01:18.30 | *** join/#asterisk ipengineer (~zconkle@108-206-120-218.lightspeed.mckntx.sbcglobal.net) |
01:29.09 | *** join/#asterisk acidfu_ (~acidfoo@24-212-247-227.cable.teksavvy.com) |
02:29.12 | *** join/#asterisk pppingme (~pppingme@unaffiliated/pppingme) |
02:35.58 | *** join/#asterisk jkroon (~jkroon@2c0f:f720:1:0:fcb3:e235:8b0a:1f6b) |
02:38.25 | *** join/#asterisk fstd_ (~fstd@unaffiliated/fisted) |
02:38.35 | *** join/#asterisk Oatmeal (~Suzeanne@99-103-96-198.lightspeed.iplsin.sbcglobal.net) |
02:46.59 | *** join/#asterisk celer (~celer@pc1.robertwaltersjp-unet.ocn.ne.jp) |
02:47.22 | *** join/#asterisk dobson (~dobson@68.ip-149-56-14.net) |
02:47.34 | *** join/#asterisk O47m341 (~Suzeanne@99-103-96-198.lightspeed.iplsin.sbcglobal.net) |
02:57.19 | *** join/#asterisk hiyo (~hiyo@unaffiliated/hiyo) |
03:09.32 | hiyo | Hello I am having a problem with calling over OpenVPN, I get silent calls. However when in the LAN of the server, it works fine |
03:10.22 | *** join/#asterisk genpaku (~genpaku@107.191.100.185) |
03:16.10 | *** join/#asterisk boris_t (~boris_t@363103629.convex.ru) |
03:16.10 | *** join/#asterisk genpaku (~genpaku@107.191.100.185) |
03:30.44 | *** join/#asterisk chazzam (~chazz@vps.roguewerks.com) |
03:34.44 | *** join/#asterisk SunGod (~SunGod@92.55.116.125) |
03:47.29 | *** join/#asterisk SunGod (~SunGod@92.55.116.125) |
03:50.41 | *** part/#asterisk hiyo (~hiyo@unaffiliated/hiyo) |
04:36.06 | *** join/#asterisk Oatmeal (~Suzeanne@99-103-96-198.lightspeed.iplsin.sbcglobal.net) |
04:59.02 | *** join/#asterisk bof22 (~Thunderbi@185.13.183.107) |
04:59.05 | *** join/#asterisk babak (uid19622@gateway/web/irccloud.com/x-ldfxgrmaopkvtipr) |
05:19.44 | *** join/#asterisk ruben23 (~OpenDIAL@112.201.10.245) |
05:20.38 | ruben23 | sterisk*CLI> dahdi show channels -----> No such command âdahdi show statusâ (type âcore show help dahdi showâ for other possible commands) |
05:21.12 | ruben23 | * asterisk*CLI> dahdi show channels |
05:21.16 | ruben23 | any ide aguys |
05:27.05 | *** join/#asterisk miralin (~Thunderbi@194.8.128.48) |
05:45.43 | *** join/#asterisk Rasputin3711 (~Rasputin3@87.255.254.66) |
05:58.07 | *** join/#asterisk miralin (~Thunderbi@195.19.212.23) |
06:25.18 | *** join/#asterisk bof22 (~Thunderbi@185.13.183.107) |
06:33.12 | *** join/#asterisk salz212 (2ac9eee3@gateway/web/freenode/ip.42.201.238.227) |
06:47.04 | *** join/#asterisk tirej (~errorist_@95.9.123.3) |
07:01.07 | *** join/#asterisk tzafrir (~tzafrir@local.xorcom.com) |
07:02.06 | *** join/#asterisk TriJetScud (znc@van-app-svr.ad.v10networks.ca) |
07:02.06 | *** join/#asterisk nanoha-sama (znc@van-app-svr.ad.v10networks.ca) |
07:05.00 | *** join/#asterisk hehol (~hehol@gatekeeper.loca.net) |
07:06.06 | *** join/#asterisk pchero_work (~pchero@109.70.54.56) |
07:06.47 | *** join/#asterisk bmg505 (~leon@165.255.122.53) |
07:18.57 | *** join/#asterisk CeBe (~CeBe@a81-14-235-90.net-htp.de) |
07:37.41 | ChannelZ | ruben23, yeah.. DAHDI isn't loaded |
07:43.31 | *** join/#asterisk O47m341 (~Suzeanne@99-103-96-198.lightspeed.iplsin.sbcglobal.net) |
08:02.01 | *** join/#asterisk jkroon (~jkroon@uls-154-73-35-201.wall.uls.co.za) |
08:11.43 | *** join/#asterisk todwetsprock (~florian@p5099a594.dip0.t-ipconnect.de) |
08:12.23 | *** join/#asterisk teaspoon (~teaspoon@ppp59-167-135-77.static.internode.on.net) |
08:15.08 | teaspoon | Is it possible to dynamically create peers and add them to a hunt group? |
08:16.08 | teaspoon | The example I have in mind right now is for an asterisk+iaxmodem+hylafax monstrosity, where I want to be able to spin up a couple of extra iaxmodem instances without having to update iax.conf |
08:52.35 | *** join/#asterisk tirej (~errorist_@95.9.123.3) |
09:00.03 | *** join/#asterisk bounceman (~bounceman@185.32.9.250) |
09:00.20 | bounceman | Hi, if I enable tls in manager.conf. Will that only make it possible to do requests over https? |
09:13.23 | *** join/#asterisk matrix1233 (~matrix123@197.0.177.30) |
09:13.32 | *** join/#asterisk puzzled (~puzzled@2001:982:1097:1::1:3) |
09:25.57 | bounceman | Earth is calling |
09:27.02 | *** join/#asterisk SunGod (~SunGod@92.55.116.125) |
09:27.37 | *** join/#asterisk infernix (nix@unaffiliated/infernix) |
09:58.51 | *** join/#asterisk celer (~Celer@2400:2410:27c2:3200:e35d:6073:93b:b88b) |
10:03.08 | *** join/#asterisk defswork (~aporter@212.38.66.116) |
10:04.22 | *** join/#asterisk wasanzy (~wasanzy@41-66-254-58-dedicated.4u.com.gh) |
10:04.25 | wasanzy | Hi |
10:04.51 | wasanzy | is there a way to determine if an audio has not finished playing? |
10:22.42 | *** join/#asterisk sekil (~sekil@nat-73.net011.net) |
10:35.15 | jkroon | hi all, using htop I'm seeing that at least one asterisk thread is starting to approach 100% cpu. what can I do to figure out what that thread is. |
10:36.27 | jkroon | I'm afraid that it's sporadically hitting 100% and causing quality problems, so I'd like to confirm which thread this is so that I can determine what I need to do to fix the problem> |
10:37.19 | Samot | What do the logs say is happening? |
10:37.29 | Samot | What are the actual resources being used? |
10:38.41 | jkroon | samot, overall CPU >= 100% (meaning more than one core is being consumed). |
10:38.52 | jkroon | logs themselves are all the normal logs that I expect to see. |
