sr-users August 2013

sr-users@lists.kamailio.org

93 participants
177 discussions

tls nat
by hiro 12 Sep '13

12 Sep '13

After many failures because of broken libraries I managed to decrypt my problematic TLS sessions providing the private key of kamailio to wireshark. With TLS+SRTP enabled my nokia phones send session progress and pracks with rtp port. This breaks NAT/rtpproxy in kamailio, which replaces the rtpproxy port in session progress, but then forgets about it for the 200 OK. Attached is a tree overview and the conversations of each phone with kamailio. hiro

2 5

RPC command dlg.list (for large amounts of dialogs)
by Alex Balashov 11 Sep '13

11 Sep '13

Hello, I am having trouble getting a large list of dialogs (800+) out of Kamailio using the dlg.list binrpc command. I posted a while back that the list of dialog data was coming back truncated, and was told to increase: modparam("ctl", "binrpc_max_body_size", ...) modparam("ctl", "binrpc_struct_max_body_size", ...) I did that, increasing the buffers to 20 MB. Then, when I ran sercmd -s unixs:/tmp/kamailio_ctl dlg.list I got: ERROR: reply too big which clearly seemed to be on the client side now. So, I looked in the sources for sercmd (now sercmd), in sip-router/utils/sercmd/sercmd.c, and found this: if ((ret=get_reply(s, reply_buf, MAX_REPLY_SIZE, cookie, &in_pkt, &msg_body))<0){ switch(ret){ case -1: goto error_read; case -2: goto error_parse; case -3: goto error_cookie; case -4: goto error_toobig; So, I adjusted: #define MAX_REPLY_SIZE to several megabytes as well, and tried the same with MAX_BODY_SIZE too. Now, when I run the dlg.list command, it just hangs and never returns anything. It was stuck there for well over 20 minutes. An 'strace' on sercmd reveals that it receives what appears to be the full listing, but keeps wanting to read() more. I don't have the exact output handy at the moment, but it looked like this: socket(PF_FILE, SOCK_STREAM, 0) = 3 connect(3, {sa_family=AF_FILE, path="/tmp/kamailio_ctl"}, 110) = 0 writev(3, [{"\241\3\v%s\230\4", 7}, {"\221\tdlg.list\0", 11}], 2) = 18 read(3, "\241\27\16\257%s\230\4\221Hhash:551:9521 state:4 "..., 786432) = 340000 read(3, [just hangs here] Now, I would have assumed that perhaps there is more to read and the output is not returning due to some endless locking of the dialog profile table, which is constantly changing. However, that cannot be the case because kamctl fifo dlg_list works fine for any amount of dialogs, and always returns all the data near-instantly! Any assistance is much appreciated! -- Alex [1] I don't have the output handy at the moment, but it was a read() value: -- Alex Balashov - Principal Evariste Systems LLC 235 E Ponce de Leon Ave Suite 106 Decatur, GA 30030 United States Tel: +1-678-954-0670 Web: http://www.evaristesys.com/, http://www.alexbalashov.com/

3 9

dialog save/restore
by Victor V. Kustov 10 Sep '13

10 Sep '13

Hi all! I make failover scheme with two kamailio+carp. Kamailio use dialogs for accounting. Solved tasks: K. as registrar (mysql DB, shared with asterisk) auth/accounting via radius usrloc DB in mysql. Need to solve: dialogs. Today if K1 server down, K2 take sip requests. New calls(dialogs) work perfectly, but if dialog created by K1, K2 had nothing about start and cant complete accounting. Of cause, on K2 i may load dialogs from DB. I think on timer or on event from carp. But may be exist better solutions. -- WBR, Victor JID: coyote(a)bks.tv JID: coyote(a)bryansktel.ru I use FREE operation system: 3.10.4-calculate GNU/Linux

