Re: [OpenSIPS-Users] ACK looping issue with end-point co-located with OpenSIPS
Daniel, have you added perhaps your IP address to the domain table? It is not supposed to be there and should be removed, otherwise opensips thinks 192.168.117.4:5070 is itself as others pointed out. Also, if you have any checks in your script like: if ( !is_from_local() !is_uri_host_local() ) you should replace them with the version which doesn't use domain table: if ( from_uri != myself uri != myself ) HTH. Andrew On 10/16/2012 02:33 AM, Ali Pey wrote: What do you have for alias in your opensips.cfg? Opensips seems to think 192.168.117.4:5070 http://192.168.117.4:5070 is itself. Regards, Ali Pey On Mon, Oct 15, 2012 at 6:33 PM, Daniel Eiland daniel.eil...@gmail.com mailto:daniel.eil...@gmail.com wrote: Hi folks, I've got an issue with ACK messages being looped when they are sent to an endpoint that is co-located with my OpenSIPS proxy. I've got OpenSIPS located on server A @ port 5060 and two conference endpoints: C01 on server A @ port 5070 and C02 on a separate server. When my client calls into C01, the ACK message is continually routed by OpenSIPS back into itself instead of the conference endpoint listening on port 5070. This doesn't happened when I call into C02, OpenSIPS properly routes it to the right destination. When I compare the two ACKs, they are fairly similar and when I looked at some OpenSIPS logs both messages are being routed in a similar fashion namely loose_route is true. The only difference (which I'm sure if the problem, I'm just not sure why) is that the ACK to C01 has Destination User ($du) of 192.168.117.4 while the ACK to C02 has a $du of NULL. If anyone has any suggestions, I'd be grateful. Thanks, Daniel Also here are the two ACKs for comparison: ACK sip:C01@192.168.117.4:5070;transport=udp SIP/2.0 Via: SIP/2.0/UDP QWE.RTY.XYZ.ABC:2453;rport;branch=z9hG4bKPjca04290517874935af64a839e6bf9701 Max-Forwards: 70 From: deiland sip:1001@192.168.117.4 mailto:sip%3A1001@192.168.117.4;tag=d02a111d4a064b05a7cf987b006bd001 To: sip:C01@192.168.117.4 mailto:sip%3AC01@192.168.117.4;tag=ma94e0688avFj Call-ID: 5c35c0a5dcb442a9afd324c988bd0a3c CSeq: 24503 ACK Route: sip:192.168.117.4;lr User-Agent: Blink 0.2.7 (Windows) Content-Length: 0 ACK sip:C02@192.168.155.211:5070;transport=udp SIP/2.0 Via: SIP/2.0/UDP QWE.RTY.XYZ.ABC:2122;rport;branch=z9hG4bKPj5318936898454066944aadf64ad846d0 Max-Forwards: 70 From: deiland sip:1001@192.168.117.4 mailto:sip%3A1001@192.168.117.4;tag=ff6fb0856e264e7f96601529148bb206 To: sip:C02@192.168.117.4 mailto:sip%3AC02@192.168.117.4;tag=54aZ48KZ7a9Xj Call-ID: afddb9bea6714f95bd84dff2159e2b14 CSeq: 11988 ACK Route: sip:192.168.117.4;lr User-Agent: Blink 0.2.7 (Windows) Content-Length: 0 ___ Users mailing list Users@lists.opensips.org mailto:Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to generate early media until call is established?
Hi Adam, just a wild guess - try to insert t_reply(180, Ringing) before t_relay in your script. On 06/06/2012 10:34 AM, Adam Raszynski wrote: Hi All, Simple scenario: - OpenSIPS as call router to SIP termination provider - I have no control on remote gateways and can't generate early media there Current situation: - After dialing a number user hears silence until call is routed by my termination provider, call routing to mobile networks sometimes takes 10 or more seconds before RINGING or BUSY response I would like to generate call progress in early media until some meaningful response is generated by termination provider I have local FreeSwitch based media/application server and can use it to generate the tone So the only question is how to route early media to FreeSwitch while making a call and how to disable it when response comes from my provider? Kind Regards ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] RTP Proxy Between Public and Private IPs
Arjun, I'm not sure if running rtpproxy behind NAT is a good idea, but there is a patch on internet for adding advertised address setting to rtpproxy: http://opensips-open-sip-server.1449251.n2.nabble.com/Rtpproxy-behind-the-NAT-td5008041.html I've used it ok for some experiment. Also I can suggest looking at Sipwise sip:provider, which supports advertised_address setting for both signaling and RTP out of the box. If you are looking for paid support for setting up a RTP proxy behind a NAT, please contact me off list. Good luck. Andrew On 06/01/2012 11:16 AM, Arjun Shankar K S wrote: Hi All, Greeting to everyone!! I have openSIPS installed successfully behind NAT. I am facing issues in setting up RTP Proxy in the same network where we have a Public IP and an Private IP. I have configured RTP Proxy V1.2.0 in the same machine as openSIPS which has an internal IP as 10.196.15.212 and I have started the RTP Proxy using the following command, ./rtpproxy -l PublicIP/10.196.15.212 -s udp:127.0.0.1:7890 -F I am still facing issue where the Call do not get established properly due to improper NAT Traversal. Can anybody provide some support in setting up a RTP proxy behind a NAT. Any help is sincerely appreciated. Thanks, Arjun ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] RTP Proxy Between Public and Private IPs
On 06/01/2012 03:07 PM, Arjun Shankar K S wrote: # ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy Does this error mean that thr rtp proxy is unable to communicate back to the openSIP Server? Is the issue resolvable? Usually this means that rtpproxy didn't like the request from opensips. Could you please check the rtpproxy debug (either by running it in foreground or by configuring it to write to syslog)? ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] drouting dr_rules prefix only supports numeric
On 01/26/2012 08:49 PM, Anil M Pannikode wrote: Just wondering if there is a reason not to allow non-numeric chars in prefix column of dr_rules ? Anil, hope this answers your question: http://lists.opensips.org/pipermail/users/2010-July/013935.html ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How block Register attack
On 01/16/2012 06:35 AM, nick_ch...@ezmobo.com wrote: How to block register attack? That is the exact purpose of ratelimit module. You can do automatic ratelimit as defined in the params or you can do forced ratelimiting for every new REGISTER. Please check the readme of ratelimit module: http://www.opensips.org/html/docs/modules/1.7.x/ratelimit.html ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Call pickup
On 01/15/2012 11:42 PM, Dmitriy Abramov wrote: Hi, Bogdan. Where i can get full list of MI command? I saw http://www.opensips.org/Resources/DocsCoreMi17, but i want to know how i can get info about calls in opensips? /Regards,/ /Dmitriy/ You should look for the paragraph Exported MI Functions in the documentation of your module e.g.: http://www.opensips.org/html/docs/modules/1.7.x/tm.html#id294563 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS in a University Environment
You may want to get in contact with uniza.sk. I believe they are using opensource voip as they have published some excellent tutorials like http://nil.uniza.sk/sip/nat-fw/configuring-nat-traversal-using-kamailio-31-and-rtpproxy-server even though it is about kamailio and not opensips On 01/10/2012 07:46 PM, Gabriel Kuri wrote: Does anyone know of any Universities running OpenSIPS for local call routing between handsets? We're looking at replacing our old Avaya system or upgrade it, and the forklift upgrade from Avaya is ridiculously expensive (no surprise). We'd like to replace our Avaya system with a combination of OpenSIPS and FreeSWITCH and some Cisco routers for external PSTN access, but it's going to be a tough sell to our CIO, unless we can show someone else has done it already. Any pointers to other Universities would be great. Cheers, Gabe ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] About msilo module config
On 12/20/2011 02:31 AM, Kevin wrote: Who can tell me how to configure the msilo module to store all of the off line messages? Or give me a configure example? See here http://www.opensips.org/html/docs/modules/devel/msilo.html#id293310 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] nathelper config
Nick, On 10/14/2011 05:15 AM, Nick wrote: Hello It's my log , network is wifi. U 2011/10/14 10:08:47.915408 220.130.6.175:42666 - 10.10.12.91:5060 REGISTER sip:10.10.12.91 SIP/2.0. This is log for the registration, not for the call. As I said you should check yourself if the sdp for INVITE and 200 OK sent to SIP UA have the proper IP addresses - they should be rtpproxy ip address. Don't know if you can expect more help from the community mailing list for free. OpenSIPS is not like asterisk:) It is much more flexible but you should learn how these things actually work. ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] nathelper config
On 10/13/2011 11:30 AM, Nick wrote: When I start opensips. It display error. Can you give me any suggest? Thanks Oct 13 16:27:49 [20343] DBG:core:find_cmd_export_t: force_rtp_proxy not found The force_rtp_proxy() function was removed in 1.6.4 and needs to be replaced with rtpproxy_offer() / rtpproxy_answer(). http://www.opensips.org/Resources/DocsMigration163to164 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] nathelper config
On 10/13/2011 11:46 AM, Nick wrote: Oct 13 16:44:10 [26809] ERROR:rtpproxy:mod_init: no rtpproxy set specified You should define the rtpproxy command socket with rtpproxy_sock param: http://www.opensips.org/html/docs/modules/devel/rtpproxy.html#id250023 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] nathelper config
On 10/13/2011 12:21 PM, Nick wrote: Hello Andrew You mean is I need install rtpproxy server?? Only loadmodule rtpproxy.so is not active?? Thanks for your support. Yes, install rtpproxy server, configure it to start on some socket (unix of udp) and put that socket as rtpproxy_sock parameter in opensips.cfg. Note that if you use the unix socket you need to take care that opensips has write permissions into it. ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Not writing to Opensips log
On 10/13/2011 04:46 PM, Nauman Sulaiman wrote: Hi, I deleted the 1.6 opensips log i had redirected logging to /var/log/opensips.log as it had become huge. However opensips is not logging to it anymore. I've done a chmod 777 on the new log file but that still does not help. Can anyone tell me how to get it logging again without having to reboot the whole server. I guess you need to restart your rsyslog (not sure how it is called in your distribution). Rsyslog is keeping the old log open. ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] nathelper config
Nick, There is not enough information in your email to give you a suggestion. Please collect the sip trace with ngrep; you need to verify that IP address in SDP c-line is changed to the address of your proxy in both directions. Looking at your config I don't see calls to rtpproxy_offer() To check the NAT status of the callee tell usrloc to load it into flag 6: modparam(usrloc, nat_bflag, 6) Then if any of the caller, callee are behind NAT, call rtpproxy_offer in request route and rtpproxy_answer in reply route. Do not expect that both parties will be behind NAT at the same time and it will work well. On 10/13/2011 01:15 PM, Nick wrote: Hello Andrew Thanks. I installed rtpproxy server OK. And then, I started opensips server OK. I have two iphone 4, I want to test sip and video. Network setting, one is wifi, other is 3G 3G wifi iphone - opensips - NAT device - iphone I can see wifi video and listen voice in my 3G network. But I can't see 3G video and listen voice in my wifi network. Can you give me any suggest?? Thanks advance. Nick ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] rtpproxy in bridge mode
On 10/12/2011 07:55 PM, Brad Bendy wrote: Adding in some check for SDP did take care of that problem, just appears the outbound leg upstream is not getting rewritten with the correct external address, still using the internal one. From what Ive read you can use rtpproxy_answer() and rtpproxy_offer() depending if the 183/200 has SDP in it, right now ive been using rtpproxy_offer(FAIIO) when the dst_ip is the private IP and that side is rewritten correctly, but in on_reply route doing the same thing but FAEEO no rewrite appears to be happening. Ive looked at the alg.cfg example in the nathelper examples directory Sorry, you may be missing the point. Why II and EE? You need to call rtpproxy with IE flags for the calls going out of internal network and EI for the calls from outside to internal network. They shouldn't be mixed. Also are you sure you want to use the A flag? a - flags that UA from which message is received doesn't support symmetric RTP. IIRC most UAs support symmetric RTP. Andrew ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] creating a regular expression
