
| 16:21.09 | shido6 | hrmmm...... |
| 16:22.29 | *** join/#asterisk-dev DarkRift (i=dark@bas10-montreal02-1177582051.dsl.bell.ca) |
| 16:37.40 | shido6 | ok |
| 16:37.55 | shido6 | i need user@context to forward i think |
| 16:38.17 | shido6 | n/m |
| 16:38.25 | shido6 | i just dont think its working for me |
| 16:38.35 | shido6 | using 1.4.18 |
| 16:42.51 | *** join/#asterisk-dev mcargile (n=mcargile@gateway.in-florida.net) |
| 16:43.26 | *** join/#asterisk-dev jets (n=brian@pdpc/supporter/active/jets) |
| 16:43.28 | mcargile | Did the changing of ztdummy to use the high performance timers in the kernel help with XEN virtualization? |
| 16:56.18 | *** join/#asterisk-dev ZaVoid (n=zavoid@c-67-165-25-195.hsd1.pa.comcast.net) |
| 16:56.40 | ZaVoid | morning all |
| 16:57.13 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 17:07.18 | shido6 | OK |
| 17:07.19 | shido6 | ITS THERE |
| 17:07.21 | Corydon76-dig | Lovely. So apparently Debian did something bad to key generation |
| 17:07.22 | shido6 | IT WORKS |
| 17:07.40 | shido6 | thanks a lot putnopvut |
| 17:08.03 | russellb | Corydon76-dig: yeah ... saw that |
| 17:08.38 | Juggie | hah |
| 17:08.43 | Juggie | well, when you Remove the seed |
| 17:08.53 | Juggie | thats a good way to create a key generation problem :) |
| 17:09.43 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 17:11.44 | russellb | blames file |
| 17:13.57 | *** join/#asterisk-dev JunK-Y (n=junky@modemcable153.55-201-24.mc.videotron.ca) |
| 17:14.54 | file | falls over |
| 17:15.50 | ZaVoid | it seems as if its always files fault |
| 17:15.58 | russellb | yes. |
| 17:20.41 | Juggie | its wise to blame the guy in eastern canada |
| 17:20.45 | Juggie | thats who i blame :P |
| 17:22.28 | JunK-Y | russellb: do you think i can get access to branch 1.6.0 so i can merge stuff? |
| 17:23.11 | Juggie | no way |
| 17:23.15 | Juggie | dont trust the montrealer! |
| 17:23.17 | russellb | JunK-Y: have you seen my messages on asterisk-dev? |
| 17:23.46 | russellb | your last commits were not correct |
| 17:28.26 | jmls_home | file, you weren't stuck on that train were you ? |
| 17:28.38 | jmls_home | thinks file was the host ... |
| 17:28.54 | jmls_home | that's why he's staying undercover ... |
| 17:29.12 | file | whistles |
| 17:29.26 | jmls_home | I _knew_ it |
| 17:29.52 | jmls_home | what was tha all about ? Very X-files I thought ... |
| 17:31.53 | ctooley | Not that it's directly related to "asterisk" development, but do any of you guys know how to lower the amount of time a TCP socket stays in TIME_WAIT? Our EIVR connections aren't going away fast enough and that seems to be a roadblock in our scaling tests. |
| 17:32.27 | Corydon76-dig | ctooley: SO_LINGER, but I don't think you can reduce the default |
| 17:32.42 | Juggie | .. /proc/sys/net/ipv4/tcp_fin_timeout |
| 17:32.43 | jmls_home | hey, c'mon lets stay on topic. I really want to know what happened on that train in Canada, and what files' involvement was .. |
| 17:32.46 | jmls_home | :) |
| 17:32.52 | Juggie | its an os setting i dont think you can change it in code |
| 17:33.00 | Juggie | (not with a function anyway) |
| 17:34.06 | JunK-Y | russellb: just saw it. but i dont understand why we need to unregister if we failed to properly acf register. |
| 17:34.29 | Corydon76-dig | russellb: did you see my note on that? |
| 17:34.36 | Corydon76-dig | err, JunK-Y ? |
| 17:34.37 | russellb | Corydon76-dig: yep |
| 17:34.39 | russellb | and replied |
| 17:34.44 | Corydon76-dig | Sorry |
| 17:35.25 | Corydon76-dig | JunK-Y: If you look at the full code, you'll see that the INTENT was to unregister the function ABOVE that was registered, even though that's not what the code did |
| 17:36.34 | russellb | copy/paste error. |
| 17:37.34 | russellb | code cleaning up after itself is a good thing, though |
| 17:38.22 | bkruse | M12635 |
| 17:38.24 | MuffinMan | [feedback] [Asterisk-GUI] Card Detection 0012635: Card detection not working when analog and digital both present reported by msales (Karma: neutral) http://bugs.digium.com/view.php?id=12635 |
| 17:38.25 | bkruse | ztscan bug? |
| 17:41.58 | russellb | you tell me boss |
| 17:45.14 | *** join/#asterisk-dev codefreeze-lap (n=murf@216.166.159.235) |
| 17:45.46 | *** join/#asterisk-dev PTorres (n=PTorres@200.68.87.146) |
| 17:47.31 | JunK-Y | Corydon76-dig: gotcha, cause what was in the trunk before my commit was not correct, i will modify it to fix that with what russell said in the -commits |
| 17:47.49 | russellb | already fixed .. |
| 17:48.10 | JunK-Y | ur fast, was not fixed 10 minutes ago :) |
| 17:50.51 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.137) |
| 17:53.51 | *** join/#asterisk-dev jmls_work (n=jmls_wor@mail.tessera.co.uk) |
| 17:56.33 | *** join/#asterisk-dev jmls_home (n=jmls_hom@mail.tessera.co.uk) |
| 18:02.51 | jmls_home | russellb: what are the blocking issues for 1.4.20 ? |
| 18:03.18 | russellb | putnopvut was looking at a weird SIP thingy |
| 18:03.31 | russellb | and there is a bug open talking about deadlocks in 1.4.20 .. |
| 18:03.40 | russellb | M12603 |
| 18:03.43 | MuffinMan | [feedback] [Asterisk] Channels/chan_sip/General 0012603: Asterisk locks up and cannot process incoming SIP messages on 1.4.20-rc1 reported by jyap (Karma: neutral) http://bugs.digium.com/view.php?id=12603 |
| 18:07.30 | jmls_home | mmm, ok. I'll be running a 10000+ call test tonight, sip in, zap out back into zap, so I'll let you know if I find anything. This is 1.4 svn |
| 18:08.17 | russellb | thanks! |
| 18:08.38 | bkruse | russellb: I do not know :X |
| 18:10.21 | ctooley | russellb, does putnopvut think it's fixed now? |
| 18:10.28 | ctooley | We're running into it regularly in our testing. |
| 18:10.39 | russellb | no |
| 18:10.42 | putnopvut | ctooley: nope |
| 18:10.43 | russellb | putnopvut: still looking at it? |
| 18:10.47 | putnopvut | 12603? |
| 18:10.51 | ctooley | And we're running million call jobs every couple of hours |
| 18:11.29 | putnopvut | I haven't looked at it today, but I can certainly shift my focus to it. The problem is that I'd like to get a more accurate core show locks output from the reporter (or anyone). |
| 18:11.31 | ctooley | I can update to SVN and have info for you within a couple of hours. Just let me know what the settings you want for the compile. |
