I was the last one to "hold the pen" on this one, though I came to the party pretty late (basically, taking over for Bill Fenner when he became an AD, just to shepherd it thru the last stages of approval and publication).
I remember there was an IPv6 MIB design team, that I believe set the direction for the combined v4/v6 MIBS (4022, 4113,4292, 4293). At the point I was working on it, the target was consistency with the TCP MIB (RFC 4022). One point here is that ICMPv4 and ICMPv6 are separate protocols, whereas UDP is the same protocol for v4 and v6, so the situation is more similar to the TCP situation than it is to the ICMP one. FWIW, I know HP switches to report v4+v6 in the In/OutDatagrams counters. John -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Michael MacFaden Sent: Thursday, February 28, 2013 11:07 AM To: opsawg Subject: [OPSAWG] RFC 4113 -- UDP-MIB -- stats questions In -- http://www.ietf.org/rfc/rfc4113.txt I see that the oiginal mib-2 scalar counters were kept udpInDatagrams...udpHCOutDatagrams I wonder how many implementations only report v4 vs v4 + v6 when this mib is claimed as being implemented? Also why were these were not replaced with per protocol stats to be consistent with how say icmp stats were in RFC 4293? Does anyone recall the thinking here, can't see find any discussion in the RFC. Thanks, Mike MacFaden _______________________________________________ OPSAWG mailing list [email protected] https://www.ietf.org/mailman/listinfo/opsawg _______________________________________________ OPSAWG mailing list [email protected] https://www.ietf.org/mailman/listinfo/opsawg
