[dns-operations] Anycast vs. unicast NS

[David Miller]

On 3/18/2011 10:00 AM, Simon Munton wrote:
> On 18/03/2011 12:32, Jim Reid wrote:
>> In the same way that a TLD would be much better off only using a
>> single DNS implementation and just one hardware/OS platform for
>> its servers?
>
> Hard to see how more diversity could be a bad thing. Our position is 
> we recommend customers to retain unicast node(s), and use more than 
> one anycast operator.

Retain unicast nodes to what end?

>
> why wouldn't you ?
>

Diversity is "good"(tm) except when it is "bad"(tm).  Diversity does 
not, in itself, provide protection against much.  Do you also recommend 
that your customers run *nix DNS servers and M$ DNS servers (you will 
want Win 4.0, 2000, 2008 servers in the mix - for diversity)?  Do you 
recommend BIND 8 in addition to several different versions of BIND 9 for 
customer implementations?  How about distribution, kernel, nic driver 
version?  The matrix of diversity that gives you "protection against any 
issue" approaches an infinite number of different configurations.

So, as above, retaining unicast nodes provides protection through 
diversity in what way?

> _______________________________________________
> dns-operations mailing list
> dns-operations at lists.dns-oarc.net
> https://lists.dns-oarc.net/mailman/listinfo/dns-operations


-- 
-___________________________________
David Miller
Tiggee LLC
dmiller at tiggee.com