| 18:11.50 | ctooley | probably less time than that. |
| 18:11.50 | putnopvut | ctooley: Thanks. DONT_OPTIMIZE and DEBUG_THREADS should be sufficient. |
| 18:12.07 | ctooley | ok, log data will be huge, what specifically do you want me to parse out? |
| 18:12.18 | ctooley | or do you want the entire huge log files? |
| 18:12.24 | russellb | putnopvut: oh, thought you were looking at the issue where you get errors about failing to grab the channel lock |
| 18:12.25 | ctooley | Or do you just want "core show locks" output |
| 18:12.26 | russellb | even on one call |
| 18:12.28 | russellb | or so i've heard |
| 18:12.35 | putnopvut | russellb: oh, oops. |
| 18:12.43 | putnopvut | I can look into that more too. |
| 18:13.09 | putnopvut | The reporter on 10342 now is having that error on his calls. |
| 18:13.23 | putnopvut | So that may be a place to ask for more trace information, too. |
| 18:13.49 | ctooley | BTW some of the "error" messages are invalid, if no one has noticed. |
| 18:14.05 | putnopvut | ctooley: which ones in particular? |
| 18:14.07 | ctooley | (without recompiling...) |
| 18:14.09 | ctooley | [root@plsgriffin01 ~]# /opt/asterisk-1.4.19-routing/sbin/asterisk -rx "core show locks" |
| 18:14.09 | ctooley | No such command 'core show' (type 'help' for help) |
| 18:14.17 | ctooley | "core show channels" works. |
| 18:14.31 | putnopvut | Yeah, that output sucks and is not helpful, I agree. |
| 18:14.38 | ctooley | so obviously "core show" is valid. |
| 18:14.55 | ctooley | though technically not complete. |
| 18:14.57 | putnopvut | Right, it should say "No such command 'core show locks' ... |
| 18:15.04 | putnopvut | if that's what you typed. |
| 18:15.12 | ctooley | putnopvut, do these calls need RTP? |
| 18:15.24 | putnopvut | I actually don't know, but I would assume not. |
| 18:15.33 | ctooley | putnopvut, our calls aren't even answered. |
| 18:15.41 | ctooley | so this might be good information for you. :) |
| 18:16.01 | putnopvut | ctooley: thanks for the help. |
| 18:16.05 | ctooley | np |
| 18:16.13 | ctooley | brb, gotta go pick up my lunch |
| 18:22.38 | *** join/#asterisk-dev shido6 (n=shido6@74-130-51-159.dhcp.insightbb.com) |
| 18:26.43 | ctooley | putnopvut, what specifically am I looking for so I know that the data I'm sending you is useful? |
| 18:28.13 | putnopvut | ctooley: I guess what you're looking for is for SIP calls to start failing. I'm assuming you'll also start seeing the message "We could NOT get the channel lock for ..." all over the place, too. |
| 18:28.13 | putnopvut | When that happens, a core show locks will likely help. |
| 18:28.17 | ctooley | hmmm I'm not seeing that behavior |
| 18:28.41 | putnopvut | I'm starting to get a bit confused. What were you saying was happening in your setup? |
| 18:28.42 | ctooley | I haven't seen those log messages at all. |
| 18:28.54 | putnopvut | So you may be experiencing some other lock, then. |
| 18:29.10 | ctooley | I'm having periods where asterisk will run 5-7 million calls and then it will just stop accepting new calls. |
| 18:29.47 | ctooley | sometimes it's after a few hundred calls, usually it's in the millions |
| 18:29.50 | Corydon76-dig | ctooley: what's the last uniqueid? |
| 18:30.01 | ctooley | uhhh, don't know. |
| 18:30.15 | ctooley | we're not recording cdrs as they're useless to us |
| 18:30.41 | Corydon76-dig | That would say if there's something consistent like a number overflow |
| 18:30.57 | Corydon76-dig | The number of "calls" is not always the same as the number of channels |
| 18:31.04 | jsmith | Yeah... sounds suspicious to me too |
| 18:31.57 | Corydon76-dig | Even a "core show channels" should show the uniqueid |
| 18:32.22 | ctooley | hmm |
| 18:32.51 | ctooley | Corydon76-dig, it does? |
| 18:32.54 | Corydon76-dig | I can't say that that's the problem; I'd like to know that it isn't... |
| 18:33.04 | Corydon76-dig | Or no, sorry |
| 18:33.05 | ctooley | [root@plsgriffin01 ~]# /opt/asterisk-1.4.19-routing/sbin/asterisk -rx "core show channels " |
| 18:33.05 | ctooley | Channel Location State Application(Data) |
| 18:33.05 | ctooley | SIP/54786-ac08f2d0 s@RoutingTest:5 Ring DeadExternalIVR(ivr://127.0.0. |
| 18:33.14 | ctooley | even verbose doesn't show that. |
| 18:33.21 | Corydon76-dig | "core show channel SIP/foo-1234" would show the uniqueid |
| 18:34.03 | ctooley | hmm |
| 18:34.14 | Corydon76-dig | Heh, deadexternalivr |
| 18:34.54 | Juggie | ctooley, if that app exists i'm goign to kill you |
| 18:34.58 | Juggie | *going |
| 18:35.47 | Juggie | why why why would you write DeadExternalIVR |
| 18:35.53 | ctooley | Juggie, not only does it exist, it will continue to exist until such time as I can come up with a good solution to integrate a channel that should never be answered into ExternalIvR() and have time to implement it. |
| 18:36.05 | ctooley | because because because it's been written for a long time. |
| 18:36.27 | russellb | putnopvut: svn diff $RA/tags/1.4.19.2 $RA/branches/1.4 > diff.txt |
| 18:36.28 | Juggie | ctooley, but we removed DeadAGI :( and now you add DeadExternalIVR it makes me sad. |
| 18:36.28 | ctooley | And by good solution I mean one that actually meets the requirements that I have |
| 18:36.34 | russellb | RA=https://origsvn.digium.com/svn/asterisk |
| 18:36.53 | ctooley | oh well, life goes on, and our stuff is continuing to work. |
| 18:37.01 | ctooley | at least that part does. :) |
| 18:37.31 | russellb | by the way, if anyone wants to chat ... join Digium conf 31337 |
| 18:37.34 | Juggie | ctooley, but you can just check the channel status and decide which mode to run i |
| 18:37.37 | russellb | 256-428-6060, 31337 |
| 18:37.47 | russellb | or whatever other number reaches digium, exten 6060 ... |
| 18:37.47 | ctooley | Juggie, no |
| 18:37.56 | ctooley | ExternalIVR expects to be able to answer the channel |
| 18:38.17 | ctooley | that behavior is an expected behavior and has been so since it's inception. |
| 18:38.57 | Juggie | eitherway Dead is not a proper name |
| 18:39.08 | Juggie | because dead implies that at some point the channel did exist |
| 18:39.22 | ctooley | oh the channel exists just fine. |
| 18:39.27 | ctooley | it's just not answered. |
| 18:39.44 | Juggie | right, well i mean a answered channel existed at some point |
| 18:39.59 | Juggie | the whole purpose of deadagi was to run in h() |
