From: Aki Niemi <[EMAIL PROTECTED]>

   Currently, subnot-etags offers two modes of operation. In the first, the
   body of the SUBSCRIBE-generated NOTIFY is suppressed, and in the other,
   the entire NOTIFY is suppressed, and a new response 204 (No
   Notification) is generated instead.

   The latter is a considerable change to how SIP events works, and might
   cause a middlebox that tracks the SIP dialog to basically barf, thinking
   the dialog expires when it in fact gets refreshed.

I'm tempted to put all the responsibility on the SBC, since the
fundamental "deal" for an SBC is that in return for not being a proxy,
it takes on the responsibility of keeping up with evolving standards
and behaviors.

But looking at RFC 3265, I see that all 2xx responses to SUBSCRIBEs
are required to contain an Expires header for the subscription.  And
draft-ietf-sip-subnot-etags-00, section 4.5, item 2 reiterates that
rule.  So SBCs can track subscription lifetime without seeing NOTIFYs,
and for robustness should already be doing so.

Dale


_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [EMAIL PROTECTED] for questions on current sip
Use [EMAIL PROTECTED] for new developments on the application of sip

Reply via email to