sr-users October 2018

sr-users@lists.kamailio.org

81 participants
122 discussions

Kamailio Slack channel
by Alex Balashov 25 Oct '18

25 Oct '18

Hi, I'd like to submit for the community's consideration the idea of having a Kamailio Slack channel. Emmanuel Schmidbauer took the initiative to create one: https://kamailio.slack.com/ The sign-in link is: https://join.slack.com/t/kamailio/shared_invite/enQtNDY1NDc5MzMwNDIzLWJlNTF… I for one think this is a good idea and, even if it doesn't ultimately get traction, it should be seriously considered by all. Yes, I am aware of the open-source-purist objection to Slack; it is proprietary and perceived to be a "walled garden" of sorts. Furthermore, for those of us who have been on the Internet and involved in technology for a significant amount of time, the irony is not lost that it is little more than a packaging of the "IRC" experience in the bloated form of a modern JavaScript application. I am also very reluctant to contribute in any way to the Balkanisation or fracturing of project-related communication channels. That thought gives me no comfort. These are all concerns of which I'm mindful. Nevertheless: - A great deal of professional real-time messaging and communication in many organisations has moved to Slack. This is also true of our company, and most companies we work with. Thus, a Slack channel for Kamailio would be highly complementary to a platform many other people are already using. - Slack provides a polished and cohesive user experience, which is why it has become so popular. There are clients for every platform, and a seamless user experience that is also highly compatible with mobile and tablet. One can shoehorn IRC into these media, but it clearly is not designed for them. - There are already lots of open-source discussion channels and developer forums on Slack. - Slack's inline Markdown and other conveniences make it _much_ easier to have discussions about code, configurations, etc., since it is easy to insert inline monospace and multiline blobs, attachments, etc. This is what inspired my switch to it. - A lot of people positioned within the current wave of thinking about IT and technology, and thus a big part of our candidate user base, have never used IRC. Anyway, for those interested, I propose that you join Emmanuel and I. Let's see how it works out? -- Alex -- Alex Balashov | Principal | Evariste Systems LLC Tel: +1-706-510-6800 / +1-800-250-5920 (toll-free) Web: http://www.evaristesys.com/, http://www.csrpswitch.com/

3 4

Why set this flag on registration?
by Isaac 25 Oct '18

25 Oct '18

Hello, In the example for a dual-stack IPv6/IPv4 Kamailio setup here: https://kb.asipto.com/kamailio:kamailio-mixed-ipv4-ipv6 the "FLB_IPV6" flag is being sent on IPv6 registrations. Why set this flag on IPv6 registrations? If I understand correctly, registration requests never make it beyond !save("location so it's not like the flag can be used further in the dial-plan. The code in question: route[REGISTRAR] { if (is_method("REGISTER")) { if(isflagset(FLT_NATS)) { setbflag(FLB_NATB); # uncomment next line to do SIP NAT pinging ## setbflag(FLB_NATSIPPING); } if(af==INET6) setbflag(FLB_IPV6); if (!save("location")) sl_reply_error(); exit; } } Any help is appreciated. Thanks

2 2

Kamailio Question
by Brian Longwell 25 Oct '18

25 Oct '18

I'm trying to setup a simple proxy with NAT/RTPproxy. I have started with the basic cfg file and added nat and forced rtpproxy. The proxy is to reside in between endpoints and multiple SIP servers. Currently calls from endpoints are working great but my issue is calls from the server to the endpoint. The SIP messages are good until the 200 OK sent from the phone back, Kamailio is not fixing the contact so it's sending the private IP of the phone back to the server so all subsequent traffic is attempting to route to the private IP. I have tried using the fix_nated_contact everywhere I could think but still no difference. Attached is sip capture and the kamailio.cfg i'm currently using. Thank you, Brian

1 0

RtpEngine and iLBC support
by Lars Olsson 25 Oct '18

25 Oct '18

Hi, Can anyone give a quick answer regarding RtpEngine and support for iLBC codec? I have followed the instructions how to build and enable codecs according to the GitHub manual. But iLBC codec is missing when performing "rtpengine --codecs". I have tried to google around for iLBC references, but the only thing that I have found so far is libilbc, which is a lib for ffmpeg, and it looks like ffmpeg needs to be recompiled. What is the correct/fastest way forward for getting iLBC support on RtpEngine on debian 9? The purpose it to test a-law to iLBC transcoding. Best Regards, Lars

