Sitefinder II, the sequel...

Mark Jeftovic markjr at easydns.com
Mon Jul 10 20:01:57 UTC 2006




Niels Bakker wrote:

>> Also, sitefinder created a wildcard DNS record where none existed 
>> before, breaking all kinds of applications in the process, openDNS 
>> doesn't do this.
> 
> 
> Wrong.  Asking their "big caching nameserver" for gibberish returns "IN 
> A 208.67.219.40" instead of NXDOMAIN.  Same breakage occurs, although 
> they return NXDOMAIN instead of NOERROR when queried about MX or AAAA 
> records, so ironically damage for IPv6-enabled applications is limited.
>

I stand corrected, however this is not as big a deal as when sitefinder 
did it because as we've both observed, this is voluntary. If using this 
breaks your application, don't have your application use it, with 
sitefinder you didn't have the choice.

For it's target market: end user DNS resolution, the side effects will 
be minimal if anything.

> 
> Several people have eloquently expressed why creating different views of 
> a global namespace is a bad idea before on this mailing list.
>

I don't consider this a different view of the global namespace. If they 
decide to add ORSC root glue or New.net domains then it'll be a 
different view of the global namespace. Hopefully they wouldn't be that 
reckless.

> 
> Have you switched your company over yet?
>

They way we run our applications doesn't lend itself to using it (it's 
that choice thing again), but I've got a few workstations using it and 
one of my laptops. It's also a handy offsite resolver to use to check 
DNS settings from outside our own cloud.

We also get asked our members if there is a viable resolver they can use 
and we'll be happy to recommend this.

-mark


-- 
Mark Jeftovic <markjr at easydns.com>
Founder & President, easyDNS Technologies Inc.
ph. +1-(416)-535-8672 ext 225
fx. +1-(866) 273-2892



More information about the NANOG mailing list