RE: [[e-mail]: Re: BGP4 stuff: Local Preference Computation]

["NITTMANN Michael (MSMail)" <[e-mail]>]

Looks to me as if everybody agrees: I would just generalize the probably 
narrow meaning of 'mandatory': not only in exchanged parameters/variables, 
but also in IBGP, say 'internal' variables and paramers.

Or let's do classes: mandatory external, mandatory internal.

I would not need it, since localpref is by def internal, and should be 
mandatory in all implementations to assure vendor compatibility *within* an 
AS (well, now I wait for some C word people to say this is not necessary 
since router names start with C anyways  ;=]  ).

Since we all agree that localpref is not exchanged across AS boundaries, a 
tweaking based on a purely internal parameter cannot avoid loops, since the 
AS partner has no clue. It is a reactive repair.

MED, however, is exchanged or should be mandatory external, and here we have 
then an automatic adaptation mechanism that could recognize conflicting MED 
and either adjust, or override and message.

Mike

----------
From:  Yakov Rekhter[SMTP:[e-mail]]
Sent:  Donnerstag, 5. September 1996 14:41
To:  John G. Scudder
Cc:  [e-mail]
Subject:  Re: [[e-mail]: Re: BGP4 stuff: Local Preference Computation] 


John,

> This should be fixed by reclassifying LOCAL_PREF as well-known mandatory.

"Mandatory" means that it *always* has to be present. LOCAL_PREF is not
carried between external peers - just among the internal peers.

> [Other than the text being just plain wrong, the other possibility is that
> "discretionary" is being used to mean something like "this is only
> mandatory in some cases".  This is a terrible misuse of the language if
> true, and should still be fixed, if necessary by inventing some new term
> like "well-known sometimes-mandatory" or something.  (Note that the RFC
> nowhere defines "discretionary" so we are left with the dictionary
> definition, e.g. "left to discretion: exercised at one's own 
discretion".)]

Suggestions for better terminology area always appreciated (especially
if accompanied by a "rough consensus").

Yakov.