Re: [Sip-implementors] is the JAIN SIP API's can listen any sip port given by user?

2008-11-06 Thread parasuraman kumarasami
Hi Ranga, i am Unable to find the src/examples in the given link. Can u send me exact link where the src is located? Regard's Parasuraman On Thu, Nov 6, 2008 at 7:09 PM, M. Ranganathan <[EMAIL PROTECTED]> wrote: > On Thu, Nov 6, 2008 at 7:16 AM, parasuraman kumarasami > <[EMAIL PROTECTED]> w

Re: [Sip-implementors] UPDATE RFC and dialog state inconsistency?

2008-11-06 Thread Victor Pascual Ávila
On Wed, Nov 5, 2008 at 6:03 PM, Robert Sparks <[EMAIL PROTECTED]> wrote: > You are correct. > > I think 3311 was talking about "early" vs "confirmed" when it said "state > of the dialog", but clearly the text > could be improved to avoid conflation with "dialog state" (meaning call-id, > cseqs, ta

Re: [Sip-implementors] [Sip] Route header in NOTIFY

2008-11-06 Thread DRAGE, Keith (Keith)
I am unsure where you see a problem. RFC 3265 clearly provides for the Route header to appear in the NOTIFY request if you need to do this, and has text explicitly talking about the record route mechanism working as normal in SUBSCRIBE dialogs. regards Keith __

Re: [Sip-implementors] [Sip] Route header in NOTIFY

2008-11-06 Thread Iñaki Baz Castillo
Hi, please, don't send the same mail to various maillists (AKA cross-posting). 2008/11/6 <[EMAIL PROTECTED]>: > > Hi, > > If SUBSCRIBE request and response has record route header, then should > NOTIFY messages need to have ROUTE header to follow the same route as > created in the dialog for SUBS

Re: [Sip-implementors] [Sip] Route header in NOTIFY

2008-11-06 Thread Rockson Li (zhengyli)
Yes, I think so. NOTIFY would be a mid-dialog request as constructed with Route header as normal mid-request specified in RFC3261. -Rockson From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Thursday, November 06, 2008 8:44 PM

Re: [Sip-implementors] is the JAIN SIP API's can listen any sip port given by user?

2008-11-06 Thread M. Ranganathan
On Thu, Nov 6, 2008 at 7:16 AM, parasuraman kumarasami <[EMAIL PROTECTED]> wrote: > Hello Team, > > we are developing the sip application using JAIN SIP > API's. now we are creating the socket to send and receive the sip message, > is this is possible, JAIN SIP API's can listen any particular

Re: [Sip-implementors] Why TCP keep-alive (PingPong) requires registrarreplyng single CRLF?

2008-11-06 Thread Attila Sipos
I think it is encouraged this way so that the SIP stack can have its own timeout. Some TCP/IP stacks take a while to timeout and the timeout on a TCP/IP stack isn't always configurable on a "per TCP connection" basis. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTEC

[Sip-implementors] is the JAIN SIP API's can listen any sip port given by user?

2008-11-06 Thread parasuraman kumarasami
Hello Team, we are developing the sip application using JAIN SIP API's. now we are creating the socket to send and receive the sip message, is this is possible, JAIN SIP API's can listen any particular sip port given by the user? if so, could any bady can tell me how we can implement it? Th

Re: [Sip-implementors] Why TCP keep-alive (PingPong) requires registrar replyng single CRLF?

2008-11-06 Thread Iñaki Baz Castillo
2008/11/6 Harsha. R <[EMAIL PROTECTED]>: > Just a hypothesis. not sure though. > > If registrar does not reply with CRLF, then client cant know if registrar > acknowledges the message or CRLF CRLF was received at the registar; so may > be registrar replies to stop any retransmission timers? Well,

[Sip-implementors] Why TCP keep-alive (PingPong) requires registrar replyng single CRLF?

2008-11-06 Thread Iñaki Baz Castillo
Hi, TCP keep-alive (PingPong) works as follows: - The client, after registering, sends periodically double CRLF to the registrar. - The registrar replies single CRLF. Why is required the registrar replying single CRLF in order to mantain the TCP connection open? Wouldn't be enough with the client