[comp.protocols.appletalk] KIP limitations

mpm@BERLIN.ACSS.UMN.EDU ("Mark P. McCahill") (06/24/88)

Is anyone else running into the limits of the KIP code for the
Kinetics FastPath? We think KIP is great, and that's the problem. We like
KIP so well that we have loads of Kinetics boxes running KIP, and KIP's
limitations on number of routes, number of zones, and zone name length
are getting to be a bit confining.

Ideally, I would like to ONE large AppleTalk internet for the whole University.
Solving the problem by creating several independent AppleTalk internets to
avoid KIP's routing table limitations isn't too attractive because the 
independent AppleTalk internets won't communicate with each other.
So, if there is a way to get KIP to handle around 70 - 100 Kinetics boxes
my problem would be solved. (at least for a while...)

Solutions the involve running something other than KIP aren't really
going to work for us because:

1.) We have a policy of IP-traffic-only on our ethernet backbone, and KIP's 
AppleTalk-in-IP conforms to this policy. The IP-only policy is
is enforced by some routers in the backbone that only route IP, so running
the standard Kinetics software is not a good option. 

2.) We are running CAP on several of our UNIX boxes, and CAP requires
KIP.

So... Is anybody else running into this problem? Is anyone working on
a solution or will we have to do it ourselves?

Mark McCahill
Microcomputer & Workstation Systems Group
University of Minnesota

internet:	mpm@berlin.acss.umn.edu
bitnet:		MPM@UMNACVX