2 3

rfc: accounting records time details
by Daniel-Constantin Mierla 05 Sep '13

05 Sep '13

Hello, there were old and recent discussions about the representation of time in accounting records. At this moment the acc module stores the unix timestamp as datetime value. In some countries is required to store more accuracy, beyond the seconds. Also, the datetime gives some troubles with converting back to unix timestamp, specifically with time zones and daylight saving times. I want to open a discussion that is visible for all users and devs, being something affecting probably everyone, input from anyone that is interested being relevant to select the right approach. Here are some suggestions presented so far. 1) store seconds.miliseconds as double - there is a patch (which probably needs some tunings itself) on tracker. The precision is shifted from seconds to milliseconds. Internally computed from microseconds from timeval structure. The other modules might need updates (iirc dbtext has only 2 decimals precisions for double, while miliseconds will require at least three -- could be the case for flatstore or other modules). Advantage is the computing of duration directly by subtraction of two values. 2) store seconds and microseconds as two separate values. Should be no issues with existing modules. Even more accuracy than above, no issues with other modules, but will require to use two columns (thus four values to compute the duration) Any other suggestions? Personally, at this moment, I will go with 2), but that might not meet everyone's needs. So a solution can be to make it configurable, as a bitmask of what time representation to store. Say, there will be a new parameter timestamp_mode for acc module: - bit one set - store seconds timestamp as for now (default) - bit two set - store seconds and microseconds in separate integer columns - bit three set - store seconds.miliseconds as double value in one column I could add the parameter and tune acc for bit one and two. For bit three a deeper review is needed for other modules, probably the developer that submitted to the tracker can do it. More suggestions? Pro or cons opinions? Cheers, Daniel -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio Advanced Training, San Francisco, USA - June 24016, 2013 * http://asipto.com/u/katu *

4 9

PUBLISH/SUBSCRIBE in pidf format
by Rupayan Dutta 05 Sep '13

05 Sep '13

Can anybody suggest me whether kamailio 4.0.3 supports pidf format for handling presence related stuff instead of XCAP server.If yes what are the corrosponding changes in the kamaolio cfg file? If not, then how can I test my kamailio presence with XCAP functionality with a browser based client like SIPML5.Actually I want the XCAP related stuff instead of pidf which is currently available in Dubango SIPML5

2 2

Call Baring
by arun Jayaprakash 04 Sep '13

04 Sep '13

Hello, with outgoing white list/black list I am able to put a prefix in the user preference table to block certain out going calls. In the same way is it possible to block incoming calls depending on certain prefixes ( like 1-800)? Regards, A

2 2

Modules required by Web socket
by Vijay Thakur 04 Sep '13

04 Sep '13

Hi All Members, I am trying to compiling Kamailio 4.0 with websocket support. Please tell that what modules should I compile during installation. Is there any list of all available modules. Is it possible to compile kamailio with all available modules. During compilation, I am getting the error "fatal error: radiusclient-ng.h: No such file or directory". Thanks, Vijay TH

2 1

Kamailio v4.0.2, iPhone, TCP connection and PJLIB
by Roberto Fichera 30 Aug '13

30 Aug '13

Hi All, Sorry for cross-posting this email to PJLIB, but maybe there are some things related. Anyhow! I'm having problems on kamailio v4.0.2 under Fedora 18 64bit and TCP client like iPhone using PJSIP as SIP library. Basically once the iPhone side in close the call (TCP->UDP) I'm getting the error below. Kamailio is running under a VPS without NATed network so it uses a real public address. Furthermore, note that tcp_main is answering to a 192.168.2.98 ip address which is the iPhone client. This looks really strange to me since it should answer directly to the public/port used for the registration and not to a such kind of reserved address. The kamilio configuration is basically the default with a very few changes like NAT, rtpproxy and postgresql backend. This problems doesn't happen at all when using UDP->UDP calls. But I cannot use it because as you certain know UDP connection under iPhone will not work when the application run in background mode. Can someone suggest how to solve this issue or maybe suggest a TCP working solution for iPhone? Thanks in advance. Roberto Fichera. Jul 30 16:21:53 proxy /usr/sbin/kamailio[9502]: ERROR: <core> [tcp_main.c:4432]: tcpconn_main_timeout(): connect 192.168.2.98:5060 failed (timeout) Jul 30 16:21:55 proxy rtpproxy[2262]: DBUG:get_command: received command "9483_9 D 12d1d19926c4ff742a52f0c855b1bb83@94.94.x.x:5060 as74e0c388 GROahimCK6KTrl5CkYEg7nuoPIIXZ8cj" Jul 30 16:21:55 proxy rtpproxy[2262]: INFO:handle_delete: forcefully deleting session 1 on ports 15604/17354 Jul 30 16:21:55 proxy rtpproxy[2262]: INFO:remove_session: RTP stats: 354 in from callee, 603 in from caller, 957 relayed, 0 dropped Jul 30 16:21:55 proxy rtpproxy[2262]: INFO:remove_session: RTCP stats: 5 in from callee, 2 in from caller, 7 relayed, 0 dropped Jul 30 16:21:55 proxy rtpproxy[2262]: INFO:remove_session: session on ports 15604/17354 is cleaned up Jul 30 16:21:55 proxy rtpproxy[2262]: DBUG:doreply: sending reply "9483_9 0 Jul 30 16:21:55 proxy rtpproxy[2262]: " Jul 30 16:22:04 proxy /usr/sbin/kamailio[9502]: ERROR: <core> [tcp_main.c:4432]: tcpconn_main_timeout(): connect 192.168.2.98:5060 failed (timeout) Jul 30 16:22:14 proxy rtpproxy[2262]: DBUG:get_command: received command "9496_16 D 12d1d19926c4ff742a52f0c855b1bb83@94.94.x.x:5060 GROahimCK6KTrl5CkYEg7nuoPIIXZ8cj as74e0c388" Jul 30 16:22:14 proxy rtpproxy[2262]: INFO:handle_command: delete request failed: session 12d1d19926c4ff742a52f0c855b1bb83@94.94.x.x:5060, tags GROahimCK6KTrl5CkYEg7nuoPIIXZ8cj/as74e0c388 not found Jul 30 16:22:14 proxy rtpproxy[2262]: DBUG:doreply: sending reply "9496_16 E8 Jul 30 16:22:14 proxy rtpproxy[2262]: "