| 18:40.13 | Juggie | which then got abused to run agi without it terminating your process |
| 18:40.33 | Juggie | well, first it terminated the socket, then eventually it was fixed to terminate that, and the process |
| 18:40.39 | ctooley | which is a perfectly valid use case for both AGI and EIVR |
| 18:41.16 | Juggie | right, but AGI is slightly different, it doesnt auto answer |
| 18:41.29 | ctooley | And ExternalIVR does, always has. |
| 18:41.47 | ctooley | severely breaks backwards compatibility to change that. |
| 18:41.52 | ctooley | hence the second application. |
| 18:42.49 | ctooley | more importantly... |
| 18:43.14 | Juggie | its unfortunate there is no way to change the application parameters to add something without breaking compatibiltiy |
| 18:43.21 | Corydon76-dig | ctooley: so I'm guessing Excel could provision IAX trunks if you really wanted to? |
| 18:43.37 | Juggie | i just hate adding a whole new application for something which should simply be an option/flag |
| 18:43.50 | ctooley | Tree: DeadExternalIVR() .... Forest: SIP locks and stops answering ... moving on. |
| 18:43.54 | ctooley | Corydon76-dig, well.... |
| 18:44.18 | ctooley | technically yes. But, we have no product nor servers set up to really do that as we don't use Asterisk for any of our IP Trunking solutions. |
| 18:44.25 | ctooley | Asterisk is a feature server for us |
| 18:44.53 | russellb | first call to parse_request in sipsock_read()... doesn't look bad though |
| 18:45.04 | file | russellb: yeah, I agree |
| 18:46.34 | Juggie | ctooley, what happens if you run ExternalIVR on a hungup channel |
| 18:48.20 | ctooley | Juggie, it go boom. first thing ExternalIVR does is check if the channel is hung up and exits if it is. |
| 18:48.42 | ctooley | If it's unanswered it answers it, starts an audio generator on it, and starts looking for stuff. |
| 18:48.54 | Juggie | thats not what i asked |
| 18:49.08 | Juggie | i never asked for an unanswered channel, i asked what it does on a channel thats been HUNUP. |
| 18:49.12 | Juggie | *HUNGUP |
| 18:49.35 | ctooley | Juggie, and I answered that... first. |
| 18:49.43 | mcargile | does anyone know if using ztdummy with hpt timers enabled works as a timer in XEN virtual domains? |
| 18:50.41 | Corydon76-dig | Juggie: you're fighting a losing battle |
| 18:51.07 | Corydon76-dig | It's open source; he's allowed to do whatever he likes with his own system |
| 18:52.46 | Juggie | of course, i just mean from a trunk standpoint, is there any use to having ExternalIVR run on a dead channel |
| 18:53.22 | Juggie | with the exception of setting/getting variables, i'm not sure i see the point |
| 18:54.05 | Corydon76-dig | Clearly there is, or he wouldn't be doing it |
| 18:54.28 | Juggie | Corydon76-dig: you enjoy playing devils advocate? :P |
| 18:54.54 | gitguy | where is a good place to ask for features? |
| 18:55.09 | wwalker | Burger King |
| 18:55.09 | Juggie | no where, dont ask for features, pay or write. |
| 18:55.09 | Corydon76-dig | Feature requests are handled on the wiki |
| 18:55.14 | ctooley | Juggie, Well, there are several perfectly valid use cases for it to occur, even in trunk. |
| 18:55.29 | gitguy | Juggie: sure, i can pay |
| 18:55.32 | gitguy | how much do you want? |
| 18:55.51 | Juggie | ctooley, i could see if we added more comands, perhaps you have added some which are not in doc, but i'm curious what you do on a un-answered channel? |
| 18:55.55 | gitguy | i need 2 features |
| 18:56.00 | Juggie | i guess you could also decide to answer or not |
| 18:56.14 | gitguy | 1- meetme without zaptel. 2- the ability to dial from a conference |
| 18:56.29 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:56.35 | Juggie | meetme without zaptel will exist very soon |
| 18:56.36 | ctooley | Juggie, getting and setting variables is a very important thing to things like routing, authentication based on call parameters, etc |
| 18:56.38 | putnopvut | gitguy: 1 is in a development branch. |
| 18:56.46 | Juggie | prob before end of 2008 in 1.6.x |
| 18:56.55 | gitguy | yay |
| 18:57.33 | Juggie | putnopvut, i guess both kinda are, #2 i am sure could be added to the same branch |
| 18:57.40 | file | easy as pie. |
| 18:58.01 | Juggie | file, how is that branch doing, anywhere close to being file approved? :P |
| 18:58.21 | Corydon76-dig | russell tested, file approved ? |
| 18:58.29 | russellb | O.O |
| 18:58.31 | Juggie | exactally :) |
| 18:58.33 | file | code review brought up some things, mostly done the things brought up |
| 18:58.41 | Juggie | awesome |
| 18:58.47 | Juggie | 1.6.1 1.6.2 timeframe i guess? |
| 18:58.55 | file | sure |
| 18:59.09 | Juggie | well 1.6.0 is frozen right? |
| 18:59.19 | russellb | yes |
| 18:59.23 | file | correct, so you just answered yourself |
| 19:00.00 | gitguy | asterisk owns |
| 19:00.22 | shido6 | can i use ast_set_Flag to set a variable ? |
| 19:00.25 | shido6 | err |
| 19:00.27 | shido6 | ast_set_flag |
| 19:01.03 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 19:01.04 | ctooley | putnopvut, found our "lock" problem, it's not the same thing and it's unrelated. |
| 19:01.33 | jsmith | ctooley: What did it end up being? |
| 19:01.36 | ctooley | goes off fixes a single byte of the multi-thousand line patch-set |
| 19:01.39 | putnopvut | ctooley: I'm intrigued. By "found" do you mean "solved?" |
| 19:02.20 | ctooley | was adding channels to a list and not removing them when they went away. |
| 19:02.29 | ctooley | a VERY long list after several million calls. |
| 19:02.42 | Corydon76-dig | Heh |
| 19:02.48 | Corydon76-dig | ctooley: out of memory? |
| 19:02.58 | ctooley | which explains why it was "erratic", it was only hitting that limit at 500 calls when there had been millions of calls run in jobs before that. |
| 19:03.02 | shido6 | when a user accesses vmail "vmuser" gets populated with ${EXTEN}but when the login fails and asks the user to re-enter their username that vmuser is still set as the first ${EXTEN} i want to clear that upon a failed login |
| 19:03.11 | ctooley | No, I put an artificial limit on the list as an "in case shit" |
| 19:03.41 | ctooley | apparently the logic on the if() was wrong so I wasn't getting the error message either. |
| 19:03.52 | ctooley | At least I know my limit is working. :) |
| 19:03.57 | Corydon76-dig | lol |
