IRC log for #asterisk on 20190822

00:18.49*** join/#asterisk infobot (ibot@c-174-52-60-165.hsd1.ut.comcast.net)
00:18.49*** topic/#asterisk is Take the March 2019 Asterisk User Survey! https://goo.gl/forms/xL1VUHRsf95saly13 -- #asterisk The Open Source PBX and Telephony Platform (asterisk.org) -=- LTS: 13.28.0 (2019/7/25) 16.5.0 (2019/7/25), Security Only: 15.7.3 (2019/7/11); DAHDI: 3.0.0 (2018/11/15); libpri 1.6.0 (2017/01/27) -=- Wiki: wiki.asterisk.org -=- Code of Conduct: bit.ly/1hH6P22
00:22.37sawgoodDigium Staff: why do you not allow SSH into your wonderful Digium FXO/FXS metal box gateways: it would be so nice to see Asterisk live ... its depressing to me that you don't allow it
00:36.59slackacan't you just log into your pbx box and run the asterisk shell?
00:37.12slackaor are you talking about a hosted pbx ?
00:38.10sawgoodslacka: I'd like to SSH-in to the metal box FXS Digium IP gateway itself
00:38.17sawgoodseperate box from the PBX
00:41.00SamotBecause there isn't anything looking at the asterisk logs the GUI they built for it isn't going to tell you.
00:41.24SamotPlus proprietary stuff, I'm guessing.
00:49.20slackaok taking a break from my other computer issue. couldn't upgrade to 3.3 either, but upgrading from 2.1 to 2.2 looks like it might be working
00:49.28slacka#babysteps
00:54.48slackathat upgraded the sip version but not bootrom or bootblock.
02:25.38*** join/#asterisk infobot (ibot@c-174-52-60-165.hsd1.ut.comcast.net)
02:25.38*** topic/#asterisk is Take the March 2019 Asterisk User Survey! https://goo.gl/forms/xL1VUHRsf95saly13 -- #asterisk The Open Source PBX and Telephony Platform (asterisk.org) -=- LTS: 13.28.0 (2019/7/25) 16.5.0 (2019/7/25), Security Only: 15.7.3 (2019/7/11); DAHDI: 3.0.0 (2018/11/15); libpri 1.6.0 (2017/01/27) -=- Wiki: wiki.asterisk.org -=- Code of Conduct: bit.ly/1hH6P22
02:28.16*** join/#asterisk ih8wndz (jwpierce3@001.srv.trnkmstr.com)
03:40.02*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
04:56.12*** join/#asterisk gerhard7 (~gerhard7@ip5657ee30.direct-adsl.nl)
05:39.14*** join/#asterisk tomaluca95 (~quassel@kde/developer/tomaluca)
05:40.58*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
05:42.25*** join/#asterisk AndyCap (~aoy@pdpc/supporter/sustaining/AndyCap)
05:51.45*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
06:01.43*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
06:07.54*** join/#asterisk pchero_work (~pchero@87.213.247.82)
06:08.50*** join/#asterisk pchero_work (~pchero@87.213.247.82)
06:45.33*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
06:46.40*** join/#asterisk cp (~cp@b157153.ppp.asahi-net.or.jp)
06:49.01*** join/#asterisk cp (~cp@b157153.ppp.asahi-net.or.jp)
07:20.53*** join/#asterisk jkroon (~jkroon@41.113.113.250)
07:22.30*** join/#asterisk jkroon (~jkroon@41.113.113.250)
07:43.17*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
07:55.14*** join/#asterisk kamyl (~user@unaffiliated/kamyl)
08:25.26*** join/#asterisk sekil (~sekil@109-93-36-108.dynamic.isp.telekom.rs)
08:46.08*** join/#asterisk wraythe (wraythe@2a01:7e00::f03c:91ff:fe42:8645)
09:56.57wraytheIs there a method to debug why Asterisk is consuming so much CPU?
