From tony at everestbn.com Tue Dec 1 07:27:38 2009 From: tony at everestbn.com (tony at everestbn.com) Date: Tue, 1 Dec 2009 10:27:38 -0500 (EST) Subject: [RTPproxy Users] Same IP\port for caller and callee Message-ID: <54392.216.236.133.150.1259681258.squirrel@mail.everestbn.com> Hello, I'm having random audio issues (<10% of the time) with blind call transfers on a Polycom phone. According to RTPPROXY's logs, it seems it is substituting the same IP and port for the caller's and callee's address causing one end to hear their own voice while the other end hears nothing. INFO:rxmit_packets: caller's address filled in: a.b.c.199:17856 (RTP) INFO:rxmit_packets: guessing RTCP port for caller to be 17857 INFO:rxmit_packets: callee's address filled in: a.b.c.199:17856 (RTP) INFO:rxmit_packets: guessing RTCP port for callee to be 17857 The SDP in the SIP messages has all the correct IPs and ports and it appears from a SIP overview everything should work fine. Here is some background information: I am using Kamailio 1.5.3 with RTPPROXY 1.2.1. SIP PATH is as follows: PSTN Gateway---Kamailio--SIP Proxy Server--Polycom Phone RTP PATH: PSTN Gateway---Kamailio----Polycom Phone *There are nor firewalls or filters. Call Flow: I place a call from the gateway to the ip phone.? Call is answered. When I do a blind transfer from the Polycom, the SIP Proxy sends an invite with no SDP, the gateway responds with a 200 with SDP and the SIP Proxy then sends an ACK with SDP.? As I mentioned, this works most of the time and the SIP messages look identical for when the transfer works and when it doesn't. When it doesn't work, the gateway end hears their own voice while the recipient of the Polycom IP phone transfer (usually another phone on the same proxy) hears nothing. Here is a portion of my configuration file: ..... if (is_method("INVITE")) { ? if((search("^Content-Type:[ ]*application/sdp")) || (search("^Content-Type:application/sdp"))){ ???? rtpproxy_offer("fcr"); ????????? setflag(12); ???? } ??? } ??? ??? if (is_method("ACK")) { ??? ? if((search("^Content-Type:[ ]*application/sdp")) || (search("^Content-Type:application/sdp"))) { ???????? ???? rtpproxy_answer("fcr"); ????????? ????? } ??? } ??????? ?? t_on_reply("1"); .... } onreply_route[1] { ? if (status=~"(180)|(183)|(2[0-9][0-9])"){ ?? if((search("^Content-Type:[ ]*application/sdp")) || (search("^Content-Type:application/sdp"))) { ????? if (isflagset(12)) { ?????? rtpproxy_answer("fcr"); ??????? } else { ????? rtpproxy_offer("fcrl"); ???????? }??? } ?} } Any help would be greatly appreciated! Thanks, Tony From sobomax at sippysoft.com Tue Dec 1 12:32:38 2009 From: sobomax at sippysoft.com (Maxim Sobolev) Date: Tue, 01 Dec 2009 12:32:38 -0800 Subject: [RTPproxy Users] Same IP\port for caller and callee In-Reply-To: <54392.216.236.133.150.1259681258.squirrel@mail.everestbn.com> References: <54392.216.236.133.150.1259681258.squirrel@mail.everestbn.com> Message-ID: <4B157D66.2050506@sippysoft.com> Tony, It would really help if you can attach full SIP trace for the problematic calls as well as full output of RTPproxy. You can mask IPs and other information that might be sensitive (just make sure to use unique placeholders for unique IPs), or send to me directly. tony at everestbn.com wrote: > > Hello, > > I'm having random audio issues (<10% of the time) > with blind > call transfers on a Polycom phone. According to RTPPROXY's > logs, > it seems it is substituting the same IP and port for the > caller's > and callee's address causing one end to hear their own > voice > while the other end hears nothing. Regards, -- Maksym Sobolyev Sippy Software, Inc. Internet Telephony (VoIP) Experts T/F: +1-646-651-1110 Web: http://www.sippysoft.com MSN: sales at sippysoft.com Skype: SippySoft From armangungor at hotmail.com Thu Dec 3 13:26:50 2009 From: armangungor at hotmail.com (Arman Gungor) Date: Thu, 3 Dec 2009 21:26:50 +0000 Subject: [RTPproxy Users] enabling timeout notification? Message-ID: Hi, I have a problem with enabling notification on timeout. I followed steps in Daniel Pocock's mail: ------------------------------------------------------------------------------------------ Here are some notes on how to observe the new notification behaviour: Start a timeout listener in one window: $ socat UNIX-LISTEN:/var/run/rtpproxy-test.timeout.sock STDOUT Start rtpproxy in another window: $ /tmp/rtpproxy -s unix:/var/run/rtpproxy-test.sock -n unix:/var/run/rtpproxy-test.timeout.sock -f In a third window, send a test command to rtpproxy: $ socat STDIO UNIX-CONNECT:/var/run/rtpproxy-test.sock Ua test01 192.168.1.100 1001 fromtag01 totag01 ------------------------------------------------------------------------------------------ However, nothing happens after a timeout. Is there anything else to do, in order to enable timeout notification? If we have to include something like "send timeout notification for this session", what is the format for it? _________________________________________________________________ Windows Live Hotmail: Your friends can get your Facebook updates, right from Hotmail?. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_4:092009 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.rtpproxy.org/pipermail/users/attachments/20091203/4e0f79c7/attachment.htm From sobomax at sippysoft.com Thu Dec 3 14:21:20 2009 From: sobomax at sippysoft.com (Maxim Sobolev) Date: Thu, 03 Dec 2009 14:21:20 -0800 Subject: [RTPproxy Users] enabling timeout notification? In-Reply-To: References: Message-ID: <4B1839E0.9090300@sippysoft.com> Arman Gungor wrote: > Hi, > > I have a problem with enabling notification on timeout. I followed steps in Daniel Pocock's mail: > > ------------------------------------------------------------------------------------------ > Here are some notes on how to observe the new notification behaviour: > > > Start a timeout listener in one window: > > $ socat UNIX-LISTEN:/var/run/rtpproxy-test.timeout.sock STDOUT > > > Start rtpproxy in another window: > > $ /tmp/rtpproxy -s unix:/var/run/rtpproxy-test.sock -n > > unix:/var/run/rtpproxy-test.timeout.sock -f > > > > In a third window, send a test command to rtpproxy: > > $ socat STDIO UNIX-CONNECT:/var/run/rtpproxy-test.sock > Ua test01 192.168.1.100 1001 fromtag01 totag01 > > > ------------------------------------------------------------------------------------------ > > However, nothing happens after a timeout. Is there anything else to do, in order to enable timeout notification? If we have to include something like "send timeout notification for this session", what is the format for it? Yes, it has to be requested by the signaling component specifically by passing the socket name and optional argument (i.e. session id) after all parameters (i.e. after totag). Check this file for example: http://cvs.berlios.de/cgi-bin/viewcvs.cgi/sippy/sippy/Rtp_proxy_session.py?view=markup def update_caller(self, remote_ip, remote_port, result_callback, options = '', index = 0, *callback_parameters): command = 'U' self.max_index = max(self.max_index, index) if self.rtp_proxy_client.sbind_supported and self.caller_raddress != None: if self.rtp_proxy_client.is_local: options += 'L%s' % self.global_config['_sip_tm'].l4r.getServer( \ self.caller_raddress).laddress[0] else: options += 'R%s' % self.caller_raddress[0] command += options command += ' %s %s %d %s %s' % ('%s-%d' % (self.call_id, index), remote_ip, remote_port, self.from_tag, self.to_tag) if self.notify_socket != None and index == 0 and \ self.rtp_proxy_client.tnot_supported: command += ' %s %s' % (self.notify_socket, self.notify_tag) self.rtp_proxy_client.send_command(command, self.update_result, (result_callback, 'caller', callback_parameters)) Hope it helps. Regards, -- Maksym Sobolyev Sippy Software, Inc. Internet Telephony (VoIP) Experts T/F: +1-646-651-1110 Web: http://www.sippysoft.com MSN: sales at sippysoft.com Skype: SippySoft From klaus.mailinglists at pernau.at Fri Dec 11 00:18:34 2009 From: klaus.mailinglists at pernau.at (Klaus Darilion) Date: Fri, 11 Dec 2009 09:18:34 +0100 Subject: [RTPproxy Users] correct meaning of force_rtp_proxy "E" and "I" flags in bridge mode In-Reply-To: <4B05414B.8000804@sippysoft.com> References: <4AF19C2C.7010800@pernau.at> <4B05414B.8000804@sippysoft.com> Message-ID: <4B22005A.9090904@pernau.at> Maxim Sobolev schrieb: > Klaus Darilion wrote: >> Hi! >> >> When rtpproxy is in bridging mode, and the session is created (INVITE) >> with "I" flag (internal), what command should be specified during lookup >> (200 ok response) to select the other