On 10/10/2011 03:20 PM, Toyima Dias wrote: sorry, the range is 31297 - 31336... You should be able to do it with something like ^31[2-3][90123][7-9,0-6]$ , please check. 2011/10/10 Toyima Dias toyim...@gmail.com Hello, i would like some help on a regular expression using dialplan module and regular expressions...this is what i want: construct a regular expression for the following range: 31297-313336...i heve this simple string ^31[2-3]the problem is the range from 97 to 36...any help please? Thanks ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Change the Via sent by OpenSIPS
On 09/21/2011 01:54 PM, John Quick wrote: Now we are trying to make inbound calls go to some IVR's on his LAN so I would like the Via to be the LAN address. Is there any way to change advertised_address on a per-call basis? You can use set_advertised_address() core function: http://www.opensips.org/Resources/DocsCoreFcn17#toc136 Also you need opensips to insert two Record-Route headers, and set mhomed=1, IIRC. ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Load Balancing Using OpenSIPS
On 09/13/2011 10:05 AM, Faisal Rehman wrote: Thanks for your prompt response, yeah I have seen that table specified only for the load balancing work in opensips database, but I got a task to do load balancing without any database involvement, so is that something I can do? In that case you would have to hardcode the IP addresses in the config file which is not very nice. Alternatively, you could use a simple file-based dbtext database like: id(int,auto) setid(int) destination(str) priority(int) flags(int) description(str) 1:1:sip\:10.0.0.1\:5060:1:0:box1 2:2:sip\:10.0.0.2\:5060:1:0:box2 OpenSIPS works with that just like with a normal SQL DB. Regards, Andrew ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] RTPProxy bridge mode issue
Yuri, On 17.08.2011 07:27, Yuri Kirsanov wrote: Do you have any idea why is it behaving this way? I'm using OpenSIPS 1.6.4 and RTP Proxy 1.2.1 without any patches. I'm not using mhomed=1 option, I'm using force_send_socket() on outgoing calls from Internet client to LAN and it works fine. Should I also use that command for outgoing calls from LAN to Internet client? But that shouldn't affect RTP Proxy behavior, shouldn't it? Also, I've tried to use r and w options when invoking RTP Proxy, but that doesn't help. Have you tried using options ie for calls from lan to internet, ei vice versa and ee for calls from one client on the internet to another? This used to work for me.. one flag i|e may be not enough. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] SIP tracing to plain text file
Hello all, I'm looking for a way to do SIP tracing by peer and save those messages to a separate local file. I don't want to save messages to the database, as I'm looking for something really simple - a plain text file that is well formatted and easy to parse ;-) Can you tell me what are my options? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Nathelper ping does not consistently ping all contacts
James, On 01.07.2011 06:42, James Lamanna wrote: Hi, I've noticed after a period of time, Nathelper will stop sending pings to some contacts. I've verified that the contact is still registered (it is even in the location table) but the ping process appears to skip some contacts for unknown reasons. maybe see if this fixes the problem for you: http://www.mail-archive.com/users@lists.opensips.org/msg16200.html ? Could someone please look into this? I have phones behind NAT that stop being able to receive calls because firewalls close down the UDP mapping since this feature is not working properly. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] SRV TCP lookups
Hello Jarle, AFAIK the TCP will be used based if it has the lowest order number in NAPTR response, as per order and preference based selection procedure described in rfc 2915/3403 section 2: Order A 16-bit unsigned integer specifying the order in which the NAPTR records MUST be processed to ensure the correct ordering of rules. Low numbers are processed before high numbers, and once a NAPTR is found whose rule matches the target, the client MUST NOT consider any NAPTRs with a higher value for order (except as noted below for the Flags field). I think you should be able to see the SRV queries if you enable verbose debug and run tail -f /var/log/opensips.log|grep dns. On 15.06.2011 15:57, Jarle Lervik wrote: Hi all list users, Trying to figure out the best way to make OpenSIPS do TCP SRV lookups. I see from the logs that NAPTR and UDP SRV lookups are performed, but no TCP SRV lookups. This might be expected behavior but would like to figure out how I can enable OpenSIPS to also do TCP SRV lookups. For some domains there are no NAPTR records or UDP SRV records, so looking for a way to enable this in general or to “trick” OpenSIPS to include TCP SRV lookups. Will appreciate any tips or suggestions on this. Thanks in advance! -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Route group inside of drouting
On 13.06.2011 16:40, Kent Pirlo wrote: 212555, gwlist = 3,5,1 now.. lets say gw 3 actually needs to try multiple ips for that carrier before going on to gw 5, is this possible while using drouting or do i need to scrap the drouting module to do something complex like this.. It is possible and described in the module documentation: Also the module allows the usage of groups in the destination lists. A group of destinations is delimited by semi-colon char. inside the whole destination list ( like: 2,4;5,78,23;4;7;2 ). The destinations from within a group may be act differently (like load-balancing, random selection, etc), depending of the “sort_order” parameter - more about this is available under the “do_routing()” function section. http://www.opensips.org/html/docs/modules/1.6.x/drouting.html#id294582 -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] MySQL connection error
Liviu, Could it be that opensips is installed in /usr/local and is therefore using db_url from /usr/local/etc/opensips.cfg ? It's my best guess. On 10.06.2011 15:22, Barsan Liviu wrote: Hello, We re-installed from sources the OpenSIPs server to Debian and copied back to /etc/opensips the configuration files saved from the previous working OpenSIPs. When I start the server with opensips start then we receive the error in the log: Jun 10 18:03:12 P4302 opensips[18307]: ERROR:db_mysql:db_mysql_connect: driver error(1045): Access denied for user 'opensips'@'localhost' (using password: YES) Jun 10 18:03:12 P4302 opensips[18307]: ERROR:db_mysql:db_mysql_new_connection: initial connect failed Jun 10 18:03:12 P4302 opensips[18307]: ERROR:core:db_do_init: could not add connection to the pool For sure we do not have db_url in opensips.cfg or other files which contains username 'opensips'. Checked and from command line we can connect from mysql with the credentials we set in opensips.cfg. To be sure we put back the initial opensips.cfg comming with the sources, changed the db credentials to ours and still receive this error. Do you have any suggestion? Thanks, Liviu -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] MySQL connection error
/opensips.README.Debian /usr/src/opensips-1.6.4-2-tls/packaging/debian/opensips.dirs /usr/src/opensips-1.6.4-2-tls/packaging/debian/opensips.init /usr/src/opensips-1.6.4-2-tls/packaging/debian/opensips.postinst /usr/src/opensips-1.6.4-2-tls/packaging/debian/opensips.examples /usr/src/opensips-1.6.4-2-tls/packaging/rpm/opensips.default /usr/src/opensips-1.6.4-2-tls/packaging/rpm/opensips.spec.CentOS /usr/src/opensips-1.6.4-2-tls/packaging/rpm/opensips.spec.SuSE /usr/src/opensips-1.6.4-2-tls/packaging/rpm/opensips.init /usr/src/opensips-1.6.4-2-tls/packaging/rpm/opensips.init.SuSE /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.default /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.README.Debian /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.dirs /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.init /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.postinst /usr/src/opensips-1.6.4-2-tls/packaging/debian-etch/opensips.examples /usr/src/opensips-1.6.4-2-tls/packaging/gentoo/opensips-1.6.4.ebuild /usr/src/opensips-1.6.4-2-tls/packaging/gentoo/opensips.init /usr/src/opensips-1.6.4-2-tls/opensips /usr/src/opensips-1.6.4-2-tls/opensips.8 /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.8 /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.sqlbase /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.db_berkeley /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.oracle /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.mysql /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.dbtex t /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.pgsql /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.oracle /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.dbtext /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.db_berkeley /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.fifo /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbfunc.oracle /usr/src/opensips-1.6.4-2-tls/scripts/dbtext/opensips /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.unixsock /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.pgsql /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.base /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl.ctlbase /usr/src/opensips-1.6.4-2-tls/scripts/opensipsdbctl.base /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctlrc /usr/src/opensips-1.6 .4-2-tls/scripts/opensipsctl.mysql /usr/src/opensips-1.6.4-2-tls/scripts/opensipsctl /usr/src/opensips-1.6.4-2-tls/scripts/db_berkeley/opensips /usr/src/opensips-1.6.4-2-tls/modules/perl/opensipsxs.xs /usr/src/opensips-1.6.4-2-tls/modules/xmpp/doc/opensips-xmpp.cfg /usr/src/opensips-1.6.4-2-tls_src.tar.gz /usr/src/opensips-cp_4.0.tgz /usr/bin/opensips-mi-proxy /usr/share/pyshared/mediaproxy/interfaces/opensips.py /usr/share/pyshared/opensips_mi_proxy-1.0.4.egg-info /usr/share/pyshared/xcap/interfaces/opensips.py /usr/share/pyshared/xcap/interfaces/backend/opensips.py /usr/share/pyshared/miproxy/opensips.py /usr/share/doc/opensips-mi-proxy /usr/share/man/man1/opensips-mi-proxy.1.gz /usr/share/python-support/opensips-mi-proxy.public /tmp/opensipsInstall.log /var/lib/update-rc.d/opensips-m i-proxy /var/lib/dpkg/info/opensips-mi-proxy.list /var/lib/dpkg/info/opensips-mi-proxy.conffiles /var/lib/dpkg/info/opensips-mi-proxy.md5sums /var/lib/dpkg/info/opensips-mi-proxy.prerm /var/lib/dpkg/info/opensips-mi-proxy.postinst /var/lib/dpkg/info/opensips-mi-proxy.postrm /var/lib/mysql/opensips /var/cache/apt/archives/opensips-mi-proxy_1.0.4_all.deb /var/www/opensips-cp /var/www/opensips-cp/config/tools/system/cdrviewer/opensips_cdrs_1_6.pgsql /var/www/opensips-cp/config/tools/system/cdrviewer/opensips_cdrs_1_6.mysql /var/run/opensips-mi-proxy /var/log/opensips.log /sbin/opensipsdbctl /sbin/opensips /sbin/opensipsctl /sbin/opensipsunix /share/doc/opensips /share/man/man8/opensipsctl.8 /share/man/man8/opensipsunix.8 /share/man/man8/opensips.8 /share/man/man5/opensips.cfg.5 /share/opensips /share/opensips/dbtext/opensips Thanks, Liviu -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] MySQL connection error
Hi Liviu, I think Nethra is right about the modparams, please check. On 10.06.2011 16:03, Barsan Liviu wrote: Hello, The opensips -h returns what is expected: root@P4302:~# opensips -h version: opensips 1.6.4-2-notls (i386/linux) Usage: opensips -l address [-l address ...] [options] ... I haven't any file in /usr/local/atc And I see now the first time that cfg-target can be given at make. Thanks, Liviu -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] /etc/init.d/opensips restart returns OK when opensips is down...