10:39.21 | jkroon | on RTCP stats I'm seeing some jitter, but I only started logging that *after* I first saw the problem (which is varying). |
10:39.38 | Samot | How many calls? |
10:39.54 | jkroon | CPU is the only resource that I'm concerned about based on iotop, iostat, top, htop and iftop. |
10:39.54 | Samot | What activity is happening on the box right now? |
10:40.02 | jkroon | 112 active calls - 24238 calls processed |
10:40.07 | jkroon | combination of IAX and SIP channels. |
10:40.28 | Samot | What's the resources on the box? |
10:40.39 | Samot | How much RAM and CPU? |
10:40.44 | jkroon | i'm aware that SIP is running in a single thread, but that's for port 5060 only, and shouldn't disturb RTP - which is what it seems is being disturbed. |
10:41.39 | jkroon | box has 10G RAM, of whch free -m is saying 8.2G is being used, with 486M free, and 4M shared, buff/cache @ 1.2G and 1.2G available. |
10:42.05 | jkroon | according to top asterisk is using a mere 240MB of RAM. |
10:42.22 | Samot | CPU? |
10:42.38 | jkroon | Intel(R) Xeon(R) CPU E5420 @ 2.50GHz - 8 threads. |
10:42.50 | Samot | Single core? |
10:42.55 | Samot | Dual core? |
10:42.59 | jkroon | well, 4 cores with 2 threads IIRC ... let me check google. |
10:44.07 | wasanzy | how do I determine if a user terminate a call in asterisk? |
10:44.10 | Samot | ~pb your top |
10:44.16 | Samot | ~pb |
10:44.17 | infobot | it has been said that pastebin is a web-based service where you should paste anything over 3 lines so you don't flood the channel. Here are links to a few: http://pastebin.ca, http://channels.debian.net/paste, http://paste.lisp.org, http://bin.cakephp.org/; or install pastebinit with yum or aptitude. |
10:44.58 | Samot | wasanzy: You mean you want to know if the user (extension) ended the call or the other party? |
10:45.07 | *** join/#asterisk Iamnach0 (~Iamnacho@ip24-252-4-195.om.om.cox.net) |
10:45.19 | jkroon | 4 cores as per http://ark.intel.com/products/33927/Intel-Xeon-Processor-E5420-12M-Cache-2_50-GHz-1333-MHz-FSB |
10:45.20 | wasanzy | Samot: yes |
10:45.36 | jkroon | top |
10:45.44 | jkroon | sorry, wrong window |
10:45.54 | wasanzy | eg, an audio could be playing and in the middle, a user can decide to hang the call. I want to know if that happens |
10:46.39 | jkroon | the task id of the thread consuming the most CPU is the same as the main pid. |
10:46.59 | Samot | Show it. |
10:47.09 | *** join/#asterisk zapata (~zapata@2a02:b18:581:10:f072:ea29:7a0e:e3d1) |
10:47.30 | Samot | wasanzy: It's in the logs. You're going to have to catch things on the hangup.. |
10:47.34 | Samot | in the h extension. |
10:47.51 | jkroon | and seemingly i can't c&p htop ... https://pastebin.co.za/5706163895140352 - normal top (atm, that asterisk CPU spikes up to 180% at times) |
10:48.07 | wasanzy | so I can define ext => h, blalabla? |
10:50.00 | jkroon | wasanzy, that's generally the idea yes ... although, determining from which side the HANGUP came is something I've not yet managed |
10:50.18 | jkroon | Samot, will a screenshot from htop be ok? I'm unable to copy&paste the text |
10:50.20 | wasanzy | ok |
10:50.24 | Samot | So, what is mysql doing? |
10:50.36 | *** join/#asterisk Tiffon (~name@unaffiliated/tiff0n) |
10:50.38 | Samot | Since according to this it's taking up more resources than Asterisk. |
10:51.03 | jkroon | mostly i've got a process that hooks the AMI and logs RTCP reports to MySQL. |
10:51.03 | Samot | Well more memory and big chunk of CPU. |
10:51.17 | jkroon | it's using more RAM yes. and most likely more disk IO too. |
10:51.33 | jkroon | not even one core, overall 81.9% CPU idle of the 8 cores. |
10:51.36 | Samot | And almost 50% CPU. |
10:51.48 | jkroon | i'm worried about single threads taking a whole core on it's own (ie, 100%) |
10:52.34 | *** join/#asterisk kline (~kline@freenode/staff/euncs.kline) |
10:53.48 | jkroon | oh, and I've got astdb on a ramdisk ... because that's known to cause MAJOR problems. |
10:53.50 | Samot | What happens when you stop the AMI process that's writing logs? |
10:53.56 | jkroon | well, with handling of SIP. |
10:54.00 | *** join/#asterisk kline (~kline@freenode/staff/enucs.kline) |
10:54.04 | jkroon | I wrote that specifically to try and pinpoint the problem. |
10:54.13 | jkroon | so problem started before I hooked into the AMI there. |
10:54.34 | Samot | Is it constant? |
10:54.40 | Samot | Does it go away if the call volume drops? |
10:54.47 | jkroon | it's variable. |
10:54.56 | jkroon | I haven't yet graphed call volume vs problems. |
10:55.26 | jkroon | Let me quickly see if I can create a call volume histogram from the CDR records to see if that correlates with the problems. |
10:56.24 | jkroon | we have an idea of when the problems occured. |
10:56.32 | Samot | When? |
10:56.51 | jkroon | 8:15 SAST and then again around 9:55 SAST etc ... |
10:56.56 | jkroon | so we've got approximate time slots. |
11:01.31 | jkroon | this is super strange. asterisk is using less CPU now with more calls than it did a moment back. |
11:04.47 | Samot | Active calls take less resources vs setup/tear down |
11:05.26 | jkroon | fair enough. but setup also shouldn't affect threads shuffling rtp should they? |
11:05.41 | jkroon | and it seems rtp is affected in some way or another somewhow. |
