From: Curtis Phillips (phillipscurtis@xxxxxxxxxxxx)
Date: Tue Jan 25 2000 - 16:37:20 GMT-3
Apologizing in advance for both the length and convolutedness of this post. I
am certain that I am not the only one trying to get these concests straight
though...
I am a little confused about NSAP assignment and cooperation between end
systems and the LS1010. When asked to assign an initial ATM (nsap) address
to the LS1010, where does that address come from? I have seen specific and
made-up addresses both. For the initial part of the address, it typically
starts with the 47.0091.8100. I recognize 47 as the AFI. the 0091.8100 are
(I believe) Cisco-specific, and the remainder of the non-esi bits are part
of the ATM network routing hierarchy-specific numbering.. So, what are we
saying when we randomly assign 1111.2222.3333 or whatever? That we have a
flat,single-tiered,ATM routing domain? And that this is specific for this
interface only, and for all end-systems connected to it as well?
Is it correct to think that an end-system esi address assignment overrides
the mac-address of the end-system interface? If the end system is a catalyst,
which mac address is used for the 7-byte portion of nsap? If a router, then
the bia?
How does the switch recognize which 7-bytes is in use? If you are going to
answer ILMI, then please try to be specific.
Is it true to think that nsap and nsap config is not necessary in an pvc
configuration? After all vpi/vci pairs are already known ny all through manual
config right?
Speaking of vpi/vci pairs.. are they only relavent to the uni interface
interactions? I mean, are all PNNI transactions conducted via NSAP addressing
to maintain the heirarchical ATM routing internal to the ATM cloud?
Please feel free to add and/or correct any errors in my logic.
Regards,
Curtis
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:22:46 GMT-3