| 19:04.21 | ctooley | Asterisk isn't allowed to run away with RAM on this box, Java is the only process allowed to do that. |
| 19:05.32 | ctooley | Apparently when you start with multiple gigs of data, index it 73 ways from Sunday and apply thousands of exception rules to it, the dataset can get somewhat large. |
| 19:06.42 | shido6 | somewhere around logretries++; is where i want to clear vmuser |
| 19:07.23 | ctooley | wishes his C was better, he'd quit making as much work for himself. |
| 19:08.02 | jsmith | ctooley: There's only one way to get better... |
| 19:08.14 | ctooley | jsmith, become someone else? |
| 19:08.42 | ctooley | This is a positive thing to see. :( |
| 19:08.45 | ctooley | <ctooley> [May 13 18:29:53] WARNING[11716] app_deadexternalivr.c: SIP/56574-0ab18620: Option requested but not needed 'TESTATUSX,138' |
| 19:08.45 | jsmith | ctooley: Get rich enough to hire someone else to do it! |
| 19:09.21 | ctooley | will note that the above is a result of one message getting stepped on by another one. |
| 19:09.51 | jsmith | Oh? |
| 19:09.52 | *** join/#asterisk-dev ZaVoid (n=zavoid@c-67-165-25-195.hsd1.pa.comcast.net) |
| 19:09.54 | ctooley | Two messages: V,TIMESTAMP,ROUTESTATUS=300 and X,138RESTOFTIMESTAMP |
| 19:09.56 | ZaVoid | stupid mac |
| 19:10.02 | ctooley | that'sa not good. |
| 19:10.51 | ctooley | apparently it got messed up and it became "O,TESTATUSX,138" and Asterisk interpreted that. |
| 19:11.10 | ctooley | Not only is that an invalid options message, it's a very very bad sign. |
| 19:11.25 | ctooley | wwalker, .... tis gonna be a long day. |
| 19:14.59 | ctooley | putnopvut, ... |
| 19:15.01 | ctooley | [May 13 19:10:46] ERROR[25003] /root/1.4.19.1/include/asterisk/lock.h: channel.c line 936 (ast_queue_frame): mutex '&chan->lock' freed more times than we've locked! |
| 19:15.01 | ctooley | [May 13 19:10:46] ERROR[25003] /root/1.4.19.1/include/asterisk/lock.h: channel.c line 936 (ast_queue_frame): Error releasing mutex: Invalid argument |
| 19:15.05 | ctooley | finally got it. |
| 19:15.10 | putnopvut | ooh |
| 19:15.21 | ctooley | But couldn't get a "core show locks" at the appropriate time. |
| 19:15.47 | putnopvut | Ah, but at least I can see that the error is coming from ast_queue_frame. That's a good start at least. |
| 19:18.23 | Juggie | odd: http://www.usatoday.com/travel/flights/2008-05-12-sue-jet-blue_N.htm |
| 19:19.39 | gitguy | when are you guys releasing 1.6.0? |
| 19:19.46 | gitguy | can i use beta8 already in production? |
| 19:19.51 | gitguy | or you recommend to wait? |
| 19:20.00 | russellb | if you want :) |
| 19:20.02 | russellb | just ... test first |
| 19:20.13 | gitguy | i used yesterday for like 10 minutes |
| 19:20.15 | gitguy | worked fine |
| 19:20.19 | gitguy | i like the new features in the cli |
| 19:20.34 | russellb | which ones |
| 19:20.38 | gitguy | like the number of times you dial when you do core show channels |
| 19:20.38 | gitguy | etc |
| 19:20.41 | gitguy | it rocks |
| 19:20.48 | *** join/#asterisk-dev CrashHD (n=CrashHD@65.74.161.225) |
| 19:24.50 | Juggie | why is it that 1/2 the people who post on dev have more signature in their email then content |
| 19:24.58 | ZaVoid | i must be an idiot.. i switched to comcast home voice 2 weeks ago and twice now i've lost incoming calls and dial tone :) |
| 19:25.05 | Juggie | this last email i just got has like 30 lines of signature, 1 line of content. |
| 19:25.24 | ZaVoid | Juggie: people don't even have fun signatures any more |
| 19:25.34 | ZaVoid | remember the good old days when the sig was 1/2 the bandwifth of the damn email |
| 19:26.09 | ZaVoid | these kinda sigs http://www.trussel.com/f_char.htm |
| 19:33.36 | Corydon76-dig | ZaVoid: I'd go with you're an idiot for thinking of depending upon the cable company. ;-) |
| 19:33.52 | ZaVoid | lol it was the TRIPLE PLAY MAN |
| 19:33.59 | ZaVoid | and the wife wanted to "save" money |
| 19:34.03 | ZaVoid | i only use my cell anyways |
| 19:34.13 | russellb | it's pretty awesome when people put their email address in their email signature |
| 19:35.45 | shido6 | can you change the color of the NoOp text ? ;) |
| 19:36.50 | ctooley | russellb, I do because half the time our Exchange server sends out as Chris.Tooley@Excel.com and the other half time it sends out as ctooley@Excel.com. At least what's in my sig is consistent. (And no, I do not like the fact that we're using Exchange, nor can I impact that decision) |
| 19:38.00 | Corydon76-dig | ctooley: I bet if it was down 90% of the time, they'd switch it |
| 19:38.22 | ctooley | Corydon76-dig, if it was up 10% of the time they'd be amazed. |
| 19:38.44 | Corydon76-dig | ctooley: you can take care of that... |
| 19:38.53 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 19:39.00 | ctooley | actually, I have to admit, this is by far the most stable Exchange environment I've ever seen. |
| 19:39.50 | ctooley | In my time here we've had exactly 3 outages (18 months here) and none of them for more than a couple hours. Which for a 300 person company with no Exchange admins isn't too bad. |
| 19:40.19 | Corydon76-dig | What they're not telling you is that it's a farm of 128 Exchange servers and occasionally, they're able to prevent them all from going down at the same time. ;-) |
| 19:40.51 | tzanger | Corydon76-dig: :-) |
| 19:41.17 | ctooley | Corydon76-dig, actually it's 3 servers. |
| 19:41.34 | ctooley | 1 for actual mail traffic, 2 for mail storage. |
| 19:41.56 | Corydon76-dig | thinks he prefers his P133 running uw-imapd |
| 19:41.56 | ctooley | eats up a lot of server side mail storage. :) |
| 19:42.11 | ctooley | Corydon76-dig, oh, so would I. |
| 19:42.31 | ctooley | didn't say it's the most stable mail environment he'd ever seen, just the most stable Exchange environment. |
| 19:42.41 | Corydon76-dig | Slow as crap while checking folders, but it's completely stable |
| 19:44.34 | Juggie | i've never seen my exchange server down at work here |
| 19:44.39 | Juggie | in 7 years |
| 19:53.59 | ctooley | would like to thank Corydon76-dig for fixing the "asterisk -rx" verbosity problem so that find issues in this routing exercise is easier. |
| 19:54.19 | Corydon76-dig | Heh |
| 19:54.43 | Corydon76-dig | Hey, it was a fun exercise |