3 22

kamailio load balancing
by yuanshui lee 30 Aug '13

30 Aug '13

Hello, I have a question about the load balancing module of kamailio. As the site http://kb.asipto.com/ say, Kamailio is as a SIP proxy router to scale Asterisk. Can I run a kamailio instance as load balancer, and other several instances as voice server replace of Asterisk? If I can do that, could you give me a tutorial? We are using kamailio as our server. Thank you very much.

2 1

Server silently droping Bye message.
by Helena Garcia-Nieto 30 Aug '13

30 Aug '13

Hello, Thanks in advanced for the help. I am absolutely new with kamailio and still struggling through silly problems so please forgive me if the solutions is so obvious. I have a network like Devices --à Kamailio --à pstn GW The PSTN GW is more or less out of my reach for changing the behaivour. As devices I have xlite, 3cx and an android app Kamailio is on version 4.0.2 I have problems with the calls to PSTN, When A is the Android app the server is not able to relay the Bye coming from the pstngw so the android app does not disconnect the call. This do not happened when A are xlite or 3cx. Bye is correctly sent to the PSTNGW if it comes from the android app. For calls inside the kamailio server there is no problem with the devices and all the byes are sent ok. When I call from the android app I have to drop some of the parameters on the Contact header because the pstngw do not like them and drop the call otherwise (responding with a 500 IRP error) I droped the parameters using: subst('/^Contact: (.*)>.*$/Contact: \1>/i'); #delete the parameter that is anoying pstnGW on contact header I also change request uri deleting the 00, but I do this modification on all the calls. I have tried without this ruri modification and the resoult is the same xlite ok android ko so I do not think that is affecting. subst_uri('/^sip:00(.*)/sip:\1/i'); # delete the 00 Searching on the kamailio log I found that in the android call: DEBUG: tm [t_lookup.c:1395]: t_newtran(): DEBUG: t_newtran: msg id=113 , global msg id=112 , T on entrance=0xffffffffffffffff DEBUG: tm [t_lookup.c:534]: t_lookup_request(): t_lookup_request: start searching: hash=30121, isACK=0 DEBUG: tm [t_lookup.c:492]: matching_3261(): DEBUG: RFC3261 transaction matching failed DEBUG: tm [t_lookup.c:716]: t_lookup_request(): DEBUG: t_lookup_request: no transaction found DEBUG: tm [t_hooks.c:374]: run_reqin_callbacks_internal(): DBG: trans=0x7f60cadc8d10, callback type 1, id 0 entered While on the xlite call I can see tm [t_lookup.c:1395]: t_newtran(): DEBUG: t_newtran: msg id=116 , global msg id=115 , T on entrance=(nil) tm [t_lookup.c:534]: t_lookup_request(): t_lookup_request: start searching: hash=64449, isACK=0 tm [t_lookup.c:477]: matching_3261(): DEBUG: RFC3261 transaction matched, tid=1cbf.617b985b6620c855b501dfbfced5f7a4.0 tm [t_lookup.c:733]: t_lookup_request(): DEBUG: t_lookup_request: transaction found (T=0x7f60cadb2510) When processing the BYE. So Im thinking that the problem is that Kamailio do not