interface? >> >> Do I have to use the 'E' flag to select the other interface, or do I >> have to specify again the 'I' flag and rtpproxy automatically choose the >> other interface (as it is a lookup)? > > Klaus, > > Sorry for the delay, somehow I missed this message. Basically, the flag > indicates the direction in which the current message is going. So for > the 200 OK that goes from internal to external it will have to be IE. So I allways have to specify 2 flags? E.g. from internal to internal: II and from external to external: EE ? > Please note that this older version of bridging logic is outdated, I > have much better version in place in trunk, which doesn't require any > flags from the nathelper, but instead uses host in RURI for INVITE and > host in 2nd Via in 200 OK to determine a proper media address. Great. Do you consider trunk ready to use? Does it work with current Kamailio nathelper module? regards klaus From klaus.mailinglists at pernau.at Fri Dec 11 00:34:41 2009 From: klaus.mailinglists at pernau.at (Klaus Darilion) Date: Fri, 11 Dec 2009 09:34:41 +0100 Subject: [RTPproxy Users] correct meaning of force_rtp_proxy "E" and "I" flags in bridge mode In-Reply-To: <4B05414B.8000804@sippysoft.com> References: <4AF19C2C.7010800@pernau.at> <4B05414B.8000804@sippysoft.com> Message-ID: <4B220421.5080303@pernau.at> Maxim Sobolev schrieb: > Please note that this older version of bridging logic is outdated, I > have much better version in place in trunk, which doesn't require any > flags from the nathelper, but instead uses host in RURI for INVITE and > host in 2nd Via in 200 OK to determine a proper media address. Hi Maxim! I wonder what is the logic behind this automatism - is it always possible to detect the proper interface just based on RURI/Via? Is it still possible to override it specify the interface manually? thanks klaus From sobomax at sippysoft.com Sat Dec 12 01:37:14 2009 From: sobomax at sippysoft.com (Maxim Sobolev) Date: Sat, 12 Dec 2009 01:37:14 -0800 Subject: [RTPproxy Users] correct meaning of force_rtp_proxy "E" and "I" flags in bridge mode In-Reply-To: <4B220421.5080303@pernau.at> References: <4AF19C2C.7010800@pernau.at> <4B05414B.8000804@sippysoft.com> <4B220421.5080303@pernau.at> Message-ID: <4B23644A.9020600@sippysoft.com> Klaus Darilion wrote: > > Maxim Sobolev schrieb: >> Please note that this older version of bridging logic is outdated, I >> have much better version in place in trunk, which doesn't require any >> flags from the nathelper, but instead uses host in RURI for INVITE and >> host in 2nd Via in 200 OK to determine a proper media address. > > Hi Maxim! > > I wonder what is the logic behind this automatism - is it always > possible to detect the proper interface just based on RURI/Via? > > Is it still possible to override it specify the interface manually? Yes, you can still override it by specifying exact IP that you want to use in ser.cfg. The goal is to make life of the "Average Joe" who installs package using default config easier. And it's better on system with many IPs too. Regards, -- Maksym Sobolyev Sippy Software, Inc. Internet Telephony (VoIP) Experts T/F: +1-646-651-1110 Web: http://www.sippysoft.com MSN: sales at sippysoft.com Skype: SippySoft From vragukumar at signalogic.com Fri Dec 18 13:29:52 2009 From: vragukumar at signalogic.com (Vikram Ragukumar) Date: Fri, 18 Dec 2009 15:29:52 -0600 Subject: [RTPproxy Users] Transparent bridge mode using rtpproxy Message-ID: <4B2BF450.50509@signalogic.com> Hello All, Below is a sketch of a setup that im trying to get operational. _____ ______ eth0 _____ eth1 ______ |_____|----|______|-----|_____|--------|______| Internet DSL Modem Server1 Server2 2 NIC's 1 NIC (Public IP) Server1 - Runs Kamailio and rtpproxy. It has 2 NIC's installed. Server2 - Runs Asterisk. It must be assigned a Public IP. I need to use rtpproxy to intercept data being sent to Server 2, process it and let it continue along it's original path. Are there any references you can point me to, that show how to use rtpproxy to achieve this bridging? Does the connection between eth1 of Server1 and eth0 of Server2 have to made using a crossover cable ? What needs to be done to make Server2 visible over the internet ? Thank you for all your help. Regards, Vikram.