Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

The main thing is that *proxies* don't have this behavior! I consider B2BUA to be a very generic term - it can do most anything. While SBC is, as Hadriel says, a marketing term, its more specific than B2BUA and most things called that have, or can be configured to have, this behavior. T

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

What about RFC 5853 : Requirements from Session Initiation Protocol (SIP)? That one defines what SBC requirements are. Regards, Nenad -Original Message- From: Joel Gerber [mailto:joel.ger...@corp.eastlink.ca] Sent: July-18-13 2:17 PM To: Hadriel Kaplan Cc: sip-implementors@lists.cs.col

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

Wow, thanks for sharing. I honestly didn't know about that one! Joel Gerber Network Specialist Network Operations Eastlink E: joel.ger...@corp.eastlink.ca T: 519.786.1241 -Original Message- From: Hadriel Kaplan [mailto:hadriel.kap...@oracle.com] Sent: July-18-13 1:48 PM To: Joel Gerber C

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

There kinda sorta is one, in RFC 5853: http://tools.ietf.org/html/rfc5853 Obviously it's not complete/comprehensive, though. But yeah basically "SBC" is an industry market term, and it's muddy... not a formal definition like RFC 3261 "Proxy". -hadriel On Jul 18, 2013, at 11:22 AM, Joel Gerber

Re: [Sip-implementors] Call-id length

Hi there, Most probably the indicated length of the "call-id" is rather an imposed limitation of the concerned manufacturer for its product - IBM's Websphere and Microsofts MS-Turn, quote "This protocol is a proprietary extension to the Traversal Using Relay NAT (TURN)

Re: [Sip-implementors] Bulk registrations

RFC 6140 provides a mechanism; however I'm not sure if it is what you are wanting. > -Original Message- > From: Anshuman S. Rawat [mailto:ara...@3clogic.com] > Sent: Thursday, July 18, 2013 11:12 AM > To: sip-implementors@lists.cs.columbia.edu > Subject: [Sip-implementors] Bulk registrati

Re: [Sip-implementors] Overlap signaling in a native SIP network

On Jul 15, 2013, at 9:09 PM, SIP Learner wrote: > I think there is one benefit *when the caller dialed a wrong number!* > If overlap signallig were used, the caller might be informed by the proxy > immediately. In en bloc signalling, however, the caller will only be altered > after he dialed t

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

SBC is a rather arbitrary term referring to a bunch of different functions. Typically an SBC is a B2BUA, RTP media-proxy, stateful firewall, QoS/policer, NAT traverser and a transcoder, but this is not always the case. Some vendors stick on the name SBC, and it just performs SIP ALG functionalit

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

Then stop calling it a *proxy*! It is an SBC. Thanks, Paul On 7/18/13 6:28 AM, ikuzar RABE wrote: > Ok thanks for your responses, > There is indeed an RTP proxy within the sip proxy... and it works as you > described above. > > > 2013/7/17 Paul Kyzivat

[Sip-implementors] Bulk registrations

Hi, I have a need to REGISTER several distinct AORs with distinct contacts with the same REGISTRAR. I know multiple contacts for a single AOR can be done by a single REGISTER having multiple contacts. Is there a way in SIP by which I can REGISTER multiple AORs with distinct contacts? If yes, ca

Re: [Sip-implementors] Overlap signaling in a native SIP network

On Jul 15, 2013, at 9:09 PM, SIP Learner wrote: > I think there is one benefit *when the caller dialed a wrong number!* > If overlap signallig were used, the caller might be informed by the proxy > immediately. In en bloc signalling, however, the caller will only be altered > after he dialed t

Re: [Sip-implementors] Call-id length

Not only does sip have no maximum callid length, it has no bound on the length of most things in the message. Of course you can usually control the length of things that you originate, but you must not impose a limit on those you receive if you want to be interoperable. Just get it through your

Re: [Sip-implementors] Call-id length

Ok, thanks for these responses. I 'd like to add this: In RFC 3261 page 38 it is said: Use of cryptographically random identifiers (RFC 1750 [12]) in the generation of Call-IDs is RECOMMENDED. in RFC 1750 8.2.3 it is said : probably a good minimum for a very high security cryptographic ke

Re: [Sip-implementors] Call-id length

> 1) Why don't we have this info in RFC 3261? > What does RFC says about max Call-id length? SIP has no maximum Call-ID length. > 2) What about it actually? > Today I put my call-id max length to 256 characters > but I am not sure if it is correct ... You can impose a maximum; however that me

Re: [Sip-implementors] Call-id length

Someone else on the list could probably be more specific, but I believe this is a through-back to HTTP/1.1's maximum header length. It's not defined in RFC 3261, because it is stated in HTTP/1.1's RFC. That said; I'm not sure how many SIP implementations follow this "standard", as it isn't speci

[Sip-implementors] Call-id length

Hi all, I found with the link below (IBM) that Call-id max length is 256 characters: http://pic.dhe.ibm.com/infocenter/wvraix/v6r1m0/index.jsp?topic=%2Fcom.ibm.wvraix.voip.doc%2Fsiptags.html Microsoft said the same: http://msdn.microsoft.com/en-us/library/ff595864%28v=office.12%29.aspx. 1) Why do

Re: [Sip-implementors] RTP flow's route follows SIP flow's route ...

Ok thanks for your responses, There is indeed an RTP proxy within the sip proxy... and it works as you described above. 2013/7/17 Paul Kyzivat > As others have noted, for this to happen the "proxy" (proxies?) needs to > modify the SDP to cause this to happen. If it does this it has violated > t