2 1

Odd behavior comparing strings
by Duarte Rocha 25 Oct '18

25 Oct '18

Greetings, I found an odd behaviour in Kamailio and i would like to know if it's intentional. I've made this snippet for showing : $var(teste) = $null; if($var(teste) == "Test") { xerr("TEST : If returned true"); } The "if" returns true if the $var is $null or 0. I'm using the "cmp_str()" now in order to avoid this issue. Best Regards, Duarte Rocha

2 3

Re: [SR-Users] kamailio + asterisk + webrtc
by arish haque 25 Oct '18

25 Oct '18

Hi, I am using this command for starting RTPengine - /usr/sbin/rtpengine --table=0 --interface=192.168.1.x\!182.75.xx.yy --listen-ng=127.0.0.1:22222 --pidfile=/var/run/rtpengine.pid --log-level=6 --log-facility=local1 in kamailio.cfg --> listen=MY_SIP_ADDR advertise 182.75.xx.yy:5060 But still there is no RTP from outside network. Still not able to sort out what is the issue.If pfsense is blocking rtp packets or any other. but i have very config on pfsense. Thanks & Regards, Arish Haque

2 1

Re: [SR-Users] : hight shared memory usage
by Laurent Schweizer 25 Oct '18

25 Oct '18

Hello, No, both server have the same script, the only difference is that they are connected to different DB. We increase the max TLS connections because we plan to move some customer on this proxy and we want to be ready to accept more than 2K TLS connection . Laurent -----Message d'origine----- De : Henning Westerholt <hw(a)kamailio.org> Envoyé : mercredi, 24 octobre 2018 22:46 À : sr-users(a)lists.kamailio.org Cc : Laurent Schweizer <laurent.schweizer(a)peoplefone.com> Objet : [JUNK-SPAM] Re: [SR-Users] : hight shared memory usage Am Mittwoch, 24. Oktober 2018, 15:24:05 CEST schrieb Laurent Schweizer: > I have 2 proxy running and we have a very big difference of memory > usage between them, moreover we have more connected users on the proxy > with less memory usage. > > We have this issue on the first proxy, today we have to increase the > shared memory as it was no more possible to setup new TCP connection. > > We use the same version of the proxy, script is quite the same, one > difference, on the proxy with the issue we increase the number of max > TLS connection to 20K > > Any idea ? > is TLS consuming more memory than TCP or UDP ? Hello Laurent, TLS is using more memory than TCP and than UDP, due to the higher workload and overhead for crypto/connection tracking. But looking to your registered user count this should not make a big difference in the end. Do you have maybe a different usage pattern on the first proxy comparing to the second one? So that different parts of the cfg or the code are executed? Why do you increased the number of max TLS connections on the first proxy, you had also some issues here? > First proxy (with high memory usage): > > > [cycle #: 3; if constant make sure server lives] Kamailio Runtime > Details: > kamailio 5.0.6 (x86_64/linux) > now: Wed Oct 24 15:18:55 2018 > up_since: Wed Oct 24 13:21:12 2018 > uptime: 7063 > > Transaction Statistics: > tmx:UAS_transactions = 56240 tmx:active_transactions = 4 > tmx:UAC_transactions = 0 tmx:inuse_transactions = 18 > > Stateless Server Statistics: > sl:sent_replies = 134317 sl:sent_err_replies = 0 > > UsrLoc Statistics: > usrloc:location_ch-contacts = 1463 usrloc:location_ch-users = 1181 > usrloc:location_ch-expires = 1076 usrloc:registered_users = 1181 > > Core Statistics: > core:rcv_requests = 214220 core:rcv_replies = 179145 > core:fwd_requests = 5495 core:fwd_replies = 73 > > Shared Memory Statistics: > shmem:fragments = 529 shmem:real_used_size = > 15862690496 shmem:free_size = 36566109504 shmem:total_size = > 52428800000 shmem:max_used_size = 16198927224 shmem:used_size = > 15854961104 > > > > cat /proc/net/netstat | cut -d" " -f12-17 ; cat /proc/net/sockstat TW > TWRecycled TWKilled PAWSPassive PAWSActive PAWSEstab > 339118 0 684 3195 0 4648 > InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts > InECT0Pkts > 14620680 0 0 19142166496 9683 133135 > sockets: used 1951 > TCP: inuse 1029 orphan 0 tw 2 alloc 1036 mem 15 > UDP: inuse 540 mem 446 > UDPLITE: inuse 0 > RAW: inuse 0 > FRAG: inuse 0 memory 0 > > > > ------------------------------------------- > > > Second proxy (memory is OK) : > > [cycle #: 2; if constant make sure server lives] Kamailio Runtime > Details: > kamailio 5.0.6 (x86_64/linux) > now: Wed Oct 24 15:19:39 2018 > up_since: Mon Jul 2 00:31:48 2018 > uptime: 9902871 > > Transaction Statistics: > tmx:UAS_transactions = 86712048 tmx:active_transactions = 92 > tmx:UAC_transactions = 0 tmx:inuse_transactions = 178 > > Stateless Server Statistics: > sl:sent_replies = 641635175 sl:sent_err_replies = 0 > > UsrLoc Statistics: > usrloc:location-contacts = 3971 usrloc:location-users = 3845 > usrloc:location-expires = 286685 usrloc:registered_users = 3845 > > Core Statistics: > core:rcv_requests = 752644955 core:rcv_replies = 1181197213 > core:fwd_requests = 14482059 core:fwd_replies = 206909 > > Shared Memory Statistics: > shmem:fragments = 28437 shmem:real_used_size = 56447456 > shmem:free_size = 2091036192 shmem:total_size = 2147483648 > shmem:max_used_size = 82099768 shmem:used_size = 43354160 > > > cat /proc/net/netstat | cut -d" " -f12-17 ; cat /proc/net/sockstat TW > TWRecycled TWKilled PAWSPassive PAWSActive PAWSEstab > 273625 0 0 0 0 3524 > InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts > InECT0Pkts > 2010 0 1 179031265416 81564 21432766 > sockets: used 4964 > TCP: inuse 576 orphan 0 tw 3 alloc 681 mem 105 > UDP: inuse 3908 mem 3117 > UDPLITE: inuse 0 > RAW: inuse 0 > FRAG: inuse 0 memory 0 Best regards, Henning -- Henning Westerholt - https://skalatan.de/blog/ Kamailio security assessment - https://skalatan.de/de/assessment

