Authors (and coworkers!), a few quick comments as I catch up on my GROW mailing list backlog.
As written, I'm not clear what the intended encoding for the route type is. A type 0 is defined, but I'm suspecting it's intended to be a placeholder? If so, this suggests the type is the TBD values listed in the IANA considerations. For the subtypes, starting at code point 0 is not required. I'd suggest simply numbering them in alignment with the EVPN route type registrations and leaving value 0 as "reserved". As more general discussion for grow, a motivation for these new types rather than leveraging the existing types defined in RFC 7854 and the existing rib-stats draft is the family-specific name space requirements for the subtype. The working group should consider whether this format generally addresses similar requirements that families such as MVPN might also have. The related discussion point for the WG is whether the rib-stats that are per-AFI-SAFI are good enough for typed families like EVPN or whether those might need per-type refinements? -- Jeff _______________________________________________ GROW mailing list -- [email protected] To unsubscribe send an email to [email protected]
