Hi Håvard, On Nov 23, 2017, at 11:36, Havard Eidnes <h...@uninett.no> wrote:
>>> Secondly: can someone please explain to me why the idea of a >>> "primary master" where the zone data originates from and where >>> updates are performed is considered archaic? >> >> I think the only in-protocol use of the MNAME field is to >> specify the name to which UPDATE messages are sent. > > Really? I read the NOTIFY RFC (1996) so that if the MNAME name > server is included in the NS set, it will by default be excluded > from receiving NOTIFY messages from the other name servers when > they reload the zone, ref. the definition of "Notify set" in RFC > 1996. Oh, good catch! I did not even know that text was there. I stand corrected! That doesn't change the wider opinion I was trying to describe, though: in the modern/transfer graph scenario I was painting, NOTIFY configuration is far more nuanced than as described in RFC1996 and in practice (in BIND parlance) is an automatically-managed set of also-notify configuration. Joe _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