| 19:56.09 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:02.19 | putnopvut | So ctooley, when you had that error message print, were there any other errors reported immediately before or after? |
| 20:04.25 | ctooley | oh those messages are long gone |
| 20:04.50 | putnopvut | right, but I just was curious if there were others when those came up or if they were the only ones. |
| 20:05.41 | ctooley | there's 60-70 log messages associated with every call and 110 calls a second. I'm sure there was, but those messages when the way of the "rm -f" |
| 20:05.51 | ctooley | had to, ran out of disk space |
| 20:05.54 | putnopvut | Gotcha. |
| 20:06.52 | ctooley | which is what happens when you have half a meg per call times a 5 million calls * 2-3 runs per day. |
| 20:07.15 | putnopvut | Perfectly understandable to delete that mess. |
| 20:07.19 | ctooley | (sometimes we try just cranking up the volume to see what happens) |
| 20:07.30 | putnopvut | IT GETS LOUDER WHEN YOU DO THAT |
| 20:07.34 | putnopvut | ;) |
| 20:07.38 | jmls_home | whistles at ctooley's setup |
| 20:07.56 | ctooley | putnopvut, not to mention it takes a VERY long time to grep through those files.... gotta get the grep right the first time around. :) |
| 20:08.03 | jmls_home | that *has* to be running on a single server, right ... '< |
| 20:08.22 | ctooley | yes, that is running on a single server |
| 20:08.39 | ctooley | goal is 250 calls/sec/server |
| 20:08.40 | jmls_home | Jesus, I was joking ... |
| 20:09.23 | ctooley | jmls_home, it's a routing server, the "calls" are simply an INVITE, a 100 Trying, a 300 Multiple Choices, and an ACK. roundtrip for a route request is sub 70ms |
| 20:09.46 | jmls_home | oooohhhhh |
| 20:09.57 | ctooley | jmls_home, I knew you were joking, which is why I made sure to respond. :) |
| 20:10.24 | jmls_home | pats his little Dell. It was sweating a little bit hearing those numbers ... |
| 20:11.22 | ctooley | Although... I'd like to hear how anyone that's scaled asterisk to more calls/sec and more concurrent bridged channels than us succeeded in doing it. And that's not bragging, I'd love to know, we're always looking for ways to improve those two things. (All SIP BTW) |
| 20:11.45 | ctooley | s/us succeeded/we succeeded/g |
| 20:17.37 | ZaVoid | i got 250 cooley :) |
| 20:17.44 | ZaVoid | i think the peak was 300.. but that went to crap |
| 20:17.53 | ZaVoid | putting it nicely |
| 20:17.59 | ZaVoid | 5 second sound file was stretched to 15 seconds |
| 20:18.34 | ctooley | 250 calls per second? |
| 20:18.38 | ctooley | or 250 concurrent? |
| 20:18.58 | ZaVoid | concurrent :) |
| 20:19.10 | Juggie | i'm betting disk io would ill that.. might be worth trying your audio files on a ramdisk |
| 20:19.16 | ZaVoid | no way asterisk is doing 250 Callsetups a second |
| 20:19.19 | Juggie | ill=kill |
| 20:19.42 | ZaVoid | yeah thats a thought |
| 20:19.47 | ctooley | ZaVoid, :) That's our goal. It's handling 110/sec in less than 10% of the CPU |
| 20:19.51 | ZaVoid | i just keep throwing more asterisk servers at it |
| 20:19.51 | ctooley | but with no RTP |
| 20:19.57 | ZaVoid | oh no rtp |
| 20:20.06 | ZaVoid | i had 500 calls on when box when i wasn't routing any rtp |
| 20:20.15 | ctooley | Juggie, if there is enough RAM they'll be in disk cache anyway. |
| 20:20.17 | ZaVoid | when = one |
| 20:20.56 | ZaVoid | nothing like seeing that load average hit 80 then 90 though huh ct? |
| 20:21.03 | ctooley | ZaVoid, We can handle upwards of 1000 concurrent per server with RTP, but that's running 4 installations of Asterisk per server. |
| 20:21.09 | ZaVoid | yeah you told me that |
| 20:21.14 | wwalker | ctooley: that doesn't look like messages stepping on one another. That looks like the code throwing away bytes. the message characters are all in order, no interleaving |
| 20:21.16 | ZaVoid | i'm gonna give that a try sometime next week |
| 20:21.35 | ctooley | yup |
| 20:21.44 | ZaVoid | i was thinking maybe those tc400b crapo cards might help with the sound files being played but i like your idea |
| 20:25.16 | Cresl1n | ZaVoid: Have you had issues with the tc400b? |
| 20:25.34 | ZaVoid | tons |
| 20:25.39 | ZaVoid | thats why we sent the two we had back |
| 20:25.42 | ZaVoid | but they claim its fixed now |
| 20:25.47 | ZaVoid | so i might try again to offset some load hopefully |
| 20:25.54 | Cresl1n | yeah, we've got a guy putting some sweet loving into that driver |
| 20:25.55 | ZaVoid | we tried em about a year ago |
| 20:26.08 | ZaVoid | the whole g723 r53 r63 issue bugged(hehe) me out |
| 20:26.09 | Cresl1n | they should be fixed |
| 20:26.19 | ZaVoid | they'd be more usseful to me now if they did liek speex too |
| 20:33.36 | russellb | putnopvut: http://pastebin.ca/1017249 |
| 20:34.07 | mvanbaak | russellb: you pinged me couple of days ago |
| 20:34.21 | *** join/#asterisk-dev Deeewayne (n=Deeewayn@216.207.245.1) |
| 20:34.21 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 20:35.03 | russellb | mvanbaak: yeah, i was just going to remind you about the hangup cause in the hangup control frame changes ... |
| 20:35.11 | russellb | wondering if you were going to make them |
| 20:35.16 | mvanbaak | ah yeah |
| 20:35.19 | mvanbaak | I have them done |
| 20:35.23 | russellb | oh :) |
| 20:35.26 | mvanbaak | but I'm stuck with one issue |
| 20:35.37 | *** join/#asterisk-dev PTorres (n=PTorres@200.68.87.146) |
| 20:35.38 | mvanbaak | and it's too embarrissing to ask ;) |
| 20:36.49 | mvanbaak | I cant get data into frame->data |
| 20:36.53 | mvanbaak | </shame> |
| 20:37.17 | russellb | frame->data = (void *) (unsigned long) hangup_cause; |
| 20:37.31 | russellb | casting magic |
| 20:37.34 | tzafrir | How do I do an svn export with externals? |
| 20:37.56 | mvanbaak | russellb: hhmm, that will do it |
| 20:38.04 | mvanbaak | I tried with (int) |
| 20:38.06 | kpfleming | russellb: well... that will complain going the other way on a 64-bit machine |
| 20:38.11 | mvanbaak | that did give me some nice warnings |
| 20:38.34 | kpfleming | i can propose something... don't know if people will like it |
| 20:38.52 | mvanbaak | do tell |
| 20:39.02 | mvanbaak | this is the one issue why I did not commit it yet |
| 20:39.17 | mvanbaak | I created the new function, changed all occurences in the code |
| 20:39.18 | kpfleming | replace 'void *data'; in ast_frame with 'union { void *data; int data_int;};' or something to that effect |
