Hi Dale, ext [EMAIL PROTECTED] wrote: > > Often a subscriber will send the SUBSCRIBE to a URI which is not the > > request-URI of the SUBSCRIBE when it reaches the notifier agent. This > > happens due to normal SIP request routing, but the routing can also > > change with time. > > Correct, but are you saying there might be more than one agent with a > different idea on the entity-tags of entities pertaining to a certain > resource? In other words, that the resource has multiple agents that may > have a different view of the resource state? > > I believe this is disallowed per RFC3265 already. > > I doubt things could work with multiple agents that have a different > view of the resource state. > > But there might be multiple agents with different addresses, from whom > you request the state of the same resource using different SIP URIs. > (Which ultimately might all be destinations reached by sending > requests to the same AOR.) > > I don't think it interferes with the subnot mechanism, but my point is > that a resource might *not* have a single, specific SIP URI, either as > the request-URI when the SUBSCRIBE reaches the agent(s) which can > deliver its state, or as the address to which UAs subscribe.
That's fine. I think aliases work as long as the relationship is clear. A single resource can have multiple URIs pointing to it, and each resource will only ever have a single agent responsible for it. > All you need to make subnot work is that every UA agrees on which > subscriptions are "for the same resource", and that all notifiers for > "the same resource" generate the same etags for the same state, and > different etags for different state. Right, that would need to be spelled out in section 5.1., I think. Cheers, Aki _______________________________________________ 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
