> -----Original Message-----
> From: M. Ranganathan [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, June 13, 2001 5:14 PM
> To: [EMAIL PROTECTED]
> Subject: [Sip-implementors] Stateful vs Stateless vs something in
> between?
>
>
> Hello!
>
> There is a distinction made in the RFC between stateful and stateless
> proxies. For example, stateful proxies are supposed to keep
> transaction
> state and are responsible for re-transmission of requests.
>
> Given that a user agent is oblivious to whether it is talking
> to a stateful
> or stateless proxy at least some of these functions (such as
> re-transmission
> of requests) have to be supported by the user agent anyway leading to
> redundancy. ( Clearly, there are instances where proxies need to be
> stateful to support certain kinds of funcitonality whereas
> for things such
> as re-transmission, I see these as being performance
> enhancements in the
> proxy. )
>
> Is it required that proxies choose one mode versus the other
> (i.e. stateful
> versus Stateless) or can they adopt an intermediate strategy?
I'm not sure what you mean by an "intermediate strategy". A proxy can choose
to be stateful or stateless for each transaction on a case by case basis.
However, if you want to implement some kind of new level of state (for
example, where you retransmit the request in a proxy twice, and then become
stateless), you really can't do that. In most cases, it will effect end to
end reliability. Correct operation is guaranteed by compliance to the
defined FSMs. If you deviate from them, bets are off.
-Jonathan R.
---
Jonathan D. Rosenberg, Ph.D. 72 Eagle Rock Ave.
Chief Scientist First Floor
dynamicsoft East Hanover, NJ 07936
[EMAIL PROTECTED] FAX: (973) 952-5050
http://www.jdrosen.net PHONE: (973) 952-5000
http://www.dynamicsoft.com
_______________________________________________
Sip-implementors mailing list
[EMAIL PROTECTED]
http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