2 1

DMQ and DIALOG
by Phillman25 Kyriacou 24 Oct '18

24 Oct '18

Hi List! I currently have 3 Kamailio's (Version 5.1) in an ACTIVE-ACTIVE-ACTIVE scenario behind a load balancer all 3 nodes are listening and advertising the same virtual ip of the load balancer. Im using dmq module together with dialogue module to sync the state of the dialog to all 3 nodes. Is there a way or module that i can use to also sync every transaction? Problem that im facing right now is that an INVITE will come to Kamailio1 and the 100 TRYING or 183 RINGING or 200OK will go to Kamailio2 and since Kamailio1 hasn't received a response will retransmit the INVITE. Or is there a way with dmq to also update the other nodes for responses like 100 ,183 or 200 OK so that Kamailio1 knows not to retransmit? Maybe im going down the wrong path? :) Any assistance would be greatly appreciated My config for dmq and dialog: #=====================DIALOG==================================== modparam("dialog", "enable_dmq", 1) #=====================DMQ======================================= modparam("dmq", "server_address", "sip:x.x.x.x:5080") modparam("dmq", "notification_address", "sip:y.y.y.y:5080") modparam("dmq", "multi_notify", 1) modparam("dmq", "num_workers", 4) modparam("dmq", "ping_interval", 30) request_route { # MANAGE ALL DIALOGS dlg_manage(); if(is_method("KDMQ")) { dmq_handle_message(); } Thanks! Phillip