11:05.48 | Samot | Well it's real time |
11:05.55 | Samot | So if the server is being lagged. |
11:06.11 | Samot | Then it's going to impact performance. |
11:06.32 | Samot | So yes, poor audio quality is a result of a taxed server. |
11:07.01 | jkroon | at 100 concurrent calls? |
11:07.20 | jkroon | we've done 130+ before ... same version of asterisk. i'm trying to figure out what changed ... but cannot pinpoint anything. |
11:07.25 | Samot | What else is being done on the server? |
11:07.43 | jkroon | asterisk + mysql ... and a handful of helper processes. |
11:08.07 | Samot | What are the users doing? |
11:08.08 | jkroon | also started tcpdump in an effort to confirm the udp flows for rtp + sip (and also capturing 5060 tcp). |
11:08.11 | Samot | Are their queues? |
11:08.21 | Samot | How many Ring Groups, IVRs? |
11:08.25 | Samot | Conference calls? |
11:08.34 | jkroon | no ring groups, no IVRs, no conference calls. |
11:08.39 | jkroon | no queues. |
11:09.00 | jkroon | this box purely switches between incoming channel and outgoing channel, and then sporadically some voicemail. |
11:09.15 | voipmonk | not a true statement. |
11:09.35 | jkroon | ok, ... truth to the best of my understanding ... why do you disagree? |
11:09.41 | voipmonk | offload mysql to another box , offload tcpdump to another box and use a mirror port on a beefy switch that can handle it then you might be able to say "purely" |
11:09.43 | jkroon | udp 768 0 0.0.0.0:29590 0.0.0.0:* 11057/asterisk |
11:10.00 | voipmonk | and tell me more about astdb |
11:10.14 | jkroon | http://jkroon.blogs.uls.co.za/it/voip/asterisk-massively-speeding-up-those-register-requests |
11:10.29 | jkroon | we've mounted a ramdisk on /var/lib/asterisk/astdb and told asterisk to put astdb in there. |
11:10.44 | jkroon | the reason for that is the fsync() that sqlite executes causes the sip thread to block. |
11:10.49 | voipmonk | its horrible that you are only doing 120+ calls on this box - |
11:11.12 | voipmonk | http://jkroon.blogs.uls.co.za/it/voip/asterisk-massively-speeding-up-those-register-requests |
11:11.14 | voipmonk | is this you? |
11:11.23 | jkroon | voipmonk, and thus why i'm here looking for help. previous estimates and calculations showed we should be doing a lot more. |
11:11.23 | jkroon | yes. |
11:11.43 | voipmonk | this shouldn't happen until you get into the 1200's at least |
11:12.01 | voipmonk | where'd you dig up this proc? |
11:12.07 | voipmonk | cat /proc/cpuinfo |
11:12.17 | jkroon | yes, from /proc/cpuinfo. |
11:12.48 | voipmonk | what switch are you using? |
11:13.22 | voipmonk | did I read that you are suggesting that calls are dropping? |
11:13.49 | voipmonk | heh - can you explain what the customer experience is? |
11:14.05 | Samot | How many endpoints on registering to the system? |
11:14.11 | voipmonk | yes |
11:14.12 | voipmonk | good one |
11:14.37 | jkroon | 1595 sip peers [Monitored: 380 online, 401 offline Unmonitored: 12 online, 802 offline] |
11:14.46 | voipmonk | yeah |
11:14.50 | jkroon | 376 iax2 peers [142 online, 16 offline, 218 unmonitored] |
11:15.36 | jkroon | switch is an HP 1910-24G, neither it's error counters nor that on the server is reporting any problems. |
11:15.37 | Samot | What is the avg Registration Expires set for the endpoints? |
11:15.47 | jkroon | 60 minutes IIRC. |
11:16.02 | jkroon | let me confirm. |
11:16.19 | jkroon | rx_no_buffer_count: 27128059 |
11:16.25 | jkroon | I lied. error counters on eth0. |
11:16.35 | Samot | So when you are doing your tcpdumps or sip debugs do you see register attempts from endpoints that are not yours? |
11:16.49 | jkroon | logs aren't reporting failed registration attempts. |
11:17.11 | Samot | I didn't ask about logs. |
11:17.18 | Samot | I asked about live data. |
11:17.32 | jkroon | didn't check tcpdump, will do a quick check. |
11:17.52 | jkroon | i have increased the rx and tx ring buffers on the ethernet interfaces now, that may (or may not) help. |
11:19.24 | Samot | And while you're at it, see how many call attempts are being made as well. |
11:19.34 | Samot | They don't need to register to try and send calls through you. |
11:19.37 | jkroon | INVITE? |
11:19.46 | Samot | Yes |
11:20.22 | Samot | Is this box directly on the Internet or on a LAN? |
11:20.53 | jkroon | server => switch => firewall => switch => 100Mbps link => open internet. |
11:21.04 | jkroon | that 100Mbps is a peering link. |
11:21.11 | Samot | So it's behind a LAN |
11:21.19 | jkroon | currently running at around 30-60Mbps variable during the day. |
11:21.20 | Samot | That doesn't tell me if it's on a LAN |
11:21.32 | jkroon | well, strictly speaking it's always on some kind of LAN. |
11:21.43 | jkroon | you're asking whether port 5060 is internet accessible and the answer is yes. |
11:22.02 | Samot | Are the IPs on the box private LAN IPs? |
11:22.09 | jkroon | no, public IPs, no NAT. |
11:22.11 | Samot | Or does it listen directly on a public IP? |
11:22.31 | Samot | So the answer is yes, the box is directly on the Internet. |
11:23.30 | jkroon | ok, yes, the box is directly on the internet, and the firewall doesn't filter a lot of traffic for this machine, all SIP is allowed in, along with the RTP ranges, and IAX/2 traffic. Logs on the firewall doesn't show any traffic being dropped by the firewall, and hardware counters seems sane too. |