09:57.09wraytheno dialplan active, no trunks, 1 extension
09:58.26wraythehttps://snipboard.io/HOREz0.jpg
09:58.46wraytheI'm running 2 Asterisk 16 instances in Docker, (Ubuntu 18.04 base)
09:59.07wraytheno real logic in it right now, but for some reason it's sucking up more CPU than chrome on a busy day :P
10:08.04*** join/#asterisk hehol (~hehol@130.180.55.250)
10:15.47*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
10:17.42wdoekeswraythe: I'd begin with an: strace -fp `pidof asterisk`
10:20.08wdoekesalternately, isolate specific threads using cpu with: top -H -p `pidof asterisk`
10:20.22wdoekesand then attach gdb to see what that thread is doing
10:45.43*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
10:55.07*** join/#asterisk ruben23 (~Opendial@49.151.159.26)
11:14.49*** join/#asterisk emsjessec (~emsjessec@pool-173-54-255-231.nwrknj.fios.verizon.net)
11:37.53*** join/#asterisk bengoa (~alberto@194.50.55.200)
11:43.28wraythethanks - will do :)
12:06.15*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
12:15.25*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
12:38.30*** join/#asterisk brad_mssw (~brad@66.129.88.50)
12:39.28*** join/#asterisk K0HAX (~michael@gateway/tor-sasl/k0hax)
12:47.30*** join/#asterisk [devil-boy] (~dude@lfbn-bor-1-188-9.w90-50.abo.wanadoo.fr)
12:55.28*** join/#asterisk [TK]D-Fender (~joe@216.191.106.165)
12:57.27*** join/#asterisk tomaluca95 (~quassel@kde/developer/tomaluca)
13:01.05*** join/#asterisk gerhard7 (~gerhard7@ip5657ee30.direct-adsl.nl)
13:12.40*** join/#asterisk lagzilla (lagzilla@2600:3c03::f03c:91ff:fe6e:5ed9)
13:13.07lagzillaUsing mysql (I realize it's deprecated) is it possible to get multiple rows as an array?
13:16.02sibiriaif you're asking strictly about mysql, you can concatenate multiple rows into one result, as for example CSV
13:16.04*** join/#asterisk NWATechSupport (~NWARSciGu@wsip-72-204-193-12.ks.ks.cox.net)
13:16.09sibiriato produce something that ARRAY() can interpret
13:16.35sibiriain mariadb/mysql this aggreggation is done using GROUP_CONCAT()
13:18.27*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
13:25.51*** join/#asterisk bford (uid283514@gateway/web/irccloud.com/x-gqmdqqwotooorrfd)
13:25.51*** mode/#asterisk [+o bford] by ChanServ
13:27.51*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
13:30.18[TK]D-FenderOr you could just do it in AGI
13:35.06*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
13:51.06*** join/#asterisk sekil (~sekil@nat-73.net011.net)
14:03.01*** join/#asterisk jkroon (~jkroon@41.113.113.250)
14:03.11*** join/#asterisk Samot (sid133316@gateway/web/irccloud.com/x-giqcjkbqdclngsar)
14:33.31*** join/#asterisk kharwell (uid358942@gateway/web/irccloud.com/x-nzfojghravsuihwa)
14:33.31*** mode/#asterisk [+o kharwell] by ChanServ
14:37.02*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
14:44.58*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
15:02.31*** join/#asterisk scgm11_ (~scgm11@r186-52-238-223.dialup.adsl.anteldata.net.uy)
15:15.09*** join/#asterisk emsjessec (~emsjessec@pool-173-54-255-231.nwrknj.fios.verizon.net)
15:25.18*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
15:39.05*** join/#asterisk ganbold (~ganbold@66.85.186.234)
15:47.30*** join/#asterisk scgm11_ (~scgm11@r186-52-238-223.dialup.adsl.anteldata.net.uy)
15:51.25*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
15:54.13scgm11_is there any known problem using two mixmonitors on the same channel? for example to be able to have 2 recordings one full and other a part of the same call? any locking issue or anything like that?
15:55.56SamotWhy would you do that?
15:57.07scgm11_so I have in other file that part of the recording and a user can define where it begins and where it ends
15:58.53igcewielinguse post processing to extract the partial recording from the complete recording and/or each call has 2 channels.  you can put one mixmonitor on each channel.
15:59.47scgm11_post processing is more resource intensive for many files
16:00.27sibiriait's not really demanding for sox to rip part of a wav file out
16:00.55scgm11_Im using gsm not wav
16:01.17sibiriasame. it's not a variable bitrate format
16:01.35scgm11_so you think is better to postprocess those files?