| 20:39.32 | file | chuckles |
| 20:39.33 | mvanbaak | but I'm stuck with the 4 or 5 places where I have to put data into frame->data |
| 20:39.37 | kpfleming | shouldn't require any changes to existing code, but you can refer to frame->data_int |
| 20:39.39 | PTorres | hey , remember the state change thing ? we believe we have found the problem :D |
| 20:39.49 | kpfleming | wonders why file is chuckling |
| 20:40.27 | mvanbaak | kpfleming: that sounds ok to me |
| 20:41.22 | tzafrir | goes back to rsync -Ca |
| 20:44.15 | kpfleming | mvanbaak: actually, i'd make that 'unsigned int data_int;' just to be safe |
| 20:46.47 | mvanbaak | kpfleming: okies |
| 20:47.11 | mvanbaak | after I'm done uploading my holiday pics (in about 10 minutes) I'll finish the work I've done |
| 20:47.11 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.80.140) |
| 20:47.15 | *** join/#asterisk-dev paravoid (n=paravoid@zebra.cslab.ece.ntua.gr) |
| 20:47.24 | mvanbaak | you want me to put it on mantis or can I commit it to trunk ? |
| 20:47.45 | russellb | kpfleming: yeah, the union is a better idea :) |
| 20:48.03 | russellb | you can commit to trunk, we'll tell you if it's not right ;) |
| 20:48.16 | *** join/#asterisk-dev RoyK (n=roy@72.84-48-10.nextgentel.com) |
| 20:48.47 | anonymouz666 | jpeeler: I just figure out that chan_sip.c includes two times the dnsmgr.h |
| 20:48.50 | anonymouz666 | fun |
| 20:48.51 | anonymouz666 | :D |
| 20:49.16 | jpeeler | that's just the storage for the address |
| 20:49.41 | mvanbaak | lol russellb |
| 20:50.22 | jpeeler | anonymouz666: just look for every instance of "dnsmgr" |
| 20:52.14 | anonymouz666 | that's my idea too |
| 20:53.21 | russellb | putnopvut: http://pastebin.ca/1017271 |
| 20:53.58 | anonymouz666 | jpeeler: but chan_sip is pretty different between the versions |
| 20:54.08 | jpeeler | i don't doubt that |
| 20:54.41 | tzafrir | ok, so what's the deal with those voicebus files keep getting rebuilt in zaptel? |
| 20:55.53 | anonymouz666 | jpeeler: did you change anything on dnsmgr.c code? |
| 20:56.07 | anonymouz666 | or you just used what was there |
| 20:56.30 | jpeeler | there were some small modifications |
| 20:56.44 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:57.21 | jpeeler | mostly done to support SRV records i believe |
| 20:57.23 | *** part/#asterisk-dev Shazaum (n=shazaum@200.175.61.250.static.gvt.net.br) |
| 20:57.43 | russellb | putnopvut: http://pastebin.ca/1017283 |
| 20:59.10 | *** part/#asterisk-dev mcargile (n=mcargile@gateway.in-florida.net) |
| 21:02.03 | anonymouz666 | jpeeler: I just started the modification here, if you don't care to look after I finish.... |
| 21:02.54 | jpeeler | anonymouz666: sorry, but i really don't have any interest at all looking at 1.2 code |
| 21:08.29 | mvanbaak | oh joy, need to regenerate all my ssh keys and stuff :( |
| 21:08.32 | ctooley | ahh how soon we move away from those people running on "old" releases |
| 21:09.04 | mvanbaak | oh no, all my https certs as well |
| 21:09.09 | mvanbaak | this is BAD |
| 21:09.24 | jsmith | ctooley: It's not moving away from the people, it's just not having resources to look at the old *code* |
| 21:09.28 | ctooley | mvanbaak, welcome to the wonderful world of all new SSL |
| 21:09.40 | jsmith | ctooley: I would hope that everyone would appreciate the difference :-) |
| 21:10.01 | mvanbaak | ctooley: yeah. thanks to the debian team :( |
| 21:10.04 | ctooley | jsmith, I certainly can't we _never_ upgrade. :) |
| 21:10.21 | ctooley | hugs his Fedora box. |
| 21:10.31 | tzafrir | BTW: what's the effect of the recent Debian openssl advisory on RSA keys used with Asterisk? |
| 21:10.33 | ctooley | install, update, get stuff working, never apply updates again. :) |
| 21:10.38 | jmls_home | oh! oh! putnopvut! |
| 21:10.38 | mvanbaak | kicks his 300+ debian boxen |
| 21:10.44 | putnopvut | jmls_home: yessssss |
| 21:10.50 | putnopvut | Did it happen again? |
| 21:10.51 | jmls_home | is excited |
| 21:10.53 | mvanbaak | tzafrir: you have to regenerate them |
| 21:10.55 | tzafrir | (made in Debian systems, that is. It was a Debian-specific screwup) |
| 21:11.06 | jmls_home | no, but I saw a commit 115990 |
| 21:11.12 | Corydon76-dig | putnopvut: Parseltongue? |
| 21:11.23 | putnopvut | Corydon76-dig: I don't get the reference. |
| 21:11.29 | tzafrir | ok, time to go home... |
| 21:11.29 | russellb | jmls_home: that wasn't it |
| 21:11.37 | russellb | jmls_home: but there is an even more fun one coming ... |
| 21:11.38 | mvanbaak | tzafrir: yup, and resign |
| 21:11.39 | jmls_home | bugger |
| 21:11.44 | russellb | i'm not sure if it's going to fix your issue though ... |
| 21:11.46 | Corydon76-dig | putnopvut: Harry Potter |
| 21:11.58 | russellb | how heavily do you use local channels on the setup where that problem was happening? |
| 21:11.58 | putnopvut | Corydon76-dig: ah, I haven't read any of those. |
| 21:12.08 | jmls_home | russellb: all the time |
| 21:12.10 | Corydon76-dig | putnopvut: seen any of the movies? |
| 21:12.14 | russellb | jmls_home: gotcha ... |
| 21:12.14 | putnopvut | Corydon76-dig: nope. |
| 21:12.24 | ctooley | putnopvut is missing out |
| 21:12.28 | putnopvut | jmls_home: you may enjoy the next commit from russell then. |
| 21:12.38 | jmls_home | it is always a local channel that called the macroexclusive |
| 21:12.50 | putnopvut | the key word is "may" ... |
| 21:12.52 | jmls_home | waits in hope ... |
| 21:12.54 | russellb | chan_local </3 |
| 21:13.05 | Corydon76-dig | putnopvut: I'll bring my projector and we can watch the first 3 movies back to back... |
| 21:13.26 | putnopvut | Corydon76-dig: no need to bring your projector. Digium has conf rooms with projectors. |
| 21:14.00 | Corydon76-dig | after work. These require your undivided attention |
| 21:15.06 | mvanbaak | kpfleming: you still here ? |
| 21:15.11 | ctooley | Corydon76-dig, the second one? I'm not so sure on that one. But, by the 3rd, definitely. |
| 21:15.21 | mvanbaak | kpfleming: my wife is ok with the mail you sent about the prompts |
| 21:15.35 | Corydon76-dig | ctooley: Parseltongue figures prominently in the second one |
| 21:15.58 | mvanbaak | I've setup audacity for her and she's going to record all the files in the coming 2 weeks |