11:23.45 | jkroon | error counters ... |
11:24.26 | jkroon | so nothing dropped by interface/switch itself that I can find. the switch did report 30 input packet errors overall on the switch over something like 3 billion packtes, so I think we can leave that out of the equation. |
11:28.23 | Samot | So it's on the Internet and the firewall let's pretty much everything through.. |
11:28.43 | Samot | Need to see how many people are actually banging at the door on your box. |
11:32.38 | *** join/#asterisk troyt (~troyt@2601:681:4600:7641:44dd:acff:fe85:9c8e) |
11:36.32 | Samot | You have things like allowguest=no and allow_sip_anon=no? |
11:39.47 | *** join/#asterisk sekil (~sekil@jabber.net011.net) |
11:43.39 | *** part/#asterisk AviiNL (~AviiNL@185.21.52.255) |
11:44.21 | *** join/#asterisk AviiNL (~AviiNL@185.21.52.255) |
11:49.55 | jkroon | Samot, allowguest = no |
11:51.24 | jkroon | I cannot find that SIP has a allow_sip_anon setting. |
11:51.45 | *** join/#asterisk Tiffon (~name@unaffiliated/tiff0n) |
11:55.12 | *** join/#asterisk Tiffon (~name@unaffiliated/tiff0n) |
11:58.30 | *** join/#asterisk acidfu_ (~acidfoo@modemcable002.114-70-69.static.videotron.ca) |
12:03.57 | Samot | I think that goes in extensions.conf |
12:04.20 | *** join/#asterisk SunGod (~SunGod@92.55.116.125) |
12:07.00 | jkroon | Samot, that seems to be a freepbx setting? |
12:07.01 | Samot | Well allow_sip_anon is more a global variable for dialplan actions. |
12:07.11 | jkroon | which we're not using. |
12:07.15 | Samot | It's not a SIP setting. |
12:07.35 | jkroon | looks like a dialplan global variable, specific to freepbx based on what I can find. we're not using freepbx. |
12:07.36 | Samot | It's a global variable...yes..I was looking in the wrong SSH tab on my boxes. |
12:10.17 | jkroon | no problem. |
12:10.35 | jkroon | we've been running in approximately this config for ~8 years now, and definitely no changes since Feb. |
12:10.55 | jkroon | well, other than adding more endpoints ... so I'm thinking it's load related somehow. |
12:20.14 | *** join/#asterisk sekil (~sekil@jabber.net011.net) |
12:23.38 | jkroon | i picked up that the hardware rx buffers was at the default 256. I increased those to 4096. time will tell if this makes a difference. |
12:24.14 | WIMPy | Hardware? |
12:24.26 | jkroon | ethtool -g eth0 ... those settings. |
12:24.52 | WIMPy | Ah |
12:25.20 | jkroon | ethtool -S eth0 showed rx_no_buffer_count: 27128059 |
12:25.20 | jkroon | that's 27 *million* packets that won't show up in other error counters which I was previously looking at and monitoring. |
12:25.39 | jkroon | from a switch perspective the tx is successful, and from the rx_errors the packet was successfully received but due to no space in the buffers couldn't be passed to the OS. |
12:26.41 | jkroon | mostly in past experience for most drivers/hardware implementations (I'm not actually too clued up on where/how those counters are implemented and it seems to be driver specific) no buffer accounted to rx_errors and more specifically rx_missed_errors which doesn't seem to be the case here. |
12:26.45 | WIMPy | That sounds pretty bad once again. |
12:27.02 | jkroon | which is WHY my existing monitoring didn't pick up on it. |
12:28.16 | jkroon | 27128059/56564770852 = 0.05% over 570 days, 17:08. |
12:28.43 | jkroon | so most likely the problem just became noticable now it was probably there for a longer time. |
12:30.04 | WIMPy | That's the point where I expect someone to say it wouldn't have happened on BSD... |
12:32.03 | jkroon | well, with ping to another destination on another network I'm still seeing 0.008 % loss (ping -f). not sure if it's worth chasing that down. |
12:32.56 | jkroon | and if it wouldn't (by logical reason, not just because someone said so) have happened on BSD I'd be willing to put that to a test. If I can rig one. |
12:50.50 | *** join/#asterisk [TK]D-Fender (~joe@216-191-106-165.dedicated.allstream.net) |
13:01.06 | *** join/#asterisk guest9103 (~bounceman@185.32.9.250) |
13:01.37 | guest9103 | Hi, in the manager.conf. How should I specify the permit option if I wish to allow multiple IP-addresses? I find no docs explaining this. |
13:07.50 | [TK]D-Fender | I found a ton just googling it right now |
13:07.52 | [TK]D-Fender | http://www.voip-info.org/wiki/view/Asterisk+config+manager.conf |
13:07.57 | [TK]D-Fender | shows multiple used directly |
13:08.08 | [TK]D-Fender | the sample config itself shows the ACL option |
13:08.17 | [TK]D-Fender | https://wiki.asterisk.org/wiki/display/AST/Named+ACLs |
13:08.32 | [TK]D-Fender | Which is pretty clearly documented on the official wiki. |
13:11.02 | guest9103 | What the heck is "Named ACL"? |
13:11.28 | *** join/#asterisk stac (~stac@april-fools/2013/runnerup/stac) |
13:12.57 | [TK]D-Fender | ACL. |
13:13.07 | [TK]D-Fender | a csettting right there in the sample config |
13:13.58 | [TK]D-Fender | [TK]D-Fender> https://wiki.asterisk.org/wiki/display/AST/Named+ACLs <- I think the very first paragraph answers that already |
13:19.40 | *** join/#asterisk brad_mssw (~brad@66.129.88.50) |
13:29.50 | guest9103 | Great! |
13:36.34 | *** join/#asterisk matrix1233 (~matrix123@197.0.177.30) |
13:36.37 | *** join/#asterisk O47m341 (~Suzeanne@c-68-51-46-116.hsd1.in.comcast.net) |