On 06.06.2011 14:15, Toyima Dias wrote: If opensips has not started (ps -ef | grep opensips doesn't show any opensips pid), how is possible that /etc/init.d/opensips returns [OK] when you execute /etc/init.d/opensips restart...is this beacuse of mysql? is that what you mean...this script should return 1 (not succesfull, different than 0) it no pid file has been created, am i right? Correct, but there can be a case when opensips starts but then some module's initialization function fails due to mysql connect error or incompatible table version and the opensips will exit. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS 1.6.5 release
Hello, I'm also interested in the 1.6.5. On 11.05.2011 23:49, Bobby Smith wrote: Howdy, There are some critical bug fixes/changes around rtpproxy module and codec manipulation in trunk but not in an official gold release yet. Is there any tentative info on when we might see a 1.6.5 release, or perhaps a 1.7.X? I know there have been comments on the future with OpenSIPS 2.0, but at this point from a business perspective I'm just trying to sell a gold release to update and get these changes in. Thanks much, Bobby Smith -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to change transport protocol of R-URI
Nick, Have you tried t_relay() with parameters? http://www.opensips.org/html/docs/modules/devel/tm.html#trelay-1 On 18.05.2011 07:56, n...@uni-petrol.com wrote: Dear All! I need to change transport protocol of R-URI from UDP to TCP and vice verse. But unfortunately $rP variable (reference to transport protocol of R-URI) is read only. How to do this? Thanks in advance! -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] incorrect port 0 in reply from rtp proxy' error appears after DoS attack
Hi, I've recently experienced an opensips outage. As I see from the logs there was a brief DoS attack by sipvicious, then the attacker was blocked by fail2ban. But after that appeared a huge number of incorrect port 0 in reply from rtp proxy errors and the calls weren't going through. I was not able to monitor this myself, eventually the opensips was restarted. I'm going to upgrade from 1.6.2 to 1.6.4 and enable rtpproxy debug so as to get more info next time that it happens. I'm just wondering now if it this some known bug. Thanks, grep ERROR opensips.log May 17 11:53:42 sip /usr/sbin/opensips[26161]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:53:42 sip /usr/sbin/opensips[26161]: ERROR:core:parse_cseq: bad cseq May 17 11:53:42 sip /usr/sbin/opensips[26161]: ERROR:core:get_hdr_field: bad cseq May 17 11:53:42 sip /usr/sbin/opensips[26162]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:53:42 sip /usr/sbin/opensips[26157]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:53:42 sip /usr/sbin/opensips[26159]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:53:42 sip /usr/sbin/opensips[26162]: ERROR:core:parse_cseq: bad cseq May 17 11:53:42 sip /usr/sbin/opensips[26157]: ERROR:core:parse_cseq: bad cseq May 17 11:53:42 sip /usr/sbin/opensips[26161]: ERROR:core:pv_get_callid: cannot parse Call-Id header ... May 17 11:54:06 sip /usr/sbin/opensips[26164]: ERROR:core:pv_get_callid: cannot parse Call-Id header May 17 11:54:06 sip /usr/sbin/opensips[26162]: ERROR:core:pv_get_callid: cannot parse Call-Id header May 17 11:54:06 sip /usr/sbin/opensips[26155]: ERROR:tm:t_lookup_request: too few headers May 17 11:54:06 sip /usr/sbin/opensips[26156]: ERROR:tm:t_lookup_request: too few headers May 17 11:54:06 sip /usr/sbin/opensips[26155]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:54:07 sip /usr/sbin/opensips[26156]: ERROR:core:parse_cseq: expecting CSeq EoL May 17 11:54:07 sip /usr/sbin/opensips[26155]: ERROR:core:parse_cseq: bad cseq May 17 11:54:07 sip /usr/sbin/opensips[26156]: ERROR:core:parse_cseq: bad cseq May 17 11:54:07 sip /usr/sbin/opensips[26155]: ERROR:core:get_hdr_field: bad cseq May 17 11:54:07 sip /usr/sbin/opensips[26156]: ERROR:core:get_hdr_field: bad cseq May 17 11:54:07 sip /usr/sbin/opensips[26155]: ERROR:core:pv_get_callid: cannot parse Call-Id header May 17 11:54:07 sip /usr/sbin/opensips[26156]: ERROR:core:pv_get_callid: cannot parse Call-Id header May 17 11:55:01 sip /usr/sbin/opensips[26156]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy May 17 11:55:02 sip /usr/sbin/opensips[26165]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy May 17 11:55:02 sip /usr/sbin/opensips[26157]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy May 17 11:55:03 sip /usr/sbin/opensips[26155]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy May 17 11:55:04 sip /usr/sbin/opensips[26159]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy May 17 11:55:07 sip /usr/sbin/opensips[26165]: ERROR:nathelper:force_rtp_proxy_body: incorrect port 0 in reply from rtp proxy -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] [OT] SIP presence and BLF testing tools
Is there any framework available to do functional and regression testing of SIP presence and BLF? Ideally I'm looking for a tool for asserting the signaling flow and contents of the XML body (state, version numbers etc), also I'd like to be able to extract the tags and call-id into variables so as to create INVITE with Replaces. Working with message body variables in sipp is a pain. I'm also familiar with SIPr, Net::SIP and a few others, but that doesn't look suitable for someone without much development experience. Spirent is a perhaps a little closer than others in implementing pickup scenario but awkward when you go beyond the default test scenarios. So I'm interested in any information, tips, suggestions, commercial tools etc. I'm asking here 'cause judging from the quality of their BLFpresence OpenSIPS and Kamailio got the testing process just right. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] ACK not Relayed to correct destination on RE-INVITE
Hi Ash, I don't see the re-INVITE in your trace but I see that ACK after 200 OK from Yealink is not routed properly. I think your Cisco 877 has some sort of SIP ALG enabled: Note 200 OK message from FreeSwitch contains: Record-Route: sip:1.1.108.70;lr;ftag=1063201394;did=212.ef2d2d26 Contact: sip:7772@1.1.108.68:5060;transport=udp After that Yealink sends an ACK: ACK sip:7772@2.2.239.241:1037;transport=udp SIP/2.0 Route: sip:1.1.108.70;lr;ftag=1063201394;did=212.ef2d2d26 Route header is correct, but Request-URI is wrong: it must contain remote target address from the Contact header: ACK sip:7772@1.1.108.68:5060;transport=udp I think that if you collected the SIP trace from Yealink you would have found that router put its own IP address in Contact of 200 OK, hence it put router's IP into Request-URI. The very first INVITE message from the Yealink is not quite right too: source ip:port is 2.2.239.241:1034 but Contact and Via contains 2.2.239.241:1029. This might be caused but STUN not working properly, but since the problem occurs in both ways and with different UAs, Cisco 877 is likely modifying the SIP headers. On 29.04.2011 08:28, Ash wrote: Hi there, I have been trying for the last week to configure the load balancing in OpenSIPS. I am trying to configure a load balancer as per the wiki on the Freeswitch page - http://wiki.freeswitch.org/wiki/Enterprise_deployment_OpenSIPS, I have also tried the sample on the OpenSIPS site and I am getting the same results. I am attempting to have all my registrations and Invites proxied to the the Freeswitch server which will do the call processing. My phone is told to point to the SIP Domain voip2.siptest.net.au http://voip2.siptest.net.au which resolves to OpenSIPS. This is a quick layout of the path: Yealink T20 - VOIP Phone - 10.2.0.2 (Private IP) | (NAT) Cisco 877 DSL Router (2.2.239.241) (Public IP) | INTERNET | OpenSIPS (1.1.108.70) (Public IP) | FreeSWITCH (1.1.108.68) (Public IP) | External VOIP Provider (Assume 2.2.239.X and 1.1.108.X are public ranges) I register to the OpenSIPS load balancer using domains in this test case I am using 6132...@voip2.siptest.net.au mailto:6132...@voip2.siptest.net.au where voip2.siptest.net.au http://voip2.siptest.net.au is pointed to 1.1.108.70. The registration appears normal on both the OpenSIPS and Freeswitch. The destination I am calling is via another SIP provider which is routed by Freeswitch on the external profile. The problem I am seeing is that it looks like when a RE-INVITE happens the ACK gets sent back to 2.2.239.241 instead of being relayed to 1.1.108.70, I can see the ACK from 2.2.239.241 but OpenSIPS then replies and sends the ACK message to 2.2.239.241 where it should be seeing that it needs to send it to 1.1.108.68. Freeswitch will then keep sending 200 OK to OpenSIPS and then hang the call up after 30 seconds as there has been no ACK received. If there is no RE-Invite then the calls seems to work fine. It only seems to be when a RE-Invite is sent by the Phone. (I have tried a Siemens Gigaset and get the same issue). If I register the phone directly to Freeswitch I don't seem to have these issues. I have seen this issue mentioned on mailing lists in the past and I have tried the suggestions but none seem to work for me. I have provided the following which may assist: opensips.cfg http://pastebin.com/hG2GUHWV SIP Trace http://pastebin.com/8fahFPj3 OpenSIPS debug http://pastebin.com/T9ULEeXt Hopefully someone out there might have some ideas. Any advice would be appreciated. Cheers, Ash. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] using dialplan to prefix dialed number for specific accounts
Hi, for some accounts identified by ANI I need to add a prefix e.g. 001234 before the dialed number. Initially I've been using the dialplan module with ANI being used as a key for lookup by dpid in dialplan table. But dpid is declared int so I cannot accommodate account numbers greater than 2147483647. And it doesn't look like I can ALTER dpid to bigint or something like that, without having to hack with the source code. Still I'd like to use the dialplan module for that. Any hints? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] CANCEL INVITE
Hi Piotr, This sounds familiar to the problem I experienced some time ago - make sure to check comments here: https://sourceforge.net/tracker/?func=detailatid=1086410aid=2940556group_id=232389 I haven't been able to replicate that setup to confirm that the attached patch works. You are welcome to try it though :) Note RFC states it clearly that if no response has been received from the UAS at all, we should not attempt to send a CANCEL there. But it seems that in your case you received some provisional response so the issue has to do with the order in which CANCEL is fired - exactly what the patch is intended to fix. On 05.04.2011 15:56, Piotr Sobolewski wrote: I'm having problem with specific gateway to which OpenSIPS sends INVITE and then another INVITE (CallForward on no Aswer). The problem is when after sending first INVITE to gateway (without getting final response), OpenSIPS hits failure route and then sends another INVITE (with different RURI) toward gateway before CANCEL is sent, so the gateway responds to second INVITE with 482 Request merged (and gateway does not attempt to make second connection). Is there a way to send CANCEL before sending second INVITE ? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] New module: registrant
Ovidiu, great news! This is exactly what I've been looking for. But I'm not sure, does timer_interval affect the distribution of registration load in time or it affects only re-register? In what units if the hash_size given, do you have any examples e.g. how big hash_size we need to distribute in time (a few seconds apart) 10, 100 and 1000 registrations? Thank you. On 11.03.2011 07:13, Ovidiu Sas wrote: Hello all, There is a new module available for opensips: registrant. This module allows opensips to register itself on a remote registrar server. For more info, check the README file: http://www.opensips.org/html/docs/modules/devel/registrant.html Regards, Ovidiu Sas -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] terminating early dialogs with BYE
Hi, RFC3261 paragraph 15 Terminating a Session says: When a BYE is received on a dialog, any session associated with that dialog SHOULD terminate. A UA MUST NOT send a BYE outside of a dialog. The caller's UA MAY send a BYE for either confirmed or early dialogs, and the callee's UA MAY send a BYE on confirmed dialogs, but MUST NOT send a BYE on early dialogs. However early dialog termination with BYE appears to be not supported in OpenSIPS. If there any known solution to that? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips+asterisk: signalling not working?