16:02.04sibiriai don't know of you can run two mixmonitors at the same time but something about the idea irks me
16:02.20sibiriaso i'd just attach a shell script via MONITOR_EXEC
16:02.35*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
16:03.31sibiria(or whatever the environment variable is that MixMonitor wants)
16:03.38scgm11_Im concerned about locking as mixmonitor is attached to the stream mixmonitor is not doing the same as monitor
16:05.30scgm11_but I may try post processing
16:05.57sibiriaasterisk *should* spawn a separate thread for the command you supply to MixMonitor
16:06.12sibiriaif you're worried you could "nice" sox to make it run at lower prio
16:06.39scgm11_Im more concerned of IO rather than cpu
16:06.52scgm11_not sure if nice would change that
16:07.08sibiriaare you running thousands of calls in parallel?
16:07.16sibiriaif the answer is "no, hundreds" then don't worry
16:07.34scgm11_it could go up to more than a thousend channels
16:07.54scgm11_anyway not all bridged
16:08.02sibiriais the machine on a prev. generation IDE port, or on an old mechanical drive?
16:08.20*** join/#asterisk techquila (~techquila@2407:7000:9125:e400:a453:6053:e9c0:bace)
16:08.35sibiriaor some incredibly low-spec IO-rated SSD setup on AWS EC2
16:08.53scgm11_could be on cloud environments (GCP) or on premise hardware
16:09.16igcewielinglooks around and discretely hugs his NVMe drives.
16:09.17sibiriaeven at a thousand parallel calls it's unlikely even half that number of channels will hang up the same second
16:09.38sibiriai wouldn't worry about the i/o for this
16:10.34scgm11_ok I will check wich is the mixmonitor variable for after processing and do that on a bash script using sox
16:10.49scgm11_and see how it goes
16:11.09sibiriait may be an env variable like Monitor(), or maybe it's a dedicated parameter, can't recall
16:11.12sibiriabut same same
16:11.23sibiriait offers the same functionality
16:11.40igcewielingyou could even move the files to another system for postprocessing
16:12.34scgm11_Im trying to avoid to move the fiels to other system
16:15.35scgm11_thanks for the feedback
16:19.13SamotOK so the goal here is to have a fully recorded call but then another recording for just the agents portion of the call?
16:20.14sibiriai think he wants to extract snippets in the middle of the call, rather than just two separate legs of the same call
16:20.26sibiria(surely he knows that the latter is just one parameter change away)
16:20.42*** join/#asterisk dakudos (~dakudos@c-73-217-33-206.hsd1.co.comcast.net)
16:21.46scgm11_the goal is having one full record and separated parts of the recording selected by the user
16:22.28SamotHow does the user select the portions?
16:23.05sibiriathey fill in a form and mail it to a poste restante in Alaska
16:23.14sibiriai mean it's slow but it works
16:23.18scgm11_the implementation I can do in different ways, setting variables on the channel via the UI for the start stop of each
16:23.36scgm11_part
16:23.40SamotBut how does the user select the portions?
16:23.51SamotSo you'll have a GUI for them?
16:23.52scgm11_press a buton to start stop that part
16:23.56scgm11_sure
16:24.12scgm11_there is a gui for them to select when start and when stop that part
16:24.13SamotSo you'll need a method to track the monitor IDs.
16:25.08scgm11_I dont really need that I have everything I need becuase I know the channel
16:25.20scgm11_so I can add varaibles and check those in the dialplan
16:25.45SamotNo.
16:25.50SamotFor the stopping of the call.
16:26.09SamotTo stop the MixMonitor you need to know the exact MixMonitorID or it will kill them all.
16:26.26scgm11_Im not stopping the mixmonitor
16:26.31scgm11_that will stop at the end of the call
16:26.48SamotYes but if the user is doing it...
16:26.53Samotthen it's on _their channel_
16:26.59SamotNot the channel the call originated on.
16:27.04scgm11_the user is just adding new variables into the channel
16:27.06SamotSo now you'll have it on two channels.
16:27.15SamotNo, to _their channel_
16:27.22SamotNot the channel the call came in on.
16:27.51*** join/#asterisk lbazan (~lbazan@fedora/LoKoMurdoK)
16:27.51SamotIf I call ext 100, the MixMonitor is on my channel.
16:28.01SamotIf 100 hits it, then the MixMonitor is on their channel.
16:28.05*** join/#asterisk slacka (~slacka@unaffiliated/slacka)
16:28.06scgm11_is on a bridge the variable will be __ so I can even set them in both channls
16:28.10slackahowdy
16:28.29SamotMixMonitor command executes on the channel.
