RE: Executive summary of: RE: hard questions: request routing

[Oliver Spatscheck <spatsch@research.att.com>]

Abbie Barbir writes:
 > 
 > 
 > > -----Original Message-----
 > > From: Francis Zane [mailto:francis@research.bell-labs.com]
 > 
 > > My main concern with fully general forwarding mechanisms remains the
 > > ability to deal with different (particularly non-DNS) RRS mechanisms
 > > and all the cases which could arise.  As a (mostly thought-experiment)
 > > example, here's a weird case that concerns me: Suppose I have an RRS
 > > which spans two specific RR mechanisms: DNS (based on local DNS IP)
 > > and some form of L7 (based on client IP).  Is there a way to guarantee
 > > correct (ie, non-looping) behaviour?  (CDN A DNS redirects to CDN B
 > > based on local DNS, which HTTP redirects to CDN A based on client IP).
 > > 
 > 
 > yes, the matrix approach. Content hops are cumulative, you always know where
 > content
 > came from. You know the whole path.
 > 
 > abbie

Abbie,

In the case of DNS you said you would carry the path in the DNS name so for
this RRS it is clear to me why it works.  However, you still haven't answered
Francis question of how you would address this in a combination of non DNS RRS
since carrying it in the DNS name doesn't work for those RRS. You also stated,
earlier that you do not want to use global state, so now I am confused again,
how do you address the problem in the general RRS case.

Oliver