| 21:15.59 | ctooley | Corydon76-dig, true, but "undivided attention" isn't necessarily required. |
| 21:16.29 | Corydon76-dig | ctooley: the details will be important to fully understanding everything in the 7th |
| 21:17.19 | Corydon76-dig | It all comes together and you realize the importance of the diary |
| 21:17.21 | ctooley | Corydon76-dig, Assuming that the 7th isn't missing large parts. I think they have done a rather good job, but books 5-7 should really be two movies each. There's just no good place to split them. |
| 21:17.32 | russellb | A116038 |
| 21:17.34 | russellb | jmls_home: ^^^ |
| 21:17.37 | Corydon76-dig | ctooley: or 4 hour movies |
| 21:18.02 | Corydon76-dig | I want another Dances with Wolves type movie... |
| 21:18.20 | Corydon76-dig | or Godfather |
| 21:18.22 | ctooley | _loves_ sitting in a theater for 4 hours with a massive bucket of diet coke and dumpster full of popcorn to make him utterly uncomfortable 3 quarters of the way into it. |
| 21:19.14 | ctooley | At home I have a pause button (and with the new projector and surround sound system, a roughly equivalent movie experience). |
| 21:19.14 | putnopvut | mmmm...diet coke |
| 21:19.30 | russellb | wonders why his last chan_local commits haven't shown up in -commits |
| 21:20.13 | russellb | there it goes. |
| 21:20.16 | russellb | A116038 |
| 21:20.17 | putnopvut | That was slow |
| 21:20.29 | russellb | looks at MuffinMan |
| 21:20.47 | jmls_home | ok, I have no idea what you are talking about in that commit ;) |
| 21:20.49 | jmls_home | wonders at the language that is english. So many words that I don't know ;) |
| 21:20.50 | putnopvut | commit asterisk 116038 |
| 21:20.54 | MuffinMan | [116038] Committed by russell on 2008-05-13T21:17:23.250040Z: Fix a deadlock involving channel autoservice and chan_local that was debugged |
| 21:20.54 | MuffinMan | ..and fixed by mmichelson and me. |
| 21:20.54 | MuffinMan | .. |
| 21:20.54 | MuffinMan | ..We observed a system that had a bunch of threads stuck in ast_autoservice_stop(). |
| 21:20.55 | MuffinMan | ..The reason these threads were waiting around is because this function waits to |
| 21:20.57 | MuffinMan | ..ensure that the channel list in the autoservice thread gets rebuilt before the |
| 21:21.01 | MuffinMan | ..stop() function returns. However, the autoservice thread was also locked, so |
| 21:21.03 | MuffinMan | ..the autoservice channel list was never getting rebuilt. |
| 21:21.06 | MuffinMan | .. |
| 21:21.08 | MuffinMan | ..The autoservice thread was stuck waiting for the channel lock on a local channel. |
| 21:21.11 | MuffinMan | ..However, the local channel was locked by a thread that was stuck in the autoservice |
| 21:21.14 | MuffinMan | ..stop function. |
| 21:21.16 | MuffinMan | .. |
| 21:21.18 | MuffinMan | ..It turned out that the issue came down to the local_queue_frame() function in |
| 21:21.21 | MuffinMan | ..chan_local. This function assumed that one of the channels passed in as an |
| 21:21.24 | MuffinMan | ..argument was locked when called. However, that was not always the case. There |
| 21:21.27 | MuffinMan | ..were multiple cases in which this channel was not locked when the function was |
| 21:21.30 | MuffinMan | ..called. We fixed up chan_local to indicate to this function whether this channel |
| 21:21.33 | MuffinMan | ..was locked or not. The previous assumption had caused local_queue_frame() to |
| 21:21.36 | MuffinMan | ..improperly return with the channel locked, where it would then never get unlocked. |
| 21:21.38 | MuffinMan | .. |
| 21:21.40 | MuffinMan | ..(closes issue #12584) |
| 21:21.41 | Cresl1n | wow |
| 21:21.43 | MuffinMan | ..(related to issue #12603) |
| 21:21.49 | mvanbaak | MuffinMan: please use pastebin for every paste longer than 4 lines ! |
| 21:21.49 | putnopvut | Sorry for the flood. |
| 21:22.28 | ctooley | returns from a PM session with MuffinMan to find that he's flooded the channel as well... convenient to wait on that twice. :) |
| 21:22.57 | jmls_home | goes off to update his 1.4 |
| 21:24.40 | russellb | I like to give verbose commit messages :) |
| 21:24.58 | russellb | because I won't remember what the issue was tomorrow ... |
| 21:26.47 | ctooley | russellb, so I've noticed... maybe I should start submitting literary works along with patches to be used as commit messages too. :) |
| 21:26.49 | mvanbaak | russellb: that's great |
| 21:26.56 | russellb | ctooley: ha, sure! |
| 21:26.57 | mvanbaak | but too bad MuffinMan cannot pastebin it |
| 21:27.56 | jpeeler | pbx_find_extension, should it return something if the exten passed in is null? |
| 21:32.19 | mvanbaak | kpfleming: I cannot make it 'unsigned int' in ast_frame because there are some places that try to set it to '-1' |
| 21:51.17 | mvanbaak | hhmm |
| 21:51.22 | mvanbaak | that union is not working |
| 21:54.10 | mvanbaak | struct ast_frame fr = { AST_FRAME_CONTROL, AST_CONTROL_HANGUP, .data_int = AST_CAUSE_DESTINATION_OUT_OF_ORDER }; |
| 21:54.19 | mvanbaak | that gives a compile error |
| 21:59.53 | mvanbaak | ugh |
| 21:59.56 | mvanbaak | ignore that ;) |
| 22:00.26 | mvanbaak | Wed May 14 00:00:25 CEST 2008 |
| 22:00.37 | mvanbaak | and I only got 1 redbull and 3 cups of coffee |
| 22:10.33 | *** join/#asterisk-dev RoyK (n=roy@ip-99-58-149-91.dialup.ice.no) |
| 22:14.14 | jmls_home | why would I be getting a WARNING[26714]: app_queue.c:3014 try_calling: The device state of this queue member, Single Member Interface, is still 'Not in Use' |
| 22:14.20 | jmls_home | if I am using a local channel ? |
| 22:14.47 | jmls_home | the UPGRADE file talks about call limits on SIP channels only |
| 22:15.04 | putnopvut | jmls_home: right, that's typically why a channel will remain not in use. |
| 22:15.20 | putnopvut | But for a local channel, I'll need to look a bit more closely. |
| 22:15.32 | putnopvut | This is 1.4, correct? |
| 22:15.35 | jmls_home | yup |
| 22:15.54 | jmls_home | I need the local channel to be capable of taking several calls at once through the local channel |
| 22:16.15 | *** join/#asterisk-dev RoyK (n=roy@ip-117-23-149-91.dialup.ice.no) |
| 22:16.18 | putnopvut | Then, if I'm not mistaken, it's not exactly a bad thing that it's still not in use. |
| 22:16.31 | jmls_home | I have a "single queue member" that uses a local channel, which uses curl to get the "real" agent to send the call to |