16:28.44SamotNo variable to select a channel.
16:28.51SamotJust the channel it was called on.
16:29.10scgm11_I know that but to post process the filesdescriptor is closed
16:29.26scgm11_so I really dont care wich channel executes the post process part
16:29.40scgm11_and I know the exact file
16:29.40SamotYou'll have two.
16:29.45SamotThat's the point.
16:29.47SamotIt's two files.
16:29.55SamotTwo MixMonitor commands so two post functions.
16:29.58scgm11_no is just one, I never stop the mixmonitor
16:30.11SamotYou're not following what I am saying.
16:30.20scgm11_probably
16:30.30SamotYou have an inbound call that comes in over your trunk...
16:30.50SamotYou start the MixMonitor on it. That is one call to the function with the filename, options and post command.
16:30.56SamotIt will be on THAT CHANNEL.
16:31.06SamotNow you ring ext 100, they answer the call.
16:31.24SamotThey hit the button to start MixMonitor, that is another call with another filename, options and post command.
16:31.32SamotOn ext 100's channel.
16:31.45scgm11_Im not starting mixmonitor or stoppin it on the ext 100
16:32.03sibiriait will be two files only if you tell mixmonitor that you want it to keep us/them separated
16:32.14scgm11_the hit button will just add a variable that is for example:   startpart:6 (6 is the second)
16:32.17Samot12:21:48 PM <scgm11_> the goal is having one full record and separated parts of the recording selected by the user
16:32.17Samot12:23:54 PM <scgm11_> press a buton to start stop that part
16:32.17Samot12:23:59 PM <scgm11_> sure
16:32.17Samot12:24:14 PM <scgm11_> there is a gui for them to select when start and when stop that part
16:32.59scgm11_start stop part is just from a functional point of view Im just setting variables on the channel
16:32.59SamotOK, so you're going to record one file then split it up based on the variables sent in the post?
16:33.17scgm11_that is what some of you were suggesting here
16:33.33SamotSo you'll need to use MASTER_CHANNEL
16:33.38scgm11_my first aproach was doing parts with a mixmonitor over other mixmontor
16:34.03scgm11_not really in my backend I know both channels and the bridge so I can set the variables anywhere I want
16:34.14SamotOK.
16:35.31SamotJust remember _ and __ are not setting global variables.
16:35.40scgm11_I know
16:36.08SamotThey are setting up inheritance for the child channels. So if you set the var on the agents channel, it doesn't go on the master channel you're running mixmonitor on.
16:36.58scgm11_that is why as I know the bridge I can set those variables in the channel that mixmonitor is attached
16:37.36Samotso you've tested that?
16:37.47scgm11_also mixmonitor uses the audiohook not like monitor
16:37.53sibiriakeep in mind that variables for MixMonitor are evaluated when the function call occurs
16:38.31sibiriathe command portion isn't evaluated separately when mixmonitor ends
16:38.50SamotThere is that.
16:39.08sibiriaso you may very well be forced to do this after mixmonitor, or handle it at hangup
16:39.29sibiria(when you have collected "their" start/stop points for the audio excerpt)
16:39.32scgm11_yes that is the idea as I will be filling variables into the channel at hangup time I will process that
16:39.43scgm11_and create the parts
16:39.56scgm11_I need on a bash script passing all the variables I need
16:41.22scgm11_have some experience touching app_mixmonitor, app_monitor, app_queue and others so I know a bit about the internals if I need some change there I can do it
16:42.08scgm11_my original thoughts was use 2 mixmonitors but I was concerned about locking
16:42.37scgm11_as mixmonitor uses the audiohooks as chan_spy I think there wont be an issue there but.....
16:43.20scgm11_maybe the question was more suitable for the dev channel but as some of you mentioned post processing I will try it
16:46.02scgm11_and the suggestion to use MASTER_CHANNEL is ok too if I dont want to check by myself where to add the variables
16:47.20scgm11_thanks
16:48.17SamotWhy is this something for the dev channel?
16:48.23SamotAre you developing something for Asterisk?
16:48.39SamotLike Asterisk itself? Or just something that is using Asterisk as is?
16:49.09scgm11_becuase my question was about if there was any locking issue know if I use more than one mixmonitor on the same channel
16:49.33SamotFunny story, Asterisk devs are ops in this channel.