On 06.10.2010 15:30, Vallimamod ABDULLAH wrote: You are right: you should not mix record_route_preset() and record_route(). Try to replace record_route with record_route preset. And if it does not work, make a ngrep capture on your opensips server to see sip dialog between opensips and asterisk (command line: ngrep -qt -d ethX -W byline port 5060.) Btw, I encourage you to use a public ip on your server if you have the possibility: putting opensips behind nat is*bad* as everybody will tell you;-) Right, Stefano: make sure you have not added the opensips IP addresses or domain names already listed in alias core parameter to the domain table. If the address in RURI is considered local it does routing after strict. The RURI gets rewritten with the URI in the Route header and like in your case opensips relayes ACK to itself. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to change Contact header
David, Also if rewriting Contact with the opensips address is what you want to achieve, you should look no further than b2b_logic top hiding scenario: http://www.opensips.org/Resources/B2buaTutorial16#toc12 On 06.10.2010 14:15, Bogdan-Andrei Iancu wrote: Hi David, take when using such regexps as the contact hdr may have multiple syntaxes: Contact: sip:u...@domain Contact: sip:u...@domain;hdr_params Contact:sip:u...@domain;hdr_params Contact:sip:u...@domain;uri_params;hdr_params Contact: displaysip:u...@domain;hdr_params Contact: displaysip:u...@domain;hdr_params etc So having a regexp to match all case may be difficultbetter try to focus only on the domain part, like matching the @IP part, like (Contact: .*@)[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3}.[0-9]{1,3} Regards, Bogdan David Santiago wrote: Solved! Adding a \r did the thing... if ( subst('/^Contact:sip:([0-9]+)@(.*)$/Contact: sip:\...@new_ip_address_here\r/ig') ) { xlog(contact modified!); }; On Tue, Oct 5, 2010 at 6:34 PM, David Santiago david.santi...@almiralabs.commailto:david.santi...@almiralabs.com wrote: Hi all, I need to modify the host part of a contact header. I'm trying something like: if ( subst('/^Contact:sip:([0-9]+)@(.*)$/Contact: sip:\...@new_ip_address_here/ig') ) { xlog(contact modified!); }; but the resulting Contact header is wrong and cannot be processed. Having a look at the header with wireshark shows that the Contact Binding entry is missing the ending , but the Contact, URI or SIP contact address have the at the end :L May be this is not the right way to modify a Contact header... Thanks in advance, David -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS swallows BYEs
David, If you rewrite Contact header with OpenSIPS address it is expected that BYE won't go any further than OpenSIPs proxy. You should use the B2B top hiding scenario as I suggested in other email. On 06.10.2010 15:50, David Santiago wrote: Hi all, I have a running OpenSIPS installation that I'm using for testing purposes. The fact is that I'm forwarding requests from a voip provider to a jain slee server and everything is working fine (INVITEs, ACKs, RTP flow,...), except for the BYEs generated from the server side. They reach the OpenSIPs proxy and are not forwarded to the voip provider in order to finish the call. I'm not sure if I have to manually setup a route for this to happen, or if this behaviour is only available by using the B2BUA approach in OpenSIPS. Thanks a lot! David -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips+asterisk: signalling not working?
Stefan, Please try removing ip addr and domain of opensips from domains table. It is sufficient to have listen=ip and alias=domain lines in config. Domain module will learn the ip and domain from config automatically. On 06.10.2010 16:02, Stefano Sasso wrote: Hi Andrew, thank you for the reply. I'm a new opensips user, how can I check what you said? The domain and ip address of opensips server is listed in domains table, but I don't know how to see if it's in aliases. In opensips.cfg I don't have anything pointing to aliases except modparam(alias_db|auth_db|usrloc|uri_db, use_domain, 1) thanks -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips+asterisk: signalling not working?
On 06.10.2010 16:36, Stefano Sasso wrote: nothing happened. It still loops (ACKs and BYEs) Hm, I will have to check in detail what you wrote here. This ACK should reach the asterisk: U 2010/10/06 14:43:42.736777 192.168.6.130:5060 - 77.238.yy.zz:5060 ACK sip:77.238.yy.zz:5060;lr;ftag=931ba062;did=12c.0478d917 SIP/2.0. ... but then there is another ACK to itself. Are you doing NAT 77.238.yy.zz to 192.168.6.130 (opensips itself)? How do you reach the asterisk? I think it should have a mapped routable IP address to. About the correctness of your config, you may remove the record_route() from loose_route block which is marked with even if in most of the cases is useless.. comment. You only need this: # record routing if (!is_method(REGISTER|MESSAGE)) record_route_preset(77.238.xx.yy:5060); IP should be the same as in advertised_address setting. Also add force_rport() at the very top of the main route. Note 1: you do need the advertised_address setting. Note 2: after removing IPs from domain table you may need to replace if (!is_uri_host_local()) .. with equivalent check: if(!uri==myself) for outbound routing. At least it worked for me. Anyway the main question is how do you reach the asterisk. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips+asterisk: signalling not working?
On 06.10.2010 17:25, Stefano Sasso wrote: So I can resolve dnatting i.e. port 5061 to .131 and 5062 to .132 and having in load_balancer 77.238.xx.yy:5061 and 77.238.xx.yy:5062? Am I right? Yes, this should help. It seems that asterisk will append bindport to externip automatically now so correct IP will be advertised in Contact header: https://issues.asterisk.org/view.php?id=11858 So give it a try. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Asterisk Cluster Scenario
On 25.09.2010 20:45, Stefano Sasso wrote: In the OpenSIPS features we read load balancing with failover, but we could not find any useful and complete configuration example. Is OpenSIPS able to know if an asterisk server is UP or DOWN, or must we use a 3rd part tool, like mon? Anyone has experience in use OpenSIPS as SIP load balancer (not to work as real SIP proxy)? Any hint/advice for this part, or for the global setup? Stefano, I guess a lot of people here use OpenSIPS as a SIP load balancer. You may refer to the load balancer tutorial link suggested above. The load_balancer module is able to determine if an asterisk server is UP or DOWN by using gw probing (OPTIONS ping), as well as dispatcher module: http://www.opensips.org/html/docs/modules/1.6.x/load_balancer.html http://www.opensips.org/html/docs/modules/1.6.x/dispatcher.html However, since you say the call-center works both inbound and outbound you probably need to do not only dispatching of inbound calls to asterisk but also dynamic routing of some kind. In this case I recommend that you check the powerful drouting module: http://www.opensips.org/html/docs/modules/1.6.x/drouting.html It is able to accommodate both inbound and outbound calls routing. I can comment on its gateway probing implementation, though it's consistent with load_balancer. OpenSIPS will send an OPTIONS ping to each gateway each N seconds. If the gw doesn't respond to ping - mark it disabled; if it responds to the next ping successfully automatic reenabling kicks in. Also, if opensips fails to terminate the call through the gw - you can disable it from failure route. If the next ping succeeds the gw will be automatically re-enabled. I would only disable the gw from failure route on some response codes, that indicate a server error. 500 and 603 probably are the good candidates (but check your applications). Otherwise there is a good chance of false positive, if the number was misdialed or something. BTW You can still take the gateway out of service, which happens to respond to the OPTIONS ping, if you disable in manually via MI command (supported in all 3 modules). This will completely stop the probing hence automatic re-enabling will not occur. BTW2 Do you mean your asterisk servers are running behind NAT? If so how are they reached from outside? How do you send the calls in and out of your network? You may contact me with off the list if you need any further help or clarification. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] [SR-Users] [OT] Any utility to create cool SIP flows in HTML format?
Iñaki, You may want to check the siplogview (http://siplogview.sourceforge.net) tool available from here: http://sourceforge.net/scm/?type=cvsgroup_id=117322 It may not suit your needs out of the box, but at least you will be able to use it as a good starting point for customization. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Authentication On Failure Routes
On 14.09.2010 14:30, Ross Beer wrote: I am having a problem where gateways require authentication which works perfectly for standard calls, however when a gateway failure occurs the next gateway fails auth. I can see that the packet contains the Authentication header which fails as the security relate to the previous gateway. How can I get around this issue, do I just remove the header so that the auth request is requested once more from the next gateway or is there a way I can get OpensSIPS to add the auth header? That would be handy, AFAIK such feature was developed in kamailio 3.1: http://kamailio.org/docs/modules/stable/modules_k/uac.html#id2885303 -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How opensis can manage different ports.
On 18.09.2010 07:36, mayamatakeshi wrote: If you are dealing with more than 2 ports, this might get complicated. But in case of just 2 ports, you could check the ReceivedPort and set a bflag during handling of the REGISTER request: if($Rp == 5060) { setbflag(BFLAG_RECEIVED_ON_PORT_5060); } That's correct. It will also work with more than 2 ports, but you need to enumerate each socket with a flag. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] crash when calling get_source_group() with uninitialized db_url
Hello Bogdan, Thank you - it works fine: Aug 19 13:59:14 box01 /usr/local/sbin/opensips[6137]: INFO:permissions:init_address: db_url parameter of permissions module not set, disabling allow_address Aug 19 13:59:14 box01 /usr/local/sbin/opensips[6137]: ERROR:permissions:get_src_grp_fixup: get_source_group() needs db_url to be set! Aug 19 13:59:14 box01 /usr/local/sbin/opensips[6137]: ERROR:core:fix_actions: fixing failed (code=-1) at cfg line 125 Aug 19 13:59:14 box01 /usr/local/sbin/opensips[6137]: ERROR:core:main: failed to fix configuration with err code -1 On 18.08.2010 18:21, Bogdan-Andrei Iancu wrote: Hi Andrew, You are right - please check the trunk version where I made a fix (rev 7142) - if ok, I will do a backport. Regards, Bogdan -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] crash when calling get_source_group() with uninitialized db_url
Hello Bogdan, I've noticed that permissions module in 1.6.3 dies to core if get_source_group() is called and db_url was not initialized. This is not something would normally happen - but nevertheless... Core was generated by `/usr/local/sbin/opensips -P /var/run/opensips.pid'. Program terminated with signal 11, Segmentation fault. [New process 5847] #0 0x2addea1c05fb in get_source_group (msg=0x793700, pvar=0x78f840 N) at address.c:586 586 group = find_group_in_hash_table(*hash_table, (gdb) bt #0 0x2addea1c05fb in get_source_group (msg=0x793700, pvar=0x78f840 N) at address.c:586 #1 0x0040e818 in do_action (a=0x789ec8, msg=0x793700) at action.c:1040 #2 0x00411d15 in run_action_list (a=value optimized out, msg=0x793700) at action.c:139 #3 0x00410b7b in do_action (a=0x78aa40, msg=0x793700) at action.c:712 #4 0x00411d15 in run_action_list (a=value optimized out, msg=0x793700) at action.c:139 #5 0x00410105 in do_action (a=0x78abf0, msg=0x793700) at action.c:706 #6 0x00411d15 in run_action_list (a=value optimized out, msg=0x793700) at action.c:139 #7 0x00412067 in run_top_route (a=0x7851e8, msg=0x793700) at action.c:119 #8 0x00456a35 in receive_msg ( buf=0x758120 INVITE sip:7...@192.168.31.15:5060 SIP/2.0\r\nVia: SIP/2.0/UDP 192.168.31.67:5061;branch=z9hG4bK-7qwngnepdz4owliv;rport\r\nMax-Forwards: 69\r\nFrom: \bob|\ sip:11165410...@192.168.31.67;tag=uyjyhekcfkp3lyw..., len=959, rcv_info=0x7fffb5aabc90) at receive.c:162 #9 0x0049b604 in udp_rcv_loop () at udp_server.c:492 #10 0x0042a43d in main (argc=3, argv=value optimized out) at main.c:818 -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Upgrade error from 1.5.3 to 1.6.3 - table version mismatch