13:37.23 | guest9103 | [TK]D-Fender: do you happen to know if the pem file needs to be generated using the as_tls script or is it okay to generate it using openssl req? |
13:38.01 | [TK]D-Fender | I don't |
13:38.19 | guest9103 | [TK]D-Fender: You are the man of the day anyway. |
13:39.00 | guest9103 | Accept |
14:09.44 | *** join/#asterisk kharwell (kharwell@nat/digium/x-wmzjtfmjleisfxqs) |
14:18.11 | *** join/#asterisk punk_ (eirik@login.blix.com) |
14:18.17 | punk_ | hello everyone :) |
14:25.25 | *** join/#asterisk putnopvut (putnopvut@asterisk/master-of-queues/mmichelson) |
14:25.25 | *** mode/#asterisk [+o putnopvut] by ChanServ |
14:31.30 | *** join/#asterisk rmudgett (rmudgett@nat/digium/x-tsvhggcutzwqqqiu) |
14:41.20 | *** join/#asterisk newtonr (RustyNewto@nat/digium/x-zlwcjwauhdutanxz) |
14:41.20 | *** mode/#asterisk [+o newtonr] by ChanServ |
15:22.32 | *** join/#asterisk jmetro (32fe06a1@gateway/web/freenode/ip.50.254.6.161) |
15:22.42 | jmetro | Can anyone remind me the name of the default hold music? :) |
15:24.07 | [TK]D-Fender | [default] |
15:24.33 | jmetro | The Mp3 <.< i think its "Cold Morning Coffee" ? |
15:24.48 | [TK]D-Fender | IIRC * includes 4 or so |
15:26.26 | jmetro | hmm... |
15:26.46 | *** join/#asterisk Chotaire (chotaire@oahu.chotaire.net) |
15:44.54 | ghoti | So ... I have a slew of what appear to be "zombie" channels "core show channels verbose" output is here: [root@pbx ~]# asterisk -rx 'core show channels verbose' |
15:44.57 | ghoti | Channel Context Extension Prio State Application Data CallerID Duration Accountcode PeerAccount BridgedTo |
15:45.01 | ghoti | SIP/222-00000532 from-internal STARTMEETME 4 Up ConfBridge 822,,, 222 4122:55: (None) |
15:45.05 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 3024:46: (None) |
15:45.09 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 1027:33: (None) |
15:45.13 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 193:05:1 (None) |
15:45.17 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 1145:44: (None) |
15:45.22 | ghoti | Bridge/0x7f256800ba0 default s 1 Up (None) (None) 4122:54: (None) |
15:45.25 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 118:28:4 (None) |
15:45.29 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 3165:56: (None) |
15:45.33 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 3024:57: (None) |
15:45.37 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 3166:07: (None) |
15:45.42 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 1196:13: (None) |
15:45.46 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 433:42:2 (None) |
15:45.50 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 1196:41: (None) |
15:45.54 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 4122:54: (None) |
15:45.58 | ghoti | Bridge/0x7f256800ba0 default s 1 Up (None) (None) 4122:54: (None) |
15:46.02 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 3165:57: (None) |
15:46.06 | ghoti | ConfBridgeRecorder/c default s 1 Up (None) (None) 118:31:2 (None) |
15:46.10 | ghoti | 17 active channels |
15:46.13 | ghoti | 1 active call |
15:46.15 | ghoti | 16876 calls processed |
15:46.17 | ghoti | CRAP, sorry about that, meant to paste the patebin.ca link. :-( |
15:46.46 | ghoti | http://pastebin.ca/3707994, fwiw. :-P |
15:47.36 | ghoti | These appear to have been valid calls that never got removed from .. some internal table. And they're affecting my stat gathering, though I don't think they're actually affecting performance or trunk usage. |
15:48.07 | ghoti | Any idea what I can do to clear them? I imagine that rebooting the box would do the trick, but I'd prefer something less .. drastic. |
15:58.37 | *** join/#asterisk Chotaire (chotaire@oahu.chotaire.net) |
16:16.29 | *** join/#asterisk matrix1233 (~matrix123@197.0.177.30) |
16:26.36 | rwb | Hi, does anyone know why the extensions don't ring? I have analog phones with FXS extentions. If I dial an extention, it doesn't ring, but when I pick it up, it connects! |
16:28.20 | WIMPy | Wrong voltage, wrong frequency, too little power. |
16:30.48 | rwb | Hmm, they are old analog phones, but there should be enough power from the card (the supply is in fine) Not sure how to test that. |
16:31.49 | Samot | Is the ringer turned on? |
16:32.44 | Samot | Or turned up? |
16:33.08 | Samot | I don't know how many times that was the cause of "service not working right" when I was doing residential services. |
16:33.19 | *** join/#asterisk miralin (~Thunderbi@194.8.128.48) |
16:35.10 | *** join/#asterisk lgaetz (~lgaetz@66.185.28.100) |
16:35.21 | rwb | yea, are, I was hoping for maybe some setting somewhere as I am quite new to this. I'll have to figure out a way to test the phones themselves. |
16:36.31 | Samot | Do you have an ATA or even a POTS line? |
16:38.10 | rwb | I don't, so far no way to test. I will be getting some "ring detectors" shipped today though, so I'll try those directly. |
16:39.21 | rwb | I'm just really hoping is not the pci board or modules! |
16:40.30 | Samot | What does 'dahdi show cadences' have? |
16:40.49 | Samot | or 'dahdi show status' |
16:43.27 | *** part/#asterisk lgaetz (~lgaetz@66.185.28.100) |
16:49.04 | rwb | I see numbers for r1: - r4 but not sure what those mean. |