16:49.45scgm11_I know
16:49.50SamotSo I'm sure one can pipe in about your question here.
16:50.04scgm11_that is why I put that here
16:50.18SamotBut why do you think you're going to have a lock?
16:50.20scgm11_I wasnt thinking in post processing
16:53.14scgm11_because there is a locking mechanism in audio hooks, using ast_audiohook_lock .. attach,
16:53.34fileeach MixMonitor has its own audiohook
16:53.37scgm11_and wasnt sure If it was meant to have more than one
16:54.14fileI can't think of any reason it wouldn't work
16:54.17scgm11_ok so wont collide or have any locking issue using more than one? that was my original question
16:54.26fileno
16:55.12filethe locking itself would not be a problem
16:55.36scgm11_ok that is what I wanted to make sure as Its not that east to be sure looking app_mixmonitor, beacuse I had some issues using mixmonitor and monitor togheter in the past
16:55.46scgm11_saw one deadlock and some dumps
16:56.04filethat would be unrelated.
16:56.57scgm11_ok, just wanted to make sure if there were any known issues using more than one
16:57.59fileany known bugs are always in JIRA, if not then either someone hasn't reported it, it doesn't exist, or it's been fixed
17:00.02scgm11_thanks file
17:11.26*** join/#asterisk dacod (~dacod@187.103.104.42)
17:11.51*** join/#asterisk scgm11_ (~scgm11@r186-52-238-223.dialup.adsl.anteldata.net.uy)
17:35.05slackawoohoo, finally got this damn phone updated to most recent.
17:35.28slacka10+ firmware upgrades later and 1 bootrom update later.
17:35.39slackafinally has a web interface.
17:45.44*** join/#asterisk jkroon (~jkroon@165.16.203.109)
17:56.36*** join/#asterisk as3ty (uid332392@gateway/web/irccloud.com/x-uwuwgeyjnmgvtgwz)
18:13.06*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
18:43.50*** join/#asterisk zapata (~zapata@2a02:1748:f71:380:bcef:d6c6:e793:50ce)
18:59.42*** join/#asterisk scgm11_ (~scgm11@r186-52-238-223.dialup.adsl.anteldata.net.uy)
19:13.55*** join/#asterisk jkroon (~jkroon@165.16.203.109)
20:13.04*** join/#asterisk NWATechSupport (~NWARSciGu@wsip-72-204-193-12.ks.ks.cox.net)
20:23.04*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
21:13.52*** join/#asterisk MLC (~MLC@63.249.40.11)
21:15.52MLCOn my local network, I have an asterisk server that registers with voip.ms (a sip provider).  When it first registers, everything is fine and I can call in and out no problem.  After a while, I can't call in or out.  I look at the state table in the firewall and the port that voip.ms thinks I'm registered on has disappeared from the state table.  Seems like maybe the firewall is be killing that state/connection because it is idle? Is t
21:15.52MLChere a way for asterisk to send some kind of "keep alive"? I set keep_alive_interval=20 in pjsip.conf, but it didn't make any difference.
21:19.57wraythequalify=yes
21:19.59wraythe?
21:21.26wraytheas long as your firewalls udp session timeout is longer than the default qualify interval, it should retain the session in its state table
21:21.42MLCis that a setting in pjsip.conf?
21:22.39SamotBecause voip.ms has no responsibility to respond to keep alives.
21:22.56SamotThe setting may not matter.
21:23.26MLCI see qualify_frequency - is that it?
21:25.28SamotYes
21:31.16MLCSeems to be doing the trick. Thank you both.
21:37.23wraytheok cool
22:23.02ruben23anyoen cna recommed a low cost voip provider.?
22:23.18MLCvoip.ms has been pretty good for me
22:24.40igcewielingI use vitelity.net
22:25.05igcewielingsince 2001, but it is very low volume.
22:25.12SamotI wouldn't count Vitelity as low cost unless you have a wholesale account.
22:33.31*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
22:36.53sibiriaruben23: a lot of it depends on your destinations
22:37.01sibiriasinch have good prices within europe
22:42.37*** join/#asterisk stevedavies (~stevedavi@197.155.252.3)
23:39.48ruben23sibiria: desitination is US
23:57.18*** join/#asterisk [TK]D-Fender (~joe@64.235.216.2)

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