On 7/18/2010 7:19 AM, Yaakov Stein wrote: > Stewart and all, > > Indeed, the limited number of reserved labels was the main drawback > of the "reserved label" approach when I first raised it. > > The other possibility of this sort was a special CoS bit combination, > but there are no universally defined CoS (yet). > > So, are we back to the GAL ? > > If so, there are still at least 4 options : > 1) define a new protocol type (plenty of openings in THAT registry!) > 2) put 1588 packet into an ACH TLV > 3) use IPv4 protocol type with 1588 over UDP/IP but mark it in some > other way (e.g., reserved field)
You *MUST* allow both IPv4 *and* IPv6 for an IETF document if you are using IP at all. If you need to know which is being used then there is a major problem here. Danny > 4) define a new first nibble (0010 ?) <ugh!> > > Y(J)S _______________________________________________ TICTOC mailing list [email protected] https://www.ietf.org/mailman/listinfo/tictoc