16:50.40 | *** join/#asterisk boris_rh (d0414927@gateway/web/freenode/ip.208.65.73.39) |
16:53.42 | boris_rh | Hello, I was looking to use ARI snoop feature: https://wiki.asterisk.org/wiki/display/AST/Asterisk+13+Channels+REST+API#Asterisk13ChannelsRESTAPI-snoopChannel |
16:54.13 | boris_rh | but then I came across this limitation: http://lists.digium.com/pipermail/asterisk-dev/2015-November/075157.html |
16:54.46 | boris_rh | does not it make whole whisper feature useless or I am missing something? |
16:56.07 | file | useless only if you have no media going to the channel |
16:58.45 | boris_rh | thanks @file can you please elaborate ? |
16:59.03 | boris_rh | if channel is in conference |
16:59.12 | file | if the channel has no media being sent to it, then the whisper can't mix itself in |
16:59.27 | file | a conference generates media always, so it would work |
16:59.41 | file | the same limitation applies to ChanSpy |
17:00.11 | boris_rh | so why guy at http://lists.digium.com/pipermail/asterisk-dev/2015-November/075157.html not happy? |
17:00.15 | boris_rh | is not it his case? |
17:00.32 | boris_rh | when he has conf bridge |
17:00.42 | boris_rh | and another bridge for spying |
17:00.48 | file | in his case the channel was NOT in a bridge |
17:01.03 | file | and ARI does not generate a constant media stream to the channel, unless you call /silence |
17:01.42 | boris_rh | Thanks @file, so if I have some bridge with two channels |
17:01.48 | file | then it's fine |
17:01.57 | boris_rh | I invoke snoop on one of the channel |
17:02.12 | boris_rh | and add snoop channel to another bridge |
17:02.27 | boris_rh | where I also adding whisperring person |
17:02.37 | boris_rh | is it right way to do? |
17:02.40 | file | yes |
17:03.13 | boris_rh | great, but it sounds to me the same case as at http://lists.digium.com/pipermail/asterisk-dev/2015-November/075157.html |
17:03.23 | boris_rh | maybe I missing something |
17:03.36 | file | in that case the channel being snooped on was doing nothing |
17:03.40 | file | it was waiting for ARI operations |
17:03.47 | file | when a channel is waiting in ARI *NO AUDIO* is sent to it |
17:03.56 | file | since there's no audio the whispering can't mix the whispered audio in |
17:04.19 | file | if the channel is in a bridge with another channel then media IS being sent to it so the whispering will work |
17:04.59 | boris_rh | OK, cool, thanks a lot @file, it is really encouraging , I will give it a try |
17:06.14 | boris_rh | can I create more then one snooping channel for the same channel? |
17:06.23 | file | yes. |
17:07.44 | boris_rh | @file: great, thanks |
17:09.22 | *** join/#asterisk babak (uid19622@gateway/web/irccloud.com/x-bxfbilmunrrhfjob) |
17:10.35 | *** join/#asterisk Euler (~Euler@pool-173-48-218-40.bstnma.fios.verizon.net) |
17:10.41 | Euler | hi all |
17:11.22 | Euler | upgrading to v13 from like v1.4, and asterisk doesnt seem to handle ENV vars same way; |
17:11.47 | Euler | that is, the ENV my AGI script has access to |
17:12.29 | [TK]D-Fender | That's nearly a decade-long jump |
17:12.47 | Euler | tried passing in via AGI() command (e.g., AGI(/path/to/script, ${ENV(VARNAME)}), but it comes through as empty string |
17:12.52 | [TK]D-Fender | Show the actual failure & code BTW |
17:14.56 | Euler | basically, i have a linux ENV var that i want to pass in and it doesnt seem to be available |
17:15.55 | [TK]D-Fender | Right from the start that'd have nothing to do with AGI, and rather that dialplan function. |
17:16.35 | Euler | i think the big difference is that it is running now as "asterisk" user, and i'm not sure that user has the same env as other users |
17:17.08 | Euler | yea, i don't think it's related to AGI() specifically; that's just where i noticed it |
17:17.53 | Euler | according to extensions.conf doc, "Unix/Linux environmental variables can be reached with the ENV dialplan function: ${ENV(VARIABLE)}" |
17:18.06 | Euler | but that seems to be empty for my env var |
17:18.20 | Euler | so i'm guessing the asterisk user doesnt have it for some reason |
17:19.27 | WIMPy | Do you start Asterisk the same way you did before? |
17:20.12 | Euler | it's a newer OS, so it uses "service" calls instead of older init script |
17:27.01 | Euler | well, looks like i can set specific env vars in the systemd script |
17:27.28 | Euler | so, i'm past that hump |
17:27.37 | Euler | thanks for listening! |
17:28.53 | Euler | (ffr, adding the var to /usr/lib/systemd/system/asterisk.service , in format like: Environment=MYVAR=someval ) |
17:31.23 | punk_ | is there any way I can manipulate an incomming call from the console? so that it gets redirected to another extension? |
17:32.05 | punk_ | I guess I could create hooks/checks in the extensions.conf file which looked at some file or database to see if the call should go somewhere else, but is there a better way? :-) |
17:34.25 | [TK]D-Fender | punk_, depends when and how you actually want these things to happen |
17:34.36 | [TK]D-Fender | punk_, because you just described 2 completely different things |
17:34.44 | [TK]D-Fender | punk_, So clarify your actual goal |
17:38.00 | punk_ | [TK]D-Fender: Customer calls the queue (choose 1 for sales... etc), at the same time I have an IRC bot that tells me who is calling/number, I want to be able to type !mine or !transfer <phonenum> and it goes directly to the correct person :-) |
