in-addr.arpa server problems for europe? [SEC=UNCLASSIFIED]

Michelle Sullivan matthew at sorbs.net
Mon Feb 15 07:04:50 CST 2010


Wilkinson, Alex wrote:
>     0n Mon, Feb 15, 2010 at 01:40:31PM +0100, Michelle Sullivan wrote: 
>
>     >Michelle Sullivan wrote:
>
>     >michelle at enigma:~$ dig +trace +bufsize=512 -x 81.255.164.225
>     >michelle at enigma:~$ dig +bufsize=4096 -x 81.255.164.225 @NS3.NIC.FR
>
> Curious, why did you modify 'bufsize' ?
>   

Well I started here:

http://sel.icann.org/node/715#fn1

and figured that it was a way to force the packet size and protocol so
that I could fit it to known constraints in the PIX

eg:

Fix to 512 bytes and if the PIX is rejecting anything over 512 bytes
there is a simple answer.
Fix to 4096 bytes and it forces to EDNS (v0) - as can be seen in the
output, to see if the PIX is just dropping all EDNS..

obviously the resulting size sent back I cannot control (except by
limiting the maximum size), so the next step was to query all (or a
selection) of the servers being traced through.

What I can't figure out is why I can query the servers directly and get
a response but the trace fails.

Any insight will be greatly appreciated.

Michelle






More information about the NANOG mailing list