On 10.08.2010 23:21, Gavin Henry wrote: opensipsdbctl migrate opensips_1_5 opensips_1_6 was used and there was one error with the subscriber db which is actually a stored procedure calling an Asterisk realtime table. Is there a guide to migrate the SQL with out using the above tool? Gavin, I sometimes checkout two versions with svn and run diff - you are correct the structure of dialog table has changed in 1.6.3, not sure why opensipsdbctl migrate .. did not convert it: diff -ruN opensips_1_5/scripts/mysql/dialog-create.sql opensips_1_6/scripts/mysql/dialog-create.sql --- opensips_1_5/scripts/mysql/dialog-create.sql2010-08-04 17:11:48.0 +0300 +++ opensips_1_6/scripts/mysql/dialog-create.sql2010-08-04 17:11:48.0 +0300 @@ -1,4 +1,4 @@ -INSERT INTO version (table_name, table_version) values ('dialog','3'); +INSERT INTO version (table_name, table_version) values ('dialog','4'); CREATE TABLE dialog ( id INT(10) UNSIGNED AUTO_INCREMENT PRIMARY KEY NOT NULL, hash_entry INT(10) UNSIGNED NOT NULL, @@ -18,7 +18,10 @@ callee_sock CHAR(64) NOT NULL, state INT(10) UNSIGNED NOT NULL, start_time INT(10) UNSIGNED NOT NULL, -timeout INT(10) UNSIGNED NOT NULL +timeout INT(10) UNSIGNED NOT NULL, +vars TEXT(512) DEFAULT NULL, +profiles TEXT(512) DEFAULT NULL, +script_flags INT(10) UNSIGNED DEFAULT 0 NOT NULL ) ENGINE=MyISAM; -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS LiveDVD
This has been already reported today on the OpenSIPS-News mailing list. You should expect a fixed ISO soon. On 04.08.2010 18:14, Nedzad wrote: I downloaded OpensipsLiveDVD and extracted it and I installed successfully Vmware player 3.1.0 build-261024, on my OS Windows 7 Ultimate, 64-bit. But when I am going to FileOpen a Virtual MachineDebian5.vmxPlay virtual machine, I get following error: File not found: Debian 5-Snapshot1.vmsn This file is required to power on this virtual machine. If this file was moved, please provide its new location. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Location info when using OpenSIPS as outbound proxy
On 01.08.2010 20:37, Bogdan-Andrei Iancu wrote: 2) ideally, for an outbound proxy, you should do the registration processing at reply time, once the main registrar accepted the registration and eventually made all the changes over it. But right now opensips does not accept registration processing for replies. Just in case - some time ago I did something like: onreply_route[3] { # Here we handle REGISTER replies xlog(L_INFO, [$mi] [$rs $rr]\n); if (status=~200) { route(3); }; route[3] { # workaround for location saving xlog(L_INFO, saving location\n); save(location,0x02); } 0x02 - do not generate a SIP reply to the current REGISTER request. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] ACK/200 to wrong port
On 23.07.2010 18:16, Marcio Veloso Antunes wrote: Ok, So in that case all Cisco ATA 186 Version 3.1.1 must use 5060 as it's own SIP port. Is there a knowlodge base which this kind of information could be stored ? I was wondering about the same thing but interop with misc devices has been own trialerror experience for me so far.. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] ACK/200 to wrong port
Marcio, This looks like a bug in Cisco ATA186. Cisco A received 200 OK (packet no. 32): Record-Route: sip:200.198.184.198:5061;lr=on;ftag=3971837741 Contact: sip:10...@187.13.212.160:5081;user=phone;transport=udp And sent ACK sip:10...@187.13.212.160 SIP/2.0 Route: sip:200.198.184.198:5061;lr=on;ftag=3971837741 But it must put remote target URI learned from Contact to Request-URI: ACK sip:10...@187.13.212.160:5081 According to RFC 3261: The UAC uses the remote target and route set to build the Request-URI and Route header field of the request. If the route set is empty, the UAC MUST place the remote target URI into the Request-URI. The UAC MUST NOT add a Route header field to the request. If the route set is not empty, and the first URI in the route set contains the lr parameter (see Section 19.1.1), the UAC MUST place the remote target URI into the Request-URI and MUST include a Route header field containing the route set values in order, including all parameters. Hope this helps.. On 23.07.2010 08:25, Marcio Veloso Antunes wrote: Hi all, I am two days working over a mistery on an ACK from a 200 that was missing. I was using 2 cisco ATA 186 both behind NAT, and both on other ports than 5060 (one on 5080 and other on 5081). When calling eachother the ACK was sent from cisco A to OpenSIPS (1.6.2) which processed the ACK and then forwarded to Cisco B. The problem is on debug Opensips showed that it forwarded but cisco B never received. Today i found what was going on. OpenSips was sending to CISCO_B:5060 and not CISCO_B:5080 as expected. I am sending a pdf with every packet of this dialog. If you wish, i can send my config script too. Thanks in advance, -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] get_source_group not working
Wesley, Are you using the new style of calling this function: get_source_group($var(x)) ? On 19.07.2010 17:09, Wesley Volcov wrote: Hey Bogdan, I have this error too. I updated my opensips from https://opensips.svn.sourceforge.net/svnroot/opensips/branches/1.6, and after this update the opensips -c seems ok, but after some seconds opensips started it die: Jul 19 11:00:46 opensips /usr/local/sbin/opensips[32027]: INFO:core:handle_sigs: child process 32029 exited by a signal 11 Jul 19 11:00:46 opensips /usr/local/sbin/opensips[32027]: INFO:core:handle_sigs: core was generated Jul 19 11:00:46 opensips /usr/local/sbin/opensips[32027]: INFO:core:handle_sigs: terminating due to SIGCHLD Jul 19 11:00:46 opensips /usr/local/sbin/opensips[32032]: INFO:core:sig_usr: signal 15 received Jul 19 11:00:46 opensips /usr/local/sbin/opensips[32034]: INFO:core:sig_usr: signal 15 received Am I missing something? Cheers! -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] get_source_group not working
Wesley, May be. I can suggest three things here: 1) try to run svn update again, recompile opensips and see if the problem goes away 2) set debug=6 and send us the log before moment of crash 3) if you get a coredump file, get a bracktrace from it and post it here Hope this helps. On 19.07.2010 17:43, Wesley Volcov wrote: Andrew, I'm using like the documentations says: if ( get_source_group($var(group)) ) { # do something with $var(group) xlog(group is $var(group)\n); }; I have deleted this lines and the error continue... Can it be a bug version ? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Fwd: Re: get_source_group not working
For some reason email from Wesley does not appear in the list, forwarding. Original Message Subject:Re: [OpenSIPS-Users] get_source_group not working Date: Mon, 19 Jul 2010 11:54:20 -0300 From: Wesley Volcov wesleyvol...@gmail.com Reply-To: wesleyvol...@gmail.com To: Andrew Pogrebennyk andrew.pogreben...@portaone.com CC: OpenSIPS users mailling list users@lists.opensips.org Andrew, I'he downgrade my opensips to 1.6.2 version and the error gone! It just occur with the svn version. Follow my coredump: warning: exec file is newer than core file. Reading symbols from /lib/ld-linux.so.2...(no debugging symbols found)...done. Loaded symbols for /lib/ld-linux.so.2 Core was generated by `/usr/local/sbin/opensips -P /var/run/opensips.pid -m 1024 -u root -g root'. Program terminated with signal 11, Segmentation fault. #0 0x080a2b10 in check_ip_address (ip=0xbfb5fc78, name=0x0, port=0, proto=0, resolver=0) at resolve.c:100 100 if ((ip-af==AF_INET6) #0 0x080a2b10 in check_ip_address (ip=0xbfb5fc78, name=0x0, port=0, proto=0, resolver=0) at resolve.c:100 he = value optimized out __FUNCTION__ = check_ip_address #1 0x004a461e in ?? () No symbol table info available. #2 0xbfb5fc78 in ?? () No symbol table info available. #3 0x in ?? () No symbol table info available. In this core, I had deleted the get_source_group function, but it crashed with check_source_address function. When I downgraded my opensips this error did not happen again. Cheers On 19 July 2010 11:49, Andrew Pogrebennyk andrew.pogreben...@portaone.com mailto:andrew.pogreben...@portaone.com wrote: Wesley, May be. I can suggest three things here: 1) try to run svn update again, recompile opensips and see if the problem goes away 2) set debug=6 and send us the log before moment of crash 3) if you get a coredump file, get a bracktrace from it and post it here Hope this helps. On 19.07.2010 17:43, Wesley Volcov wrote: Andrew, I'm using like the documentations says: if ( get_source_group($var(group)) ) { # do something with $var(group) xlog(group is $var(group)\n); }; I have deleted this lines and the error continue... Can it be a bug version ? -- Sincerely, Andrew Pogrebennyk -- Wesley Volcov Email: wesleyvol...@gmail.com mailto:wesleyvol...@gmail.com Messenger: vol...@live.com mailto:vol...@live.com Mobile: +55 11 9989-5348 Website: http://volcov.blogspot.com ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] uri_db NOT FOUND
Lucas, uri_db has been merged with uri module as of 1.6.0. Please check this out: http://www.opensips.org/Resources/DocsMigration15to16 On 17.07.2010 08:57, Lucas Alvarez wrote: Hi, I was trying to run opensips but I'm not being able because the application can't find the module uri_db. The problem is that the module wasn't compiled, so I check in the modules directory of the untar package /usr/src/opensips-1.6.2-tls/modules and the module wasn't there. Where can I get the module? Thanks in advance. Lucas -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] issue with trusted addresses listed with grp=0
Hi Bogdan, Thank you! It works fine now. On 11.07.2010 12:19, Bogdan-Andrei Iancu wrote: Hi Andrew, Andrew Pogrebennyk wrote: Hello Bogdan, Thanks for getting back to me. Two things: 1) I have bumped param_no for function get_source_group from 0 to 1 to get it running. fixed, thanks 2) Now I am getting the invalid operation 20/3/4 error - maybe opensips doesn't assume automatically that the variable is int: Jul 9 17:33:29 dev01 /usr/local/sbin/opensips[2299]: CRITICAL:core:comp_scriptvar: invalid operation 20/3/4!! Jul 9 17:33:29 dev01 /usr/local/sbin/opensips[2299]: WARNING:core:do_action: error in expression (l=681) hopefully fixed also. I hope we can resolve this before the 1.6.3 release. definitely :) Please update and test again. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] issue with trusted addresses listed with grp=0
Hello Bogdan, Thanks for getting back to me. Two things: 1) I have bumped param_no for function get_source_group from 0 to 1 to get it running. 2) Now I am getting the invalid operation 20/3/4 error - maybe opensips doesn't assume automatically that the variable is int: Jul 9 17:33:29 dev01 /usr/local/sbin/opensips[2299]: CRITICAL:core:comp_scriptvar: invalid operation 20/3/4!! Jul 9 17:33:29 dev01 /usr/local/sbin/opensips[2299]: WARNING:core:do_action: error in expression (l=681) I hope we can resolve this before the 1.6.3 release. n 10.07.2010 00:51, Bogdan-Andrei Iancu wrote: Hi Andrew, Indeed there is design bug - a script function returning 0 will break the script execution ; so if the group found by get_source_group() is zero, when returning the val, the script execution will end I change the way you use the function - instead of using the return code for group, you provide a pvar to store the result: $var(group) = get_source_group(); = get_source_group($var(group)); This fix is available on trunk (see revision 7009) - please test it and if ok, I will backport it to 1.6 also. Regards, Bogdan -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] issue with trusted addresses listed with grp=0
Hello, I am using the address table to keep the trusted IPs like this: # check trusted IPs $var(group) = get_source_group(); if ($var(group) == 32) { # use grp=32 as trusted indication xlog(L_INFO, Call from trusted peer - M=$rm RURI=$ru F=$fu T=$tu IP=$si ID=$ci\n); # do something stupid } else { if(!proxy_authorize(, subscriber)) { proxy_challenge(, 0); exit; } # process the call } Something strange happens when the INVITE source IP comes from the address listed with grp=0 - the call seems to get stuck and never gets to digest challenge: Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:permissions:get_source_group: Looking for df59f95c, 55518 in address table Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:permissions:get_source_group: Found 0 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:next_state_dlg: unref dlg 0xb4c2c944 with 1 - 2 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:next_state_dlg: dialog 0xb4c2c944 changed from state 1 to state 5, due event 1 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:dlg_onreply: dialog 0xb4c2c944 failed (negative reply) Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:unref_dlg: unref dlg 0xb4c2c944 with 1 - 1 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:unref_dlg: unref dlg 0xb4c2c944 with 1 - 0 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:unref_dlg: ref =0 for dialog 0xb4c2c944 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:destroy_dlg: destroing dialog 0xb4c2c944 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:dialog:destroy_dlg: dlg expired or not in list - dlg 0xb4c2c944 [2039:572707183] with clid 'ZmNmMjkxZjlhZmU4MzU4ZThhOWJlNTZmYzY2YjM4NTQ.' and tags 'f9383040' 'NULL' Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:core:destroy_avp_list: destroying list 0xb4c28240 Jul 8 16:20:49 dev01 /usr/local/sbin/opensips[24667]: DBG:core:receive_msg: cleaning up Jul 8 16:20:50 dev01 /usr/local/sbin/opensips[24675]: DBG:tm:timer_routine: timer routine:2,tl=0xb4c2dda0 next=(nil), timeout=70 Jul 8 16:20:50 dev01 /usr/local/sbin/opensips[24675]: DBG:tm:wait_handler: removing 0xb4c2dd58 from table Jul 8 16:20:50 dev01 /usr/local/sbin/opensips[24675]: DBG:tm:delete_cell: delete transaction 0xb4c2dd58 Jul 8 16:20:50 dev01 /usr/local/sbin/opensips[24675]: DBG:tm:wait_handler: done It looks as if something was wrong with script variables operation because with any grp!=0 there is a comparison operation after the Found line: Jul 8 16:22:31 dev01 /usr/local/sbin/opensips[25047]: DBG:permissions:get_source_group: Found 1 Jul 8 16:22:31 dev01 /usr/local/sbin/opensips[25047]: DBG:core:comp_scriptvar: int 20 : 1 / 32 However I can give up an idea that I am doing something stupid, else with 0 being default value much more people would have noticed this problem before. Any ideas? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] radius accounting: multi-leg, zero session time