17:38.45 | [TK]D-Fender | That would be an AMI Redirect you're looking for. |
17:39.24 | [TK]D-Fender | And nothing to do with a database lookup, etc. You are deciding live yourself external to the PBX. |
17:39.39 | punk_ | Thank you for the push in the right direction, I will investigate the AMI :) |
17:42.18 | *** join/#asterisk infinity_ (~brendon@web2.artsopolis.com) |
18:01.38 | punk_ | [TK]D-Fender: Quick question, probably a noob one, which channel should I use for the API when its a call coming over a SIP Trunk? |
18:02.22 | [TK]D-Fender | The channel you want to hijack clearly |
18:02.35 | punk_ | [TK]D-Fender: But I only know the call id? |
18:03.19 | *** join/#asterisk matrix1233 (~matrix123@197.0.177.30) |
18:03.31 | [TK]D-Fender | Is that ... a question? |
18:03.44 | punk_ | is that ... the same? the channel and the call id? |
18:04.07 | [TK]D-Fender | No. |
18:04.19 | [TK]D-Fender | call-id = SIP thing |
18:04.43 | [TK]D-Fender | channel = asterisk thing that encapsulates the call. * channel can be of ANY type |
18:05.44 | punk_ | I just noticed there is something called "channel redirect" in the asterisk console, I guess I could use that instead of the API? |
18:05.58 | punk_ | since my bot can do asterisk -rx |
18:06.10 | [TK]D-Fender | that could work |
18:08.55 | punk_ | [TK]D-Fender: I figured it out, great, thank you so much for your help and the push in the right direction. Have a nice day! :) |
18:09.20 | [TK]D-Fender | You're welcome |
18:12.24 | rwb | Anyone know if 24G phone wire thick enough to ring older dial phones with bells? |
18:14.54 | Nugget | takes a moment to say thanks for the fact that he'll never have to know that. |
18:25.32 | gruetzkopf | over which distance? |
18:26.03 | gruetzkopf | 24 AWG is 0.5mm diameter, that should be plenty. |
18:26.24 | gruetzkopf | source: run stuff like that (although not on a FX* pci card) |
18:37.26 | rwb | maybe 200 ft max each chan. I have a patch panel connected with a short telco cable to a 24 line FXS pci card. I'm thinking of driving some old dial phones with the 24g. I think it should be fine as well. |
18:38.36 | rwb | No "big" bells or lights without a seperate relay... |
18:46.40 | gruetzkopf | i run pretty big bells over like 1000ft of .6mm wire, you should be fine |
18:50.02 | rwb | good to know. Thanks! |
18:57.02 | *** join/#asterisk smkelly (~smkelly@mykonos.smkelly.org) |
19:01.24 | *** join/#asterisk kline (~kline@freenode/staff/enucs.kline) |
19:03.10 | *** join/#asterisk drale2k (~drale2k@46.101.230.32) |
19:04.41 | *** join/#asterisk davlefou (~davlefou@unaffiliated/davlefou) |
19:10.14 | *** join/#asterisk tzafrir (~tzafrir@bzq-179-40-173.cust.bezeqint.net) |
19:11.26 | *** join/#asterisk defsdoor (~andy@cpc8-sutt4-2-0-cust254.perr.cable.virginm.net) |
19:22.27 | *** join/#asterisk jkroon (~jkroon@105.6.32.6) |
19:47.04 | rwb | Is there a good way to trouble shoot why some FXS ports are dead? I'm pretty new on this. In my "Full Report" screen I see "State:Unavailable" for the ones that seem dead (no dial tone)? |
20:00.01 | rwb | Or another way to ask; Does it mean I have a bad board and/or FXS modules if I don't hear dial tone from ALL of the ports? |
20:00.52 | [TK]D-Fender | do you have a proper indications.conf? Set your zones right? |
20:00.53 | rwb | Or, could this be configuration error (me) |
20:01.04 | [TK]D-Fender | Tested with other phoens to validate the current draw as being sane? |
20:01.50 | rwb | So far I'm using the web admin only. Yes, I use the same phone on all the ports, and about half of them are silent |
20:02.08 | rwb | same phone, and same patch cable. |
20:02.40 | *** join/#asterisk jeffspeff (~Jeff@12.49.160.131) |
20:03.02 | rwb | I see "State:Unavailable" in the reports area for the ones that seem dead |
20:03.37 | rwb | I'm wondering if I just need to "enable" them somewhere |
20:03.45 | [TK]D-Fender | have you tried a DIFFERNT PHONE? |
20:03.54 | [TK]D-Fender | and a DIFFERENT CABLE |
20:04.13 | rwb | yes, same results with 3 different phones, 3 cables, on all ports |
20:04.38 | rwb | all old really old phones though... |
20:04.46 | [TK]D-Fender | The back to the first 2 thing I told you to look for. |
20:05.21 | rwb | ahh, I lost track. about when was that? |
20:06.05 | rwb | oh, sones and indications... I will look at those |
20:06.46 | jeffspeff | i've done a fresh install of * on a google compute instance. i'm able to run * using 'asterisk -cv' but i can't get the service to act right. here's a paste of the error and some other details http://pastebin.com/UTLmsh3a any ideas? thanks |
20:21.02 | *** join/#asterisk kubrrick (~Mutter@78.80.9.109.rev.sfr.net) |
20:24.29 | kubrrick | Hello guys, i m looking for srtp on my asterisk server, can someone tell me how to set cipher suite for srtp ?? |
20:57.52 | *** join/#asterisk Qwell (~north@asterisk/developer/Qwell) |
20:57.52 | *** mode/#asterisk [+o Qwell] by ChanServ |
21:08.19 | *** join/#asterisk Pegasus_RPG (~Icedove@47.142.203.153) |
21:09.20 | Pegasus_RPG | Hello. I'm trying to make a generic "pause queue member" application using exten => *30*,1,PauseQueueMember(,${SIPUSERAGENT}) on * 11 but it keeps saying the interface parameter is missing |