4 5

: hight shared memory usage
by Laurent Schweizer 24 Oct '18

24 Oct '18

Dear all, I have 2 proxy running and we have a very big difference of memory usage between them, moreover we have more connected users on the proxy with less memory usage. We have this issue on the first proxy, today we have to increase the shared memory as it was no more possible to setup new TCP connection. We use the same version of the proxy, script is quite the same, one difference, on the proxy with the issue we increase the number of max TLS connection to 20K Any idea ? is TLS consuming more memory than TCP or UDP ? BR Laurent First proxy (with high memory usage): [cycle #: 3; if constant make sure server lives] Kamailio Runtime Details: kamailio 5.0.6 (x86_64/linux) now: Wed Oct 24 15:18:55 2018 up_since: Wed Oct 24 13:21:12 2018 uptime: 7063 Transaction Statistics: tmx:UAS_transactions = 56240 tmx:active_transactions = 4 tmx:UAC_transactions = 0 tmx:inuse_transactions = 18 Stateless Server Statistics: sl:sent_replies = 134317 sl:sent_err_replies = 0 UsrLoc Statistics: usrloc:location_ch-contacts = 1463 usrloc:location_ch-users = 1181 usrloc:location_ch-expires = 1076 usrloc:registered_users = 1181 Core Statistics: core:rcv_requests = 214220 core:rcv_replies = 179145 core:fwd_requests = 5495 core:fwd_replies = 73 Shared Memory Statistics: shmem:fragments = 529 shmem:real_used_size = 15862690496 shmem:free_size = 36566109504 shmem:total_size = 52428800000 shmem:max_used_size = 16198927224 shmem:used_size = 15854961104 cat /proc/net/netstat | cut -d" " -f12-17 ; cat /proc/net/sockstat TW TWRecycled TWKilled PAWSPassive PAWSActive PAWSEstab 339118 0 684 3195 0 4648 InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts InECT0Pkts 14620680 0 0 19142166496 9683 133135 sockets: used 1951 TCP: inuse 1029 orphan 0 tw 2 alloc 1036 mem 15 UDP: inuse 540 mem 446 UDPLITE: inuse 0 RAW: inuse 0 FRAG: inuse 0 memory 0 ------------------------------------------- Second proxy (memory is OK) : [cycle #: 2; if constant make sure server lives] Kamailio Runtime Details: kamailio 5.0.6 (x86_64/linux) now: Wed Oct 24 15:19:39 2018 up_since: Mon Jul 2 00:31:48 2018 uptime: 9902871 Transaction Statistics: tmx:UAS_transactions = 86712048 tmx:active_transactions = 92 tmx:UAC_transactions = 0 tmx:inuse_transactions = 178 Stateless Server Statistics: sl:sent_replies = 641635175 sl:sent_err_replies = 0 UsrLoc Statistics: usrloc:location-contacts = 3971 usrloc:location-users = 3845 usrloc:location-expires = 286685 usrloc:registered_users = 3845 Core Statistics: core:rcv_requests = 752644955 core:rcv_replies = 1181197213 core:fwd_requests = 14482059 core:fwd_replies = 206909 Shared Memory Statistics: shmem:fragments = 28437 shmem:real_used_size = 56447456 shmem:free_size = 2091036192 shmem:total_size = 2147483648 shmem:max_used_size = 82099768 shmem:used_size = 43354160 cat /proc/net/netstat | cut -d" " -f12-17 ; cat /proc/net/sockstat TW TWRecycled TWKilled PAWSPassive PAWSActive PAWSEstab 273625 0 0 0 0 3524 InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts InECT0Pkts 2010 0 1 179031265416 81564 21432766 sockets: used 4964 TCP: inuse 576 orphan 0 tw 3 alloc 681 mem 105 UDP: inuse 3908 mem 3117 UDPLITE: inuse 0 RAW: inuse 0 FRAG: inuse 0 memory 0

2 2

How is ruid calculated
by Kjeld Flarup 24 Oct '18

24 Oct '18

I got two tablets registering with the same username, this used to work but now I got a bug report that only one of them gets a call. I found out that the problem is in the ruid column as they both end up with this ruid value uloc-1-5bd0abe9-83c-e4 How is that ruid value calculated, and how can I change this? -- -------------------- Med Liberalistiske Hilsner ---------------------- Civilingeniør, Kjeld Flarup - Mit sind er mere åbent end min tegnebog Sofienlundvej 6B, 7560 Hjerm, Tlf: 40 29 41 49 Den ikke akademiske hjemmeside for liberalismen - www.liberalismen.dk

1 0

← Newer
1
2
3
4
5
6
7
...
13
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

sr-users October 2018