| 22:16.36 | *** part/#asterisk-dev Cresl1n (n=matt@216.207.245.1) |
| 22:17.51 | jmls_home | putnopvut: right |
| 22:18.10 | putnopvut | Hmmm, I just tried testing it and my local channel member is stuck invalid and calls aren't routed to him. |
| 22:18.11 | putnopvut | Crud. |
| 22:18.16 | jmls_home | heh |
| 22:19.32 | jmls_home | after speaking to ctooley, I've had a few ideas on queues |
| 22:20.03 | jmls_home | the only reason I am using a queue now is A) to ensure that the oldest call gets answered first |
| 22:20.23 | jmls_home | B) I can play announcements to the caller |
| 22:20.42 | jmls_home | oh, and C) play MOH |
| 22:21.14 | jmls_home | using audio hooks in 1.6, would it be possible to interject announcements to the caller while they play MOH ? |
| 22:21.55 | putnopvut | jmls_home: that may be possible. |
| 22:22.00 | jmls_home | mmm |
| 22:22.11 | putnopvut | I guess you could put a whisper audiohook on the channel and periodically write frames to it. |
| 22:22.27 | putnopvut | The trick would be to interrupt the music in order to do it. |
| 22:22.53 | putnopvut | Because I doubt you'd want to hear the music and the announcement at the same time. |
| 22:23.08 | jmls_home | what would be _really_ cool would be to soften the moh and play the announcement over it |
| 22:23.18 | putnopvut | ooh, that would be cool. |
| 22:24.13 | jmls_home | the MOH application could take an extension and time period parameter |
| 22:24.24 | jmls_home | the extension would be called every "n" |
| 22:25.11 | jmls_home | this extension could then do "s,1,InjectAudio(-5,somefile) |
| 22:25.16 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 22:25.30 | jmls_home | where -5 is "lower decibles on current audio by -5" |
| 22:25.46 | *** part/#asterisk-dev RoyK (n=roy@ip-117-23-149-91.dialup.ice.no) |
| 22:26.10 | jmls_home | by calling the extension, you could use whatever dialplan wizardry to determine the audio file to play |
| 22:26.59 | putnopvut | That could be a nifty feature, and if you were going to do that, I think a set of audiohooks could accomplish it quite easily. |
| 22:27.07 | jmls_home | yeah .. |
| 22:27.23 | putnopvut | But it may need the "chaining" that file has talked about before. |
| 22:27.33 | jmls_home | ?? |
| 22:27.33 | putnopvut | I think that's what he called it. |
| 22:28.34 | jmls_home | the "whisper" on chanspy injects audio into the stream already |
| 22:28.34 | putnopvut | I'm just thinking that you'd need to be sure that you didn't also lower the volume of the announcement. |
| 22:28.34 | putnopvut | Just the music. |
| 22:28.34 | jmls_home | OIC |
| 22:28.34 | putnopvut | And that would require that the audiohooks are applied in the proper order, I think. |
| 22:28.46 | jmls_home | so each audiohook has a volume control ? |
| 22:29.12 | jmls_home | I tell you, this would be such a drop-dead killer feature ;) |
| 22:29.20 | putnopvut | The "manipulate" audiohook type can be used to control volume, and then the "whisper" one would actually inject audio. |
| 22:29.30 | putnopvut | You'd have to be certain that they went in that order. |
| 22:29.43 | putnopvut | Otherwise, you'd inject audio and then it would all have the volume lowered. |
| 22:29.43 | jmls_home | the application should take care of that |
| 22:30.10 | putnopvut | Right, it would, I'm just thinking out loud how it would be implemented. |
| 22:30.17 | jmls_home | InjectAudio would add "manipulate" and "whisper" in the correct order |
| 22:30.46 | jmls_home | eep. you could probably inject DTMF tones :p |
| 22:32.07 | jmls_home | putnopvut: actually, thinking about it, it would be a better idea to use an application to start and end audio injection, rather than a parameter to an application |
| 22:32.42 | jmls_home | s,n,StartInjection(context,extension,n) |
| 22:32.53 | jmls_home | s,n,EndInjection() |
| 22:33.36 | putnopvut | Oh, so you could inject multiple audio files? |
| 22:33.44 | jmls_home | that way you could add this feature to all apps |
| 22:34.06 | putnopvut | Ah, so something like Playback() could be injected automatically |
| 22:34.26 | jmls_home | yes |
| 22:34.58 | jmls_home | s,n,StartInjection(context,extension,Seconds,Volume) |
| 22:35.15 | jmls_home | then all the extension needs to do is play some audio |
| 22:36.03 | Corydon76-dig | s,n,HotBeefInjection(length,girth,Seconds,Volume) |
| 22:36.09 | jmls_home | for example, whilst waiting for READ or something, of you have a 10s timeout, you can start warning them "time is nearly up" |
| 22:36.10 | jmls_home | lol |
| 22:36.24 | putnopvut | oh Corydon76-dig, you card. |
| 22:36.37 | jmls_home | blushes |
| 22:36.59 | putnopvut | I could use a hot beef injection...by that I mean that I wouldn't mind eating a steak. |
| 22:37.01 | Corydon76-dig | giggles |
| 22:37.52 | Corydon76-dig | putnopvut: you should check the topic of #asstricks |
| 22:38.53 | putnopvut | hehe |
| 22:40.28 | putnopvut | Corydon76-dig: I especially like the Juggie quote, because it sounds REALLY bad out of context. |
| 22:40.54 | Corydon76-dig | Doesn't it? ;-) |
| 22:41.02 | *** part/#asterisk-dev PTorres (n=PTorres@200.68.87.146) |
| 22:43.24 | Corydon76-dig | CoaxD used to hang and we'd get him to say all sorts of things... |
| 22:44.18 | Corydon76-dig | and then as soon as we put them in the topic, he realized his mistake |
| 22:51.39 | *** join/#asterisk-dev RoyK (n=roy@ip-117-23-149-91.dialup.ice.no) |
| 22:52.08 | *** part/#asterisk-dev RoyK (n=roy@ip-117-23-149-91.dialup.ice.no) |
| 23:03.37 | tzanger | coaxd... now that is a name that sounds very familliar |
| 23:05.32 | *** join/#asterisk-dev jeffspeff2 (n=jeff@c-68-53-81-73.hsd1.ky.comcast.net) |
| 23:17.20 | Corydon76-dig | tzanger: at one point, we got him to say something about getting dick in his ass once every other day |
| 23:17.30 | tzanger | hahahaha |
| 23:17.32 | tzanger | I did not know that |
| 23:18.18 | Corydon76-dig | tzanger: he was hypothetically recoiling from getting it every day, but it still was funny the way he said it |
| 23:19.28 | Corydon76-dig | Something like "I think I could keep it down to dick in the ass once every other day." |
| 23:21.09 | tzanger | heh |
| 23:31.57 | *** join/#asterisk-dev jeffg (i=MCsnZc28@pdpc/supporter/active/jeffg) |
| 23:58.54 | *** join/#asterisk-dev ZaVoid (n=zavoid@c-67-165-25-195.hsd1.pa.comcast.net) |