21:09.45 | Pegasus_RPG | If I manually specify e.g. SIP/User1 that works |
21:10.09 | Pegasus_RPG | So I guess my question is: how can I strip off the transaction ID from the User agent string? |
21:10.38 | voipmonk | CUT it out |
21:11.14 | Pegasus_RPG | tries PauseQueueMember(,SubString(${SIPUSERAGENT},0,9)) |
21:11.42 | Pegasus_RPG | XD Attempt to pause interface SubString(,0,9), not found |
21:12.03 | Pegasus_RPG | reads about cut |
21:17.25 | Pegasus_RPG | tries a few things with it |
21:20.55 | Pegasus_RPG | Right now I have exten => *30*,1,Set(Member=${CUT(SIPUSERAGENT,,1)}) but it's not splitting on the - (SIPUSERAGENT contains SIP/User1-000003b ) |
21:21.26 | Pegasus_RPG | And I want just the SIP/User1 part |
21:21.30 | Pegasus_RPG | voipmonk: ^ |
21:22.14 | voipmonk | that doesn't look right - |
21:22.43 | voipmonk | hold on - bee's like my vape - and this form doesn't do well with stings - moving |
21:23.29 | *** join/#asterisk clopez (~tau@neutrino.es) |
21:23.43 | Pegasus_RPG | wait I think i found a trick with ${foo:0:-8} |
21:25.05 | Pegasus_RPG | DAMMIT exten => *30*,1,Set(Member=${SIPUSERAGENT:0:-9}) isn't working either |
21:25.15 | Pegasus_RPG | (Shows "Member=" ) |
21:28.55 | Pegasus_RPG | AUGH ${SIPUSERAGENT} * - SIP user agent (deprecated) |
21:29.01 | Pegasus_RPG | So much old info on voip-wiki |
21:31.34 | *** join/#asterisk [TK]D-Fender (~joe@64.235.216.2) |
21:43.11 | *** join/#asterisk jkroon (~jkroon@2c0f:f720:1:0:fcb3:e235:8b0a:1f6b) |
21:47.54 | Pegasus_RPG | got it! ${CHANNEL:0:-9} |
21:48.16 | Pegasus_RPG | Thanks to the real wiki https://wiki.asterisk.org/wiki/display/AST/Asterisk+Standard+Channel+Variables |
21:48.31 | Pegasus_RPG | voipmonk: ^ |
22:32.24 | kunwon1 | Pegasus_RPG: i'm not sure if the length of a channel name is a constant, that may have weird corner cases |
22:33.37 | Pegasus_RPG | kunwon1: I'm sure. I'm open to better ideas. :) |
22:38.30 | kunwon1 | Pegasus_RPG: is ${BLINDTRANSFER} populated? |
22:43.38 | kunwon1 | looks likke not |
22:47.49 | Pegasus_RPG | The question is: is the 8-digit transaction ID (or whatever it's called) always 8 digits? |
22:48.24 | Pegasus_RPG | (the -9 strips off the last 9 characters so should work fine for arbitrary length strings so long as that ID is constant length) |
22:48.27 | WIMPy | The only thing we know about channel names is that they start with a channeltype name and a slash. |
22:48.53 | kunwon1 | Pegasus_RPG: i figured it out, here you go same => n,Set(cutVar=${CUT(CHANNEL,,-3)}) |
22:49.10 | kunwon1 | that puts the device name in ${cutVar} |
22:50.00 | kunwon1 | maybe this would have corner cases too? unsure, i'm fairly certain it would work with my pure-SIP environment |
22:50.38 | *** join/#asterisk acidfu_ (~acidfoo@24-212-247-227.cable.teksavvy.com) |
22:50.39 | WIMPy | For the "corner case" of using chan_sip it will work. |
22:50.41 | kunwon1 | i wonder if pausequeuemember needs SIP/ in front? the line i just posted includes that, but adding another CUT should be able to remove it |
22:50.47 | Pegasus_RPG | WIMPy: I guess * needs an INTERFACE variable then :) |
22:51.01 | Pegasus_RPG | kunwon1: it does |
22:51.05 | WIMPy | What would that be? |
22:51.18 | kunwon1 | good times |
22:51.36 | kunwon1 | WIMPy: is chan_sip in the minority these days? |
22:51.37 | Pegasus_RPG | WIMPy: e.g. SIP/User1 |
22:51.49 | Pegasus_RPG | WIMPy: ZAP/foo |
22:51.51 | Pegasus_RPG | etc |
22:52.03 | WIMPy | Nope. Don't think so. But still there are many other channeltypes/ |
22:52.07 | kunwon1 | true |
22:52.08 | WIMPy | . |
22:52.45 | Pegasus_RPG | kunwon1: how does your CUT line differ from my substring one? |
22:52.58 | kunwon1 | my CUT line doesn't care how long the part after the final hyphen is |
22:53.03 | Pegasus_RPG | ah okay |
22:53.24 | kunwon1 | i believe the part after the final hyphen is of variable length, but i have no supporting evidence |
22:54.04 | Pegasus_RPG | Doesn't work for me though |
22:54.25 | kunwon1 | :/ |
22:54.37 | WIMPy | The only thing we know about channel names is that they start with a channeltype name and a slash. -- That means everything else is unknown! |
22:54.38 | Pegasus_RPG | sec, trying stuffing it into an intemediate varable |
22:55.03 | WIMPy | That also means that thee's no generic way to identify devices. |
22:55.36 | Pegasus_RPG | kunwon1: yep, I needed to put the cut result into a variable, then call that from PauseQueueMember() |
22:56.15 | kunwon1 | ah, good times |
22:57.25 | Pegasus_RPG | oh wait, no it still doesn't work :/ |
22:57.44 | kunwon1 | try changing -2 to -1, i have more hyphens in my device names than normal, i think |
22:58.32 | WIMPy | It might be easier to just use the CANNEL function. |
22:58.34 | Pegasus_RPG | Mine are like SIP/User1-00000059 |
22:58.53 | WIMPy | Not generic, either, though. |
22:59.21 | *** part/#asterisk kharwell (kharwell@nat/digium/x-wmzjtfmjleisfxqs) |
23:01.28 | kunwon1 | hmm, ${CHANNEL(peername)} |
23:01.30 | kunwon1 | TIL |
23:10.48 | *** join/#asterisk newtonr (~newtonr@173-21-146-94.client.mchsi.com) |
23:10.48 | *** mode/#asterisk [+o newtonr] by ChanServ |
23:12.18 | *** join/#asterisk pa (~pa@unaffiliated/pa) |
23:13.36 | *** join/#asterisk Chotaire (chotaire@oahu.chotaire.net) |
23:51.54 | *** join/#asterisk acidfu_ (~acidfoo@24-212-247-227.cable.teksavvy.com) |