Hello, I couldn't get radius accounting to work properly lately: 1) How to enable multi-leg accounting in radius? Jun 13 07:15:34 sip01 /usr/local/sbin/opensips[21050]: rc_avpair_new: unknown attribute 0 Jun 13 07:15:34 sip01 /usr/local/sbin/opensips[21050]: ERROR:aaa_radius:rad_avp_add: failure Jun 13 07:15:34 sip01 /usr/local/sbin/opensips[21050]: ERROR:acc:acc_aaa_request: failed to add RAD_LEG_SRC, 23 What id and type should the RAD_LEG_SRC, RAD_LEG_DST attributes have? In config they are defined as: modparam(acc, multi_leg_info, RAD_LEG_SRC=$avp(i:901);RAD_LEG_DST=$avp(i:902) 2) How to make opensips pass session-time attribute to radius? Sun Jun 3 08:10:12 2010 : Error: [sql] stop packet with zero session length. [user '000...@192.168.0.157', nas '127.0.0.1'] In config I'm doing just acc_aaa_request(). I'm using OpenSIPS 1.6.2 and want to do rating with cdrool. Since these are basic things, perhaps that's me doing something stupid? -- Regards, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] radius accounting: multi-leg, zero session time
Hello Bogdan, On 18.06.2010 00:36, Bogdan-Andrei Iancu wrote: ERROR:acc:acc_aaa_request: failed to add RAD_LEG_SRC, 23 What id and type should the RAD_LEG_SRC, RAD_LEG_DST attributes have? In config they are defined as: modparam(acc, multi_leg_info, RAD_LEG_SRC=$avp(i:901);RAD_LEG_DST=$avp(i:902) Have you defined the RAD_LEG_SRC and RAD_LEG_DST RADIUS AVPs in the RADIUS dictionary? No. Basically I am asking what id and type should those AVPs have in the RADIUS dictionary. 2) How to make opensips pass session-time attribute to radius? Sun Jun 3 08:10:12 2010 : Error: [sql] stop packet with zero session length. [user '000...@192.168.0.157', nas '127.0.0.1'] In config I'm doing just acc_aaa_request(). I'm using OpenSIPS 1.6.2 and want to do rating with cdrool. Since these are basic things, perhaps that's me doing something stupid? OpenSIPS does not explicitly add the session-time AVP - is this an extra accounting val or ? Well, I was thinking that OpenSIPS should supply session-time so that cdrtool would know real call duration, shouldn't it? Or cdrtool can deduct call duration based on the difference of timestamps of Start and Stop packets? I'm not sure how this works. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] installing opensips on Fedora 10 64bits
On 31.03.2010 22:39, Madovsky wrote: i found on rpmfind.net the 0.9.10 packages (xinian) as it doesn't exist for Fedora anymore 0.9.10 may not work well with your environment (gcc version or something). Try to install fresh version such as 1.06.31 from source; it worked for me. However I couldn't get mi_xmlrpc to work with abyss library supplied with the module. That's just my experience. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] OpenXCAP on CentOS dies with segmentation fault
Hello, I'm trying to get OpenXCAP 1.1.2 to work on CentOS 5.3. I have installed python-2.5.5 and all dependencies listed on http://openxcap.org/wiki/Installation under /usr/local/ either from source or using pip. I have python-gnutls-1.2.0 installed. Now OpenXCAP crashes with segmentation fault on loading gnutls.library as I see in gdb backtrace. Full backtrace is attached. Is this some known issue? -- Sincerely, Andrew Pogrebennyk backtrace.txt.gz Description: application/gzip ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenXCAP on CentOS dies with segmentation fault
On 30.03.2010 15:20, Jesus Rodriguez wrote: Replace python-gnutls 1.2.0 by 1.1.8 version. Thanks. I have compiled python-gnutls-1.2.0 against gnutls-2.6.6 and libgcrypt-1.4.4, installed it, set LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH and ran ldconfig, so it appears to work now. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenXCAP on CentOS dies with segmentation fault
There is another problem I am working on now - some issue with MySQLdb, however dsn in config.ini is correct: # /usr/local/bin/openxcap --no-fork Starting OpenXCAP 1.1.2 xcap.server.HTTPFactory starting on 443 TLS started error: Traceback (most recent call last): error: File /usr/local/lib/python2.5/threading.py, line 446, in run error: self.__target(*self.__args, **self.__kwargs) error: File /usr/local/lib/python2.5/site-packages/twisted/python/threadpool.py, line 161, in _worker error: context.call(ctx, function, *args, **kwargs) error: File /usr/local/lib/python2.5/site-packages/twisted/python/context.py, line 59, in callWithContext error: return self.currentContext().callWithContext(ctx, func, *args, **kw) error: File /usr/local/lib/python2.5/site-packages/twisted/python/context.py, line 37, in callWithContext error: return func(*args,**kw) error: --- exception caught here --- error: File /usr/local/lib/python2.5/site-packages/twisted/internet/threads.py, line 24, in _putResultInDeferred error: result = f(*args, **kwargs) error: File /usr/local/lib/python2.5/site-packages/twisted/enterprise/adbapi.py, line 372, in _runInteraction error: conn = Connection(self) error: File /usr/local/lib/python2.5/site-packages/twisted/enterprise/adbapi.py, line 33, in __init__ error: self.reconnect() error: File /usr/local/lib/python2.5/site-packages/twisted/enterprise/adbapi.py, line 70, in reconnect error: self._connection = self._pool.connect() error: File /usr/local/lib/python2.5/site-packages/twisted/enterprise/adbapi.py, line 342, in connect error: conn = self.dbapi.connect(*self.connargs, **self.connkw) error: File build/bdist.linux-i686/egg/MySQLdb/__init__.py, line 81, in Connect error: File build/bdist.linux-i686/egg/MySQLdb/connections.py, line 188, in __init__ error: exceptions.TypeError: 'reconnect' is an invalid keyword argument for this function -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to save seceive ip in location?
On 30.03.2010 15:49, CheeWii wrote: Now I used Opensips as a sip sms gateway. I used save(location) to store the register information. However, when my client is behind NAT,opensips will relay MESSAGE to an private ip address. It just as 192.168.111.100. Just use fix_nated_register(): http://www.opensips.org/html/docs/modules/devel/nathelper.html#id272036 -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenXCAP on CentOS dies with segmentation fault
On 30.03.2010 16:37, Andrew Pogrebennyk wrote: error: exceptions.TypeError: 'reconnect' is an invalid keyword argument for this function Problem solved by switching to MySQL-python-1.2.1_p2. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] drouting question
On 26.03.2010 12:30, Bogdan-Andrei Iancu wrote: the module extracts the FROM uri from the original message, so it will not see your change on the from hdr. Interesting.. I will have to re-check it but I think that it worked for me in the past for a similar purpose: # validate based on the packet IP address $var(from) = sip: + $fU + @ + $si; uac_replace_from(,$var(from)); -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Presence and Linksys phones - not working 1.6.1
James, Are you using pua_dialoginfo to get device state? If so are you telling the dialog module to monitor the interesting dialogs and calling dialoginfo_set()? Note that once you get this working you will likely need this fix: http://sourceforge.net/tracker/?func=detailatid=1086412aid=2847397group_id=232389 On 31.03.2010 08:31, James Lamanna wrote: Sorry, I realized I had a configuration error on my phone, but the presence still does not work. The phone now subscribes to the event: dialog. Here are relevant parts of my opensips config: modparam(presence, server_address, sip:s...@xxx.xxx.xxx.xxx:5060) modparam(presence, expires_offset, 10) modparam(presence_xml, force_active, 1) modparam(presence_dialoginfo, force_single_dialog, 1) I have also verified that handle_subscribe() is being called when a SUBSCRIBE message comes in. Calling the phone doesn't seem to produce any PUBLISH messages or anything pertaining to presence. Thanks. -- James -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] creating priority groups with drouting
Hi, I'm looking for a way to setup two groups of gws to route the traffic to with drouting. OpenSIPS shall use gws in second group only if all gws within first group failed. So it shall go over the groups sequentially but within the group, it's random. Think of it like second group contains expensive e.g. EC2 gateways. But as you all know drouting does not do any rule fallback - once a rule is matched, it will use only the destinations from this rule and it will not try to re-match a different rule. Also I find it kind of limiting that the priority field in drouting is considered only when the time selection is used for overlapping rules. So it looks like I can't achieve this behavior with drouting in its current shape. Actually it would be sufficient for me that if all the gws in a rule (higher priority) fail then this rule (or its gws) be automatically included in a blacklist, so if you call again the next rule (minor priority) would be taken. Makes sense? Is there anything I could try? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Getting Error When Configuring OpenSIPS + FreeRadius`
On 26.02.2010 14:33, Ahmed Munir wrote: Feb 26 14:47:47 rose /usr/local/sbin/opensips[25988]: ERROR:aaa_radius:rad_avp_add: failure Feb 26 14:47:47 rose /usr/local/sbin/opensips[25988]: ERROR:acc:acc_aaa_request: failed to add Source-IP, 13 And I also check table radacct in mysql database, no records are inserted into it. I think this means an incorrect RADIUS dictionary. You should verify that the extra attributes you have defined are present there. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] [New] Probing for drouting module
Bogdan-Andrei Iancu wrote: and the mysql format is in scripts/mysql/drouting-create.sql That's good news! Is there any chance of backporting this to 1.6 branch? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] [New] Probing for drouting module
Bogdan, On 20.02.2010 00:41, Bogdan-Andrei Iancu wrote: It will be as soon as it will prove its stability in trunk... So tester are welcome :) OK cool. Regards, Bogdan PS: Andrew, have you tried the CANCEL patch? Not yet - will do this on Monday! Thank you. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Query regarding Rtp Proxy opensips
On 28.01.2010 11:07, Indiver wrote: I forgot to mention that files are not storing by callee or caller number. Moreover it is taking its own unique caller id. How to over come this in order to modify the recording file name as callee-caller and time stamp format. File names are created using template: ${callid}=${tag}.${direction}.${pstype}, where $direction = 'a' or 'o', $pstype = 'rtp' or 'rtcp'. See rtpp_record.c and functions ropen() and rwrite(). -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Query regarding Rtp Proxy opensips
On 29.01.2010 11:29, Bogdan-Andrei Iancu wrote: I doubt you can change that as RTPproxy is not decoding the RTP stream - as the name says, the tool is only RTP aware, so cannot interpret the content. But I guess you can google for some other audio tools to help mixing the 2 streams. Check this page: http://www.rtpproxy.org/wiki/RTPproxy/FAQ I didn't try the rtpbreak/sox approach though. We are using proprietary tool here. I think that sox doesn't decode the g723 and g729 codecs. One thing to keep in mind is that the RTP headers are not written in the platform-independent format, so expect they be decoded only on the same platform as the one that created the recording. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] BYE - 404 not here
On 04.02.2010 12:56, Max Mühlbronner wrote: But if the call is established and the callee hangs up, the BYE is not received by the original calling side so it stays connected. My opensips knowledge is still very basic, so please excuse if it is some dumb routing mistake made by me. I suppose you have a problem with routing in-dialog requests. Please attach your opensips.cfg. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] ping gateways in lcr or drouting?
