On Thu, Apr 9, 2020 at 3:48 PM Brian Dickson <brian.peter.dick...@gmail.com> wrote:
> > On Wed, Apr 8, 2020 at 10:10 PM Benjamin Kaduk via Datatracker < > nore...@ietf.org> wrote: > >> >> Section 6.1, 6.2 >> >> Should we say anything about when it's safe for a new ZSK to be used to >> sign responses? >> > > I think, technically speaking, it is always safe to use a new ZSK, and the > only concern is when it's safe to remove and/or revoke the old ZSK (i.e. > that it is safe to use ONLY the new ZSK.) > Hi Brian, Your statement is true for double signature ZSK rolls (which are less common). In the pre-publish ZSK rollover scheme (much more common), the new ZSK starts signing new or modified data in the zone _exclusively_, and the corresponding old ZSK signatures are yanked out, after "propagation time + DNSKEY TTL time" has transpired. So if you do this too early, data in your zone could become unvalidatable. This is described in RFC6781, although the description is a quite dated -- since it talks about monolithic re-signing of zone data, and does not contemplate incremental signing, which is becoming far more common these days. Shumon.
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
