I'm under the impression Signers will generally not know of the existence
of other Signers before they start generating keytags.  Perhaps each Signer
should generate a salt that is likely to be independent of the salt used in
other Signers.

Steve Crocker



On Wed, Jul 9, 2025 at 11:05 PM Mark Andrews <[email protected]> wrote:

>
>
> > On 10 Jul 2025, at 06:01, Peter Thomassen <peter=
> [email protected]> wrote:
> >
> >
> >
> > On 7/9/25 20:09, Jim Reid wrote:
> >>> Can you clarify source of your confidence about this 'not causing
> issues'?
> >> Mental arithmetic. There are 2^16 possible key tags => there's a one in
> 2^15 chance a new tag clashes with an existing one.
> >
> > Uniformity of the keytag distribution is a wrong assumption, see
> https://ripe78.ripe.net/presentations/5-20190520-RIPE-78-DNS-wg-Keytags.pdf
> >
> > This deck also has some slides on impact.
>
> So just use a method that will work reasonable well with the generated key
> tags.
>
> For two signers A uses 'tag < 32768' and B uses 'tag >= 32768’ as part of
> the acceptance criteria
> when generating a new tag.
>
> Resolvers already do the bulk of the work with DNSSEC.  A few more key
> generation attempts on the
> authoritative side won’t hurt anything.
>
>
> > Have fun,
> > Peter
> >
> > _______________________________________________
> > DNSOP mailing list -- [email protected]
> > To unsubscribe send an email to [email protected]
>
> --
> Mark Andrews, ISC
> 1 Seymour St., Dundas Valley, NSW 2117, Australia
> PHONE: +61 2 9871 4742              INTERNET: [email protected]
>
> _______________________________________________
> DNSOP mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>


-- 
Sent by a Verified

sender
_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to