On Thu, 21 Feb 2019 at 12:25, Tony Finch <[email protected]> wrote:

> Mark Andrews <[email protected]> wrote:
> > > On 21 Feb 2019, at 6:30 am, Tony Finch <[email protected]> wrote:
> > >
> > > Does it need to be per-record? Why not GC the whole RRset?
> >
> > Because there are scenarios where you do want to GC as single
> > record from the RRset. AD has them with SRV.  Each server adds
> > its own SRV record to the RRset.
>
> Oh I see, that makes sense. Things like this need explaining in the
> document.
>
> Why not simplify the RDATA format so there is just one item per record,
> instead of containing a list of lists? Aren't the individual items going
> to be added/removed independently of each other?
>

Good idea.

A more radical approach would be to have just one hash per item per record.
This has the attraction of allowing a truncated hash to be used, without
the need for length fields.

--Dick
_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to