sr-users

sr-users@lists.kamailio.org

36 participants
37509 discussions

User state
by Magnus Löfqvist 19 Feb '18

19 Feb '18

Hi, We are running for a customer a kamailio server for user registration and routing of calls. Also using Asterisk for queue management and IVR. Now we are in the progress of writing a web portal where the customers users should be able to see if there collueges are busy in phone or availible. In asterisk I would use hints and listing in events in AMI. But what way are the best in Kamailio? Best Regards Magnus Sent from Nine<http://www.9folders.com/>

2 1

PR1447: first attempt at app_python3
by Anthony Alba 18 Feb '18

18 Feb '18

Hi Daniel and kamailio contributors, I have submitted a PR for app_python3 here: https://github.com/kamailio/kamailio/pull/1447. It is very rough, but would appreciate your feedback. Notes: 1. Almost a verbatim port of app_python; just changing obvious semantics. String->Unicode is practically the only large change. No change to symbol names and export functions so the two modules cannot coexist. (Don't loadmodule "app_python.so" and loadmodule "app_python3.so) 2. Module initialization semantics in Py3 have changed: add the top level modules to the inittab before calling Py_Initialize(). I have done this for KSR and Router. I.e. PyImport_AppendInittab() is called for Router and KSR *before* Py_Initialize() 3. Makefile looks for a python3 binary (PYTHON3 in the Makefile) to get include paths and link libraries. Here is an example compile line for IUS Python 3.6 on CentOS 7 (no native Python3 from RHEL) ## default: needs to find python3 and python3-config ## needs to find both python3.6 and python3.6-config make -C src/ PYTHON3=python3.6 SKIP_MODULES="${SKIP_MODULES}" quiet=verbose module_group_py="app_python3" group_include=py every-module 4. The semantics for GIL and thread management are too different for one-to-one replacements. I have therefore decided to comment out all threading calls. Fortunately, kamailio does not used threads at the moment. This is more complicated than it seems: even with Py2+thread management: there would be deadlocks if kamailio->python2 script->Cython extension with threads(e.g. grpcio). For an MVP it would be best to get it in without thread management. 5. It should backport to 5.1. 6. Tested locally, with a routing scripting making many python_exec() calls Thank you. Anthony Alba