Bogdan-Andrei Iancu wrote: None of them do support pinging to GW, but I guess it will be a nice feature for DR.. Bogdan, Another nice feature would be to add a pseudo-variable to do_routing() to take caller's URI from. LCR module already can do that. Probably I will come up with a patch for drouting. Thank you. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Bulding OpenSIPS v1.6.1-notls on 64 bit machine
On 02.02.2010 16:10, Steven C. Blair wrote: I'm trying to build a fresh copy of OpenSIPS v1.6.1-notls on a 64 bit RH system and experiencing some path issues when running opensipsdbctl create. It seems some scripts exist in /usr/local/lib64/opensips/ and some in /usr/local/src/apps/opensips-1.6.1-notls/scripts/. Is there a suggested way to resolve this different so the database build script will work? Steven, You should provide the exact console output that shows what happens. Personally I and many people here are using 1.6.1 on 64 bit machines. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to limit channel on bunch of called DIDs?
Bogdan-Andrei Iancu wrote: Hi Andrew, It will be a bit tricky (depending on your approach) as set_dlg_profile() does not accept variables for the name of the profile - so , you need to use a profile with values where the value is the name of the group. Bogdan, It already seems to work this way: first do avp_db_query(select grp from grp where username='$fU', $avp(s:group)); then use group name as uuid key in usr_preferences table to get the max number of allowed simultaneous calls per group; if it's still above the profile size, insert dialog into caller profile, where the value is the $avp(s:group). Thank you. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to limit channel on bunch of called DIDs?
I'm facing the same task now - limit the number of concurrent calls per group of accounts rather than a single number. I'm thinking of using the group module to organize numbers into groups with group module, then using get_user_group() to get group id and comparing the profile size with concurrent calls limit set for this group in usr_preferences table. I'd probably hack the get_user_group() function to return the group name instead of id for convenience reason, though. Bogdan-Andrei Iancu wrote: Hi, you do not need any loop - just set as key for profiling the DID number and add to that profile the calls related to that DID. Regards, Bogdan Johnson Pajayat wrote: Hi Bogdan, I was able to implement the channel limiting on one DID by using a variable instead of AVP and replacing all instances of $tU to $rU. Now, I want to limit the channels to a set of DIDs and I'm thinking of implementing a while loop and counter in order to achieve it. Is this an efficient way of doing the limiting on a set of DIDs? One problem I can think with the while loop and counter will be how to deduct those calls that were already hung up by the caller. Again, inputs will be greatly appreciated. Thank you very much. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] ping gateways in lcr or drouting?
Hi, Is there any feature in drouting or lcr module to stop selecting a failing gw for a particular amount of time? It would suffice to have a ping mechanism or alternatively, mark the gateway as defunct in the failure_route. I was thinking pinging mechanism is present at least in lcr, but right now I see it only in dispatcher module. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] multiple Via headers separated by comma
Josip Djuricic wrote: Transaction is not matched if request is sent with 2 or more multiline via headers and response is received with via header in one line separated by comma? Josip, This is absolutely legal if multiple values are combined in one line separated by comma. Ccheck RFC 3261 for multiple header field values combining. Section 7.3. [H4.2] also specifies that multiple header fields of the same field name whose value is a comma-separated list can be combined into one header field. That applies to SIP as well, but the specific rule is different because of the different grammars. Specifically, any SIP header whose grammar is of the form header = header-name HCOLON header-value *(COMMA header-value) allows for combining header fields of the same name into a comma- separated list. The Contact header field allows a comma-separated list unless the header field value is *. Response is matched to request using branch parameter from uppermost Via header, so I don't know why RFC compliant implementation would have problems with response matching when Via header is combined. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Avpops failure route
Hi, Perhaps someone could chime in on this.. Andrew Pogrebennyk wrote: Bogdan, You are correct. But the thing is that when fr_inv_timer hits, OpenSIPS (prematurely) sends INVITE on the next branch and only after that CANCELs the previous one. And if the gateway receives different branch on transaction to which no final reply has been sent yet - it can merge the requests. Let me know if you need the traces, but I've found this behavior to be consistent in the versions 1.3.2 - 1.5.3. Another thing I've found is that OpenSIPS resets the fr_timer in retransmission_handler() if no provisional response to INVITE has been received, or retransmission_handler() seems to affect the fr_timer somehow. Here we see that it forwards the INVITE and sets FR_TIMER as per script: -- Sincerely, Andrew Pogrebennyk PortaOne, Inc., QA Engineer andrew.pogreben...@portaone.com Tel: +1-866-SIP VOIP (+1 866 747 8647) ext. 7133 Meet us at ITEXPO East 2010 Miami Beach Convention Center January 21-22, Booth 424 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Avpops failure route
/opensips[16128]: DBG:tm:timer_routine: timer routine:0,tl=0x28701a34 next=0x2870199c, timeout=92 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:final_response_handler: stop retr. and send CANCEL (0x287017b8) Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_should_relay_response: T_code=183, new_code=408 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_pick_branch: picked branch 1, code 408 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:is_3263_failure: dns-failover test: branch=1, last_recv=408, flags=1 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_should_relay_response: trying DNS-based failover Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_check_status: checked status is 408 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:avpops:ops_pushto_avp: 1 avps were processed Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_check: start=0x287017b8 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:t_check: transaction already found! Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:pv_get_tm_reply_code: reply code is 408 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: forwarded on 408 to: sip:400[...skipped@a.b.c.d Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: 1 avps were removed Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:core:pv_get_dsturi: no destination URI Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: Request leaving server, D-URI='null' - M=INVITE RURI=sip:400[...skipped...] Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:core:check_via_address: params a.b.c.d, a.b.c.d, 0 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:set_timer: relative timeout is 50 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:set_timer: calling insert_timer_unsafe: timeout=50, final value=9250 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:insert_timer_unsafe: [4]: 0x28701b2c (9250) Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:_set_fr_retr: FR_TIMER = 20 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:set_timer: relative timeout is 20 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:set_timer: calling insert_timer_unsafe: timeout=20, final value=112 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:insert_timer_unsafe: [0]: 0x28701b48 (112) Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:relay_reply: branch=1, save=1, relay=-1 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:final_response_handler: done Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16128]: DBG:tm:timer_routine: timer routine:0,tl=0x2870199c next=0x0, timeout=92 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16127]: DBG:core:parse_msg: SIP Reply (status): Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16127]: DBG:core:parse_msg: version: SIP/2.0 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16127]: DBG:core:parse_msg: status: 100 Dec 6 22:22:32 sip2 /usr/local/sbin/opensips[16127]: DBG:core:parse_msg: reason: Trying Thank you. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Avpops failure route
Hello, I have a similar setup. It could be that you just need to call append_branch() somewhere in the failure_route[3], unless you do it in route[4]. At least it works for me. However I have faced another problem that you can't use the same gateway/UA several times as a target for sequential forwarding, since INVITEs to different targets are part of the same dialog, besides OpenSIPS sends new INVITE a few ms prior to canceling previous branch (well, I'm about to submit it as a bug). I thought you could have some clue about it since I see you are also using one IP address. I have written about it here: http://www.mail-archive.com/users@lists.opensips.org/msg07999.html but Bogdan didn't reply yet. Andrew Indiver wrote: hi bodgan, Thanks for your reply. I made some changes and call is now going to destination. But when no answer or busy it is not going to failure route. here are the changes i did. #unconditional call forward if(avp_db_load($ruri/username,$avp(s:callfwd))) { avp_pushto($ruri, $avp(s:callfwd)); avp_print(); route(4); exit; } #fwd on busy if (avp_db_load($ruri/username, $avp(s:fwdbusy))) { if (!avp_check($avp(s:fwdbusy), eq/$ruri/i)) { setflag(26); }; }; fwd on noanswer if (avp_db_load($ruri/username, $avp(s:fwdnoanswer))) { if (!avp_check($avp(s:fwdnoanswer), eq/$ruri/i)) { setflag(27); }; }; t_on_failure(3); #Failure Route# failure_route[3] { if (isflagset(26) t_check_status(486)) { if (avp_pushto($ruri,$avp(s:fwdbusy))) { avp_delete($avp(s:fwdbusy)); resetflag(26); route(4); exit; }; }; if (isflagset(27) t_check_status(408) t_check_status(487)) { if (avp_pushto($ruri, $avp(s:fwdnoanswer))) { avp_delete($avp(s:fwdnoanswer)); resetflag(27); route(4); exit; }; }; It does not going to failure route and just hanging up! ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Problem with prefix() during call forwarding
I have put b2bua between OpenSIPS and Cisco, but still no luck. The point here is that when fr_inv_timer hits, OpenSIPS prematurely sends INVITE per the next branch and only after that CANCELs the previous one. I don't think this is the correct behavior actually, and there was a similar issue mentioned in the Kamailio mailing list: http://lists.kamailio.org/pipermail/devel/2009-May/018982.html If I can make OpenSIPS to wait for the fist branch to be canceled, I will invent something in the b2bua to make it finally work. However I understand that in general serial forking to one destination won't work. Andrew Pogrebennyk wrote: Bogdan, Thanks. I'm using 1.5.3. I sort of got stuck with this serial forking scenario. I mean, OpenSIPS does what is supposed to do. The problems is the call needs to be sent to (and is originated by) the Cisco AS5300. When one destination fails OpenSIPS sends the call to the next destination, but since new INVITE is part of the same dialog for OpenSIPS is carries the same Call-ID and From tag the Cisco gets confused and sends 482 Loop Detected or 500 Internal Server Error. Perhaps delaying new INVITE by a few ms for the Cisco to invalidate the call state would have helped, but I can't find my way around this. Regards, Andrew -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Problem with prefix() during call forwarding
