CNAME Issue - Whether to use CNAME-data or