1 1

PUBLISH and ETag
by Cyrille Demaret 16 Feb '18

16 Feb '18

Hi, I'm using Kamailio with presence enabled and Asterisk PJSIP and outbound-publish. My problem is happening when I place 2 consecutive calls from Asterisk : When I make a first call Asterisk sent the following: PUBLISH sip:201@192.168.100.37 SIP/2.0 Via: SIP/2.0/UDP 192.168.100.37:5080;rport;branch=z9hG4bKPj4f9c19eb-26d8-4bb1-8f00-e69723a610 82 From: sip:201@mydomain.com;tag=a560e088-9e8a-49f2-a9b1-4a0ec31340bf To: sip:201@mydomain.com Call-ID: 5adcf0a0-f138-44d6-8c56-eaf7c3b3b183 CSeq: 10697 PUBLISH Event: dialog Expires: 180 Max-Forwards: 70 User-Agent: Asterisk PBX 14.6.0 Content-Type: application/dialog-info+xml Content-Length: 247 <?xml version="1.0" encoding="UTF-8"?> early.. Kamailio replies : SIP/2.0 200 OK Via: SIP/2.0/UDP 192.168.100.37:5080;rport=5080;branch=z9hG4bKPj4f9c19eb-26d8-4bb1-8f00-e6972 3a61082;received=192.168.100.37 From: sip:201@mydomain.com;tag=a560e088-9e8a-49f2-a9b1-4a0ec31340bf To: sip:201@mydomain.com;tag=b596189c6de9c38f624fd84638f43be6-ff39 Call-ID: 5adcf0a0-f138-44d6-8c56-eaf7c3b3b183 CSeq: 10697 PUBLISH Expires: 180 SIP-ETag: a.1518775074.19863.16.0 Server: kamailio (5.0.5 (x86_64/linux)) Content-Length: 0 When the call is done, Asterisk sent another PUBLISH telling that the call if terminated : PUBLISH sip:201@192.168.100.37 SIP/2.0 Via: SIP/2.0/UDP 192.168.100.37:5080;rport;branch=z9hG4bKPja93efb01-a518-445e-9e9b-f6f97ab8c7 52 From: sip:201@mydomain.com;tag=165fb3b2-ec0e-4786-889f-eb194ad456ce To: sip:201@mydomain.com Call-ID: 5adcf0a0-f138-44d6-8c56-eaf7c3b3b183 CSeq: 10698 PUBLISH Event: dialog SIP-If-Match: a.1518775074.19863.16.0 Expires: 180 Max-Forwards: 70 User-Agent: Asterisk PBX 14.6.0 Content-Type: application/dialog-info+xml Content-Length: 230 <?xml version="1.0" encoding="UTF-8"?> terminated.. And Kamailio replies : SIP/2.0 200 OK Via: SIP/2.0/UDP 192.168.100.37:5080;rport=5080;branch=z9hG4bKPja93efb01-a518-445e-9e9b-f6f97 ab8c752;received=192.168.100.37 From: sip:201@mydomain.com;tag=165fb3b2-ec0e-4786-889f-eb194ad456ce To: sip:201@mydomain.com;tag=b596189c6de9c38f624fd84638f43be6-48b4 Call-ID: 5adcf0a0-f138-44d6-8c56-eaf7c3b3b183 CSeq: 10698 PUBLISH Expires: 180 SIP-ETag: a.1518775074.19873.18.1 Server: kamailio (5.0.5 (x86_64/linux)) Content-Length: 0 Here, the SIP ETag is a.1518775074.19873.18.1. The problem is if I make a new call before the expiration of the previous SUBSCRIBE, Asterisk reuse this SIP ETag according to the RFC : PUBLISH sip:201@192.168.100.37 SIP/2.0 Via: SIP/2.0/UDP 192.168.100.37:5080;rport;branch=z9hG4bKPj9d13bb82-31d9-48db-9672-bd4b6b4f22 f0 From: sip:201@mydomain.com;tag=33e6b028-0444-4b3a-8bc2-4a987a291528 To: sip:201@mydomain.com Call-ID: 5adcf0a0-f138-44d6-8c56-eaf7c3b3b183 CSeq: 10699 PUBLISH Event: dialog SIP-If-Match: a.1518775074.19873.18.1 Expires: 180 Max-Forwards: 70 User-Agent: Asterisk PBX 14.6.0 Content-Type: application/dialog-info+xml Content-Length: 247 <?xml version="1.0" encoding="UTF-8"?> early. Kamailio refuse it with this error : "Trying to update an already terminated state. Skipping update." because the call is considered as terminated. The RFC is stating : When updating previously published event state, PUBLISH requests MUST contain a single SIP-If-Match header field identifying the specific event state that the request is refreshing, modifying or removing. This header field MUST contain a single entity-tag that was returned by the ESC in the SIP-ETag header field of the response to a previous publication. Why Kamailio is acting like that? Best regards, Cyrille

2 2

Problem with "transport=tcp" and rtpproxy
by Pau Giralt 16 Feb '18

16 Feb '18

Hi! I'm working with kamailio and rtpproxy. When I configure the clients to send sip messages through TCP adding the ";transport=tcp" in the sip message the rtpproxy is not working anymore. The code is never passing the point pasted below, inside route[NATMANAGE] in kamailio.cfg if (!(isflagset(FLT_NATS) || isbflagset(FLB_NATB))) return; Therefore, rtpproxy_manage("co"); is never reached. Could it be a parsing problem? ( check_route_param("nat=yes") ) The config file is exactly the same as the one provided in the kamailio repository (the default one) Thanks!

2 7

Raspberry Pi deb packages -> built
by Mititelu Stefan 15 Feb '18

15 Feb '18

Hi people! After a bit of research and lib installations, I've managed to build the full kamailio deb packages for a raspberry pi 3 (version B) running raspbian stretch (lite) that can be found at [1] (with a really few module exceptions that might be used in really specific scenarios -> I think) Just for fun, I'm planning to do some CPS stats tomorrow (if I got the time), with a basic config and also share here, but, if someone else tried it before and is able to share some stats/benchmarks, would be greatly appreciated (and interesting to compare, given my "decent" rpi overclock), because I'm also curious about VoIP stuff on raspberry pi! :D [1] https://github.com/smititelu/rpi/tree/master/kamailio/5.1.1 Thank you, Stefan

2 3

KEMI
by Aleksandar Sosic 15 Feb '18

15 Feb '18

Hi everyone, we're valuating on rewriting our kamailio routes in python. We're doing a lot of string manipulation and GraphQL API queries so the kamailio scripting language is a little bit limited for this use case. Can someone tell us if the KEMI framework is stable enough for production? Kind regards, -- Aleksandar Sosic mail: alex.sosic(a)timenet.it skype: alex.sosic cell: +385 91 2505 146

