🚀 go-pugleaf

In article <5560009@hpindda.HP.COM> tozz@hpindda.HP.COM (Bob Tausworthe) writes:
>I am writing this note in an effort to obtain some information on the
>addressing strategies employed by different organizations in their ISO
>networking products. The areas I'm concerned with are:
>
>   - CONS vs CLNP. How are you making the decision?
We use our directory service to select the transport protocol (TP0 vs TP4)
and the network layer.
>   - ES-IS protocol. Are you implementing it on ES systems?
Yes. It is the only dynamic way to determine a system's SNPA (ie Ethernet
(whoops, 802.3) address).
>   - first hop decisions for ESs. are you statically configuring this
>     information?
We have a mechanism for determining if an NSAP address is on the
local subnet (sort of a "on the same subnet of" function). If this
function decides that the NSAPs are on the same subnet, then we route
directly to the destination. Otherwise, the packet is sent to an
IS. When our system receives an ISH, a default route is generated toward
that IS. Subsequent ISHs from other ISs overwrite the previous information.
>   - CLNP over X.25. What addressing methods do you support.
We normally use OSINET addresses when using CLNP/X.25 (but this is not
strictly necessary).
>	 - TPvsCLNP SAP identifier in Call User Data
If you use 1984 X.25 you can send the NSAP in the X.25 call request user 
data (X.25 CRUD). This is only necessary when running TP4,0/CONS.
>	 - Subaddressing
What about it?
>	 - Extended address facility
If we have 1984 X.25, we use it.

Rob Hagens
UW Madison Computer Science (ARGO project)
Rob Hagens
..!{allegra,heurikon,ihnp4,seismo,sfwin,ucbvax,uwm-evax}!uwvax!hagens
hagens@rsch.wisc.edu
Weekend: A time between work and more work.