Bogdan, Thanks. I'm using 1.5.3. I sort of got stuck with this serial forking scenario. I mean, OpenSIPS does what is supposed to do. The problems is the call needs to be sent to (and is originated by) the Cisco AS5300. When one destination fails OpenSIPS sends the call to the next destination, but since new INVITE is part of the same dialog for OpenSIPS is carries the same Call-ID and From tag the Cisco gets confused and sends 482 Loop Detected or 500 Internal Server Error. Perhaps delaying new INVITE by a few ms for the Cisco to invalidate the call state would have helped, but I can't find my way around this. Regards, Andrew Bogdan-Andrei Iancu wrote: Hi Andrew, Noticed you fixed the problem, but here are some ideas/questions: 1) what version on opensips do you use? 2) keep in mind that all the changes you do before creating the transaction (which is typically done by the first t_relay()) are inherited by all the following branched (you create via failure route). If you want to do changes to affect only a specific branch, you should use the onbranch route (see http://www.opensips.org/Resources/DocsCoreRoutes#toc2) Regards, Bogdan ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Problem with prefix() during call forwarding
Hi, I'm trying to put together some configuration for unconditional call forwarding. The carrier requires me to send the call with prefix 400 in R-URI. Here are the relevant routes: route[6] { if(avp_db_load($ru/username,$avp(s:callfwd))) { avp_pushto($ru, $avp(s:callfwd)); xlog(L_INFO, forwarded to: $avp(s:callfwd)); avp_delete($avp(s:callfwd)); $avp(i:25) = 20; } route(7); } route[7] { prefix(400); rewritehost(XX.YY.ZZ.WW); t_on_failure(2); xlog(L_INFO, Request leaving server, D-URI='$du' - M=$rm RURI=$ru F=$fu T=$tu IP=$si ID=$ci\n); t_relay(XX.YY.ZZ.WW:5060); exit; } failure_route[2] { # forward on busy if(t_check_status((486)|(408)) avp_pushto($ru, $avp(s:callfwd))) { append_branch(); xlog(forwarded on $T_reply_code to: $avp(s:callfwd)); avp_pushto($du, $avp(s:callfwd)); avp_delete($avp(s:callfwd)); $avp(i:25) = 20; route(7); } } The problem is that while the call is sent to the first call forward destination correctly (with prefix 400 in R-URI), it goes to the next destination (triggered from failure_route) without the prefix in R-URI! There are the following messages in the log file: Nov 12 21:12:00 sip2 /usr/local/sbin/openser[52971]: forwarded on 408 to: sip:89151793...@xx.yy.zz.ww Nov 12 21:12:00 sip2 /usr/local/sbin/openser[52971]: 1 avps were removed Nov 12 21:12:00 sip2 /usr/local/sbin/openser[52971]: DBG:core:pv_get_dsturi: no destination URI Nov 12 21:12:00 sip2 /usr/local/sbin/openser[52971]: Request leaving server, D-URI='null' - M=INVITE RURI=sip:40089151793...@xx.yy.zz.ww F=sip:84957978...@xx.yy.zz.ww T=sip:4953801...@aa.bb.cc.dd IP=XX.YY.ZZ.WW id=470bed43-cece11de-b158f4a9-da974...@xx.yy.zz.ww Despite R-URI appears with 400, it is sent without the prefix as I've confirmed by the trace. That no destination URI line looked suspicious to me, in fact I would expect that prefix() handles destination URI as well. I thought that could be the case so I've added explicit $du = $ru; after prefix and rewritehost. D-URI looks fine now, but it is still sent on the wire without 400: ov 12 21:59:29 sip2 /usr/local/sbin/openser[53183]: forwarded on 408 to: sip:89165438...@xx.yy.zz.ww Nov 12 21:59:29 sip2 /usr/local/sbin/openser[53183]: DBG:avpops:ops_pushto_avp: 1 avps were processed Nov 12 21:59:29 sip2 /usr/local/sbin/openser[53183]: 1 avps were removed Nov 12 21:59:29 sip2 /usr/local/sbin/openser[53183]: Request leaving server, D-URI='sip:40089165438...@xx.yy.zz.ww' - M=INVITE RURI=sip:40089165438...@xx.yy.zz.ww F=sip:84957978...@xx.yy.zz.ww T=sip:4953801...@aa.bb.cc.dd IP=XX.YY.ZZ.WW id=e9c3fa8b-ced411de-b59ef4a9-da974...@xx.yy.zz.ww What is the problem? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Problem with prefix() during call forwarding
Andrew Pogrebennyk wrote: Nov 12 21:59:29 sip2 /usr/local/sbin/openser[53183]: Request leaving server, D-URI='sip:40089165438...@xx.yy.zz.ww' - M=INVITE RURI=sip:40089165438...@xx.yy.zz.ww F=sip:84957978...@xx.yy.zz.ww T=sip:4953801...@aa.bb.cc.dd IP=XX.YY.ZZ.WW id=e9c3fa8b-ced411de-b59ef4a9-da974...@xx.yy.zz.ww Here is the relevant piece from ngrep: 21:59:29.328353 IP (tos 0x10, ttl 64, id 20061, offset 0, flags [none], proto UDP (17), length 1280) AA.BB.CC.DD.5060 XX.YY.ZZ.WW .5060: SIP, length: 1252 INVITE sip:89165438...@xx.yy.zz.ww SIP/2.0 [...] -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Problem with prefix() during call forwarding
Andrew Pogrebennyk wrote: failure_route[2] { # forward on busy if(t_check_status((486)|(408)) avp_pushto($ru, $avp(s:callfwd))) { append_branch(); I have fixed my problem, of course I had to use the append_branch() after prefix() in the script. xlog(forwarded on $T_reply_code to: $avp(s:callfwd)); avp_pushto($du, $avp(s:callfwd)); This $du modification was actually an artifact left from one of my tests, I didn't really mean to modify the destination URI there :) Thanks. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] drouting questions: tech-prefixes, SRV lookup
Hi, I'm going to try switching from lcr to drouting so I've got two questions. Firstly, I'm a bit puzzled with what to do with tech prefixes. I know I can't define a prefix which contains special characters as # or *. I understand why it is done so, but in real world the numeric part of the tech prefix happens to overlap with some destination and sometimes this is beyond your control. I'm thinking of some workaround, such as using s.substr or s.select and putting the result in ruri_avp. Anyone tried to put it to work? Another thing, I've noticed that in the dr_gateways table 'address' column is varchar. Could this be used to put the DNS/SRV domain name there for SRV lookup? As far as I can see the drouting does not cache the IP addresses for destination at startup so I will get the domain name in the R-URI and OpenSIPS will be perfectly able to resolve it, but anyway I'd like to confirm this. Thanks. -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] drouting and regular expression
Airton Kuada wrote: I User's module lcr and I'm testing the module drouting to perform the migration. However, I am trying to use regular expressions ( '^ 413,312 [45]') in the column prefix of dr_rules table and is failing. Can I use regular expression in this column? If i can not, what better way to write the route expression? Hello, I was reading through the drouting threads when I spotted your unanswered question. Drouting does not support regexps (and only digits are allowed) as all the routing rules are kept in a tree (by prefix) to enhance the searching and reduce the time to find the matching rule. Couldn't you reword your routing rules without the regexps? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] New Release: OpensSIPS-CP 2.0
Iulia Bublea wrote: Hello, The release 3.0 for OpenSIPS Control Panel comes to support the lastest release of OpenSIPS Server, OpenSIPS 1.6. [..skipped..] Thanks for the good news. But I am confused, the email subject reads 2.0 and the website has both. So is it 2.0 or 3.0? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] delay 200 OK in opensips for pre-configured interval
Hi, I am looking for the ability to send media backward before a call is established, allowing the remote side to send some DTMF before a call is established. For that purpose, is it possible to delay 200 OK on OpenSIPS for a pre-configured interval? So far I have considered the b2bua module, embedding the execution of Perl function with perl.so and some alternatives like SEMS and its ann_b2b application from iptel or sippy b2bua but found nothing that allow me do this without extra considerable development effort. Sure it would be better and more correct to implement this logic on the remote side, but for some reason this is not an option. Any ideas? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] limiting the number of gateways tried in lcr module
Hello, How do I limit the number of gateways tried in lcr module? Say, try only 2 (or 3, 4) gateways even if there are 5 matching gateways in the database? I can't do that by using the script variable because failure route is reenterable and variable doesn't keep its value between entering. I also couldn't achieve that by using AVPs. Any ideas? -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] limiting the number of gateways tried in lcr module
Brett Nemeroff wrote: You should be able to track this with avps. What version are you running? I know that in at least the newer versions of opensips, this is definitely supported:http://www.opensips.org/Resources/DocsCoreVar14 The AVPS will be visible in all routes where any message (reply or request) of the transaction will be processed - branch_route , failure_route, onreply_route (for this last route you need to enable the TM parameter * onreply_avp_mode*). Thanks. I don't have the config or the logs at hand at this moment but I tried something like this yesterday: failure_route[1] { if (!isflagset(31)) { $avp(i:500) = 1; # Initialize counter of failed attempts setflag(31); } else { $avp(i:500) = $avp(i:500) + 1; # OR: # avp_op($avp(i:500), add/1); } if ($avp(i:500) 2) { t_reply(503, Couldn't complete the call); exit; } if (next_gw()) { t_on_failure(1); t_relay(); } else { t_reply(503, No gateways); exit; }; } As a result of incrementing AVP variable, no matter which way, the OpenSIPS sends out some junk and then crashes... I'm using 1.4.4. -- Sincerely, Andrew Pogrebennyk PortaOne, Inc., QA Engineer andrew.pogreben...@portaone.com Tel: +1-866-SIP VOIP (+1 866 747 8647) ext. 7133 Meet us on June 1-3 at ITW, Booth 802 Marriott Wardman Park Hotel, Washington, DC http://www.internationaltelecomsweek.com/ ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] limiting the number of gateways tried in lcr module
Thanks for your input, Brett. I tried to initialize the counter from the main route block, but it didn't work - somehow I ended up with rURI in $avp(i:500). Then I rewrote the failure_route exactly like in the example I have provided earlier - and it worked like a magic. Previously I've had the same pieces of logic, just written in a different way, so I don't know what was the problem with it. Anyway I'm happy with the way it works now. In case anyone wonders, here is my working failure_route: failure_route[1] { # - # Failover to next gateway if any # - if (!isflagset(31)) { $avp(i:500) = 1; # Initialize counter of failed attempts setflag(31); } else { $avp(i:500) = $avp(i:500) + 1; # OR: # avp_op($avp(i:500), add/1); } xlog(L_INFO, - Made $avp(i:500) failed attempts\n); if (t_check_status(403)) { xlog(L_INFO, - Got 403 response; no more gateways will be tried\n); t_reply(403, Forbidden); exit; } if ($avp(i:500) = 2) { # limit on the search depth is set here xlog(L_INFO, - Didn't get positive response from two gateways; giving up\n); t_reply(503, Couldn't complete the call); exit; } if (next_gw()) { t_on_failure(1); t_relay(); } else { t_reply(503, No gateways); exit; }; } Brett Nemeroff wrote: I don't entirely remember the way the arming of the flags works and how they persist across failure_routes. Maybe someone can comment on that? Instead of using flags, call: $avp(i:500) = 1; # Initialize counter of failed attempts from your main route block, before failure route is called. Let us know how that works for you.. -Brett -- Sincerely, Andrew Pogrebennyk ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users