[dns-operations] Against ghost domain name

[paul vixie]

On 2/10/2012 3:34 AM, Masato Minda wrote:
> On 2012/02/09 23:19, paul vixie wrote:
>
>> i don't think this is correct. if an iterative caching resolver
>> re-iterates through the registry's name servers then an NXDOMAIN would
>> stop the query. ghost domains work only when this re-iteration does not
>> occur due to caching of NS and related A/AAAA.
> Thank you for your comment. I understand it.
>
> So, I think that if the registry makes the wild-card or some synthesis
> response name server, it works well in this situation.

can you explain what effect a wild-card or synthesis, which occurs in an
authority name server, could have if the data in question is already
cached in a large number of recursive name servers, who are therefore
not re-iterating through the authority server, when stubs ask them for
the long-lived data?

> But I know that some old caching nameserver is not effective.
> Many old nameservers are still working in the internet. :-(
>
> I will make demonstration environment and test it.

i look forward to viewing your results.

please also consider points (A) and (C) from this internet draft:

http://tools.ietf.org/html/draft-vixie-dnsext-resimprove-00

paul