>> and, if we do mbz, they can then write the rfc which changes that
> The problem then is all the implementations -- or firewalls -- that 
> will check that the bits really are zero upon receipt.  If you want 
> MBZ, say "sender MUST set to zero; receiver and middle boxes MUST NOT 
> check".

as i thought it was the stated mbz formulation, i was assuming this.

> Then you have to hope that folks listen to that part.  Especially for 
> firewalls, I wouldn't count on that.  (For precedent, so to speak, look 
> at what happened with the ECN bits when a particular version of Linux 
> started using them.)

hmmm.  i think i am starting to reconsider.  idiots are soooo creative.

randy
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to