Cell Relay Archive

[calin@roedu.net]

On Mon, 14 Jun 1999, Nige wrote:

> 
> Paul Koning wrote:
> > 
> > The UNI and signalling specs say "this end system identifier can be
> > a globally unique identifier specified by an IEEE MAC address".
> > 
> > So the MAC address rules do apply.  What bugs me is the words "can be".
> > That's bogus because it defeats the purpose.
> 
> I'm sure the words "can be" are intentional - to mean that this is a 
> suggested value for the field.  But in no way is it mandatory.
> I have a paper from the ISO explaining their (careful) use of
> "shall", "should" etc.   I'm pretty sure I've seen similar from the ATMF.
> 

I think that "can be" suggested that other values could be used as ESI:
for example (SUN and Cisco put this in their user manuals) that IP address
of the interface can be (again :-) used as ESI (this is also an unique
value at least for local net). I am not sure that I have
seen other examples. Is anyone using something else except MAC/IP for ESI? 


Best luck!

--Calin Poenaru