Sun Jan 15, 2006 at 09:40:40 -0500, John Peacock wrote:
>Robert La Ferla wrote:
># host www.edmunds.com
>;; connection timed out; no servers could be reached
># host www.infiniti.com
>;; connection timed out; no servers could be reached
>
>If you are having problems with djbdns, why are you using a BIND tool that even
>the BIND people think is hopelessly broken? You should do something like this:
*snore*
>NTE: both those addresses are serviced by Akamai (a load-balancing service) so
>you are likely to get very different answers.
>
># dig www.edmunds.com @216.255.129.249
>
>The way to do that using the djbdns tools is:
>
>$ DNSCACHEIP=216.255.129.249 dnsqr a www.edmunds.com
>
>However, why are you trying to resolve that hostname with one of the Digex
>nameservers? Are you are Digex client? It appears they have their nameserver
>set up to refuse queries from random hosts (I cannot use it, for example).
Perhaps because of:
edmunds.com. 172800 IN NS mia01.digex.com.
The digex.com name servers refuse queries with the recursive flag set-
behaviour that should not be surprising to people who use djbdns.
The real problem here is that Dan refuses to fix code that breaks on a
DNS configuration (akamai's) that, while not illegal, is substantially
more complex in practice than was anticipated when he coded dnscache.
Google for akamai + dnscache.
To paraphrase
Akamai: It works for MaraDNS. It works for PowerDNS. It works for
Nominum CNS. It works for Microsoft DNS. It works for hundreds of
millions of BIND users. It just doesn't work for djb.