5 11

Re: [SR-Users] RTPEngine Transcoding
by Social Boh 15 Feb '18

15 Feb '18

Hello, I'm trying to use this configuration: Softphone A audio codec g722 and Opus Softphone B audio codec PCMA and PCMU Kamailio with rtpengine_manage with codec-transcode-PCMU codec-transcode-PCMA in the request Softphone A call Softphone B The RTPEngine DEBUG show many Decoder Error: https://pastebin.com/k1Zhw8HS Thank you Regards --- I'm SoCIaL, MayBe El 13/02/2018 a las 20:05, Richard Fuchs escribió: > On 2018-02-13 07:22 PM, Social Boh wrote: >> Thank you. >> >> which is the best way to implement ir >> >> codec-strip-all and the add codecs? >> >> Is there a document about how work the transcoding? > It's really up to you and depends on what you want to achieve. Doing a > strip-all and then adding codecs you want to offer is definitely one > way. Another is to just add codecs that you know or think the peer may > want to see. > > Cheers >

2 2

Invalid payload location
by Soltanici Ilie 15 Feb '18

15 Feb '18

Hi All, Can someone explain why i'm receiving such errors in syslog: [4839]: {INVITE (1)] | Seq=101 | Source IP=46.166.139.7 (udp)} ERROR: <core> [core/parser/sdp/sdp.c:212]: set_sdp_payload_attr(): Invalid payload location [4839]: {INVITE (1)] | Seq=101 | Source IP=46.166.139.7 (udp)} ERROR: <core> [core/parser/sdp/sdp.c:212]: set_sdp_payload_attr(): Invalid payload location [4839]: {INVITE (1)] | Seq=101 | Source IP=46.166.139.7 (udp)} ERROR: <core> [core/parser/sdp/sdp.c:228]: set_sdp_payload_fmtp(): Invalid payload location [4839]: {INVITE (1)] | Seq=101 | Source IP=46.166.139.7 (udp)} ERROR: <core> [core/parser/sdp/sdp.c:212]: set_sdp_payload_attr(): Invalid payload location This is how the SIP Packet is looking: 2018/02/15 10:51:40.916040 46.166.139.7:30189 -> X.X.X.X:5060 INVITE sip:01031203697179@X.X.X.X:5060 SIP/2.0 Via: SIP/2.0/UDP 46.166.139.7:30189;branch=z9hG4bK-5342365299746297160962993236828333618949;rport=30189 From: <sip:88893855391@46.166.139.7:30189>;tag=0vkn019bur To: <sip:01031203697179@X.X.X.X:5060> Contact: <sip:88893855391@46.166.139.7:30189> Call-ID: 0889633580294714035892119863939070195463 CSeq: 101 INVITE User-Agent: Hello Annonymous Max-Forwards: 70 Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY Supported: replaces Content-Type: application/sdp Content-Length: 356 v=0 o=SIP-Session 950029666 1307322426 IN IP4 46.166.139.7 s=SIP c=IN IP4 46.166.139.7 t=0 0 m=audio 5075 RTP/AVP 18 0 8 101 a=fmtp:101 0-15 a=rtpmap:18 G729/8000 a=rtpmap:0 PCMU/8000 a=rtpmap:8 PCMA/8000 a=rtpmap:3 GSM/8000 a=rtpmap:9 G722/8000 a=fmtp:9 bitrate=64000 a=rtpmap:104 G726-16/8000 a=rtpmap:101 telephone-event/8000 a=sendrecv Thank You.

1 0

python_exec return 0 causes script hangs; return -1 is treated like 0
by Anthony Alba 15 Feb '18

15 Feb '18

5.1.1: if python_exec method returns 0, then the script hangs If the python_exec method return -1, then expressions treat it as if it returned zero. Somewhere there seems to be an offset by 1 issue. if (!python_exec("method")) { # if you return -1, the script does not hang but treated like zero } ################## Example 1: class kamailio(object): def hello(self, msg): KSR.info("hello, world\n") return 0 Calling: python_exec("hello") will cause script hang ################### Example 2: class kamailio(object): def hello(self, msg): KSR.info("hello, world\n") return -1 Calling python_exec("hello") if (!python_exec("hello")) { ## method returns -1 ## why did we get here? xlog("L_INFO","method return zero!\n"); } shows INFO: <script>: method return zero!

2 1

[IGNORE] 5.1.1 python_exec, extra chars ' 0' appended to args
by Anthony Alba 15 Feb '18

15 Feb '18

Apologies for the noise; it is working now. Not sure how I got the extra " 0" appended to the method arguments for python_exec(). AAlba

1 0

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