[Yeti DNS Discuss] 答复: Call for comment: Experiment proposal for Yeti Algorithm Rollover

[Davey Song(宋林健)]

Thanks for your comments, reply in lines

> > The reasons of only rolling KSK is that some people suggested that
> > some resolver may not tolerate a KSK and a ZSK using different
> > algorithms in the same zone (notably PowerDNS).
> 
> This sentence is unclear. If some resolvers cannot tolerate a KSK and a
ZSK with
> different algorithms, it seems to imply the opposite: we need to change
the
> algorithm of the ZSK as well.

Yes. if there is any risk, we should avoid it and roll both ksk and zsk in
the same time. Current proposal is designed deliberately to expose this
risk. I suggest we do it in the lab environment with different major
resolvers rather than test it in Yeti testbed. 

The concern I have now in this proposal is adding a new KEY(new algorithm)
before adding RRSIG signed by the new algorithm. In prior KSK rollover the
new key is pre-published to fit RFC5011 and use double DS approach. But the
combination of RFC5011 and algorithm rollover requirement makes double DS
approach failed. I think we should use double signature approach in this
case.

> Also, reading the excellent report
> <https://labs.ripe.net/Members/anandb/dnssec-algorithm-roll-over>, it
seems
> that some resolvers (that was three years ago) complained when the records
> were not signed with *all* the algorithms listed in the DS record. So,
when the
> new KSK will be accepted as a trust anchor, it may break these resolvers.

OK I will read it and return to you soon.

Davey