[dns-operations] Input from dns-operations on NCAP proposal

Peter Thomassen peter at desec.io
Wed May 25 13:37:43 UTC 2022


Hi Thomas,

On 5/23/22 15:48, Thomas, Matthew wrote:
> In the 2012 round of new gTLDs, DNS data collected at the root server system via DNS-OARC’s DITL collection was used to assess name collision visibility. The use of DITL data for name collision assessment purposes has growing limitations in terms of accessibility, increasing data anonymization constraints, a narrow data collection time window, and the limited annual collection frequency.

I think these are valid concerns, but they don't necessarily mean that a new assessment methodology is needed. Instead, we could try to work towards reducing these limitations, i.e. improve accessibility, collection frequency etc.

> Other changes in the DNS, such as Qname Minimization, Aggressive NSEC Caching, etc., also continue to impair name collision measurements at the root.

QNAME Minimization drops labels from the left. How would it impact root traffic?

Aggressive NSEC caching covers non-delegated domains. Assuming such a record is cached, the root would not be asked for a name contained in the range, regardless of which special response strategy would be employed for such queries.

In both cases, I think it would be helpful to understand better how these mechanisms impair name collision measurements. Can you elaborate?

> In preparation for the next round of TLDs, the NCAP team is examining possible new ways of passively collecting additional DNS data while providing a less disruptive NXDOMAIN response to queries.

Before deciding what set-up best suits the data collection, I'd like to understand what data do you want to collect specifically?

> The proposed system below is an attempt to preserve the NXDOMAIN response these name collision systems are currently receiving, 
[...]
> The proposal would involve delegating a candidate TLD. The delegation process of inserting a string into the DNS root zone will make the TLD active in the domain name system. The required delegation information in the referral from the root is a complete set of NS records and the minimal set of requisite glue records.

Given the DNS protocol, these two goals seem inconsistent. If the servers referred to by the NS rrset do know the zone, they will answer NOERROR for apex queries, and depending on the query type, they will also return records (e.g. NS, or SOA). If they don't know the zone, the best practice response is REFUSED.

There doesn't seem to be a compliant way to delegate a candidate TLD and then have the auth return NXDOMAIN to queries for that domain. (If you do that, there is no guarantee of success. The set-up would be strange, and resolvers may decide to pass SERVFAIL to their clients, for example. Also, cache issues may arise, as pointed out by Vladimír.)

> Configuration 3: Use a properly configured empty zone with correct NS and SOA records. Queries for the single label TLD would return a NOERROR and NODATA response.

If I understand correctly, that's similar to what was done in 2012. Again, I'm not sure why it would not work now when it did back then. I think this is the question that should be answered first.

> The level of disruption to existing private use of such labels by this restricted form of name delegation would be reasonably expected to be /minimal/;

I think it would be "hoped", not "expected" to be minimal. :-)

Best,
Peter

-- 
Like our community service? 💛
Please consider donating at

https://desec.io/

deSEC e.V.
Kyffhäuserstr. 5
10781 Berlin
Germany

Vorstandsvorsitz: Nils Wisiol
Registergericht: AG Berlin (Charlottenburg) VR 37525


More information about the dns-operations mailing list