know where to send it. Im not sure that is the problem. Ive checked and tags and session information seem ok to me, so I do not understand what is wrong with the Bye so the server cannot find where it should be sent to. I attached a trace with both calls first xlite and then android This is the Bye on the XLite call Frame 59: 788 bytes on wire (6304 bits), 788 bytes captured (6304 bits) Ethernet II, Src: Cisco_67:50:00 (macAdd), Dst: FujitsuT_80:7d:d8 (macAdd) Internet Protocol Version 4, Src: PSTNGWIP (PSTNGWIP), Dst: ServerIP (ServerIP) User Datagram Protocol, Src Port: sip (5060), Dst Port: ServerPort (ServerPort) Session Initiation Protocol (BYE) Request-Line: BYE sip:XliteAnumber@publicXliteIP:55838;transport=tcp SIP/2.0 Method: BYE Request-URI: sip:XliteAnumber@publicXliteIP:55838;transport=tcp [Resent Packet: False] Message Header Via: SIP/2.0/UDP PSTNGWIP;branch=z9hG4bK5d4.f93633a612aca237ad1f5c341c1b2986.0 Via: SIP/2.0/UDP PSTNGWIP:5061;branch=z9hG4bK-fat3ctwlimdw3lkb;rport=5061 Route: <sip:ServerIP:ServerPort;r2=on;lr=on;nat=yes> Route: <sip:ServerIP:ServerPort;transport=tcp;r2=on;lr=on;nat=yes> Max-Forwards: 16 From: 00Bnumber <sip:00Bnumber@ServerIP:ServerPort>;tag=eihmt4yopie2j4fq.i To: <sip:XliteAnumber@ServerIP:ServerPort>;tag=2b8f312e Call-ID: NGZlMDU3OWE2YjM1ZGQ1OTEyM2NiNmI0YjM1MjA0YTg CSeq: 515 BYE Sequence Number: 515 Method: BYE Contact: Anonymous <sip:PSTNGWIP:5061> User-Agent: Sippy cisco-GUID: 521708479-278139363-3214082078-3387785402 h323-conf-id: 521708479-278139363-3214082078-3387785402 and that one is the one on the Android one: Frame 115: 766 bytes on wire (6128 bits), 766 bytes captured (6128 bits) Ethernet II, Src: Cisco_67:50:00 (macAdd), Dst: FujitsuT_80:7d:d8 (macAdd) Internet Protocol Version 4, Src: PSTNGWIP (PSTNGWIP), Dst: ServerIP (ServerIP) User Datagram Protocol, Src Port: sip (5060), Dst Port: ServerPort (ServerPort) Session Initiation Protocol (BYE) Request-Line: BYE sip:AndroidAnumber@PrivateAndroidIP:35986;transport=tcp SIP/2.0 Method: BYE Request-URI: sip:AndroidAnumber@PrivateAndroidIP:35986;transport=tcp [Resent Packet: False] Message Header Via: SIP/2.0/UDP PSTNGWIP;branch=z9hG4bKf0a2.679a023eeebfc2b9acae8c683386146f.0 Via: SIP/2.0/UDP PSTNGWIP:5061;branch=z9hG4bK-ge57vxpelnqy57yg;rport=5061 Route: <sip:ServerIP:ServerPort;r2=on;lr=on;nat=yes> Route: <sip:ServerIP:ServerPort;transport=tcp;r2=on;lr=on;nat=yes> Max-Forwards: 16 From: <sip:00Bnumber@ServerIP:ServerPort>;tag=2yapycnnrb7yjgwj.i To: <sip:AndroidAnumber@ServerIP:ServerPort>;tag=652340227 Call-ID: fa8e5892-e88f-fd9b-8f99-0545ac8e9e8e CSeq: 438 BYE Sequence Number: 438 Method: BYE Contact: Anonymous <sip:PSTNGWIP:5061> User-Agent: Sippy cisco-GUID: 521708482-278139363-3214082078-3387785402 h323-conf-id: 521708482-278139363-3214082078-3387785402 Both Devices are behind NAT. I hope I explained my problem correctly but do not hesitate to ask any other thing. Thanks in advanced Helena

2 6

← Newer
1
2
3
4
5
6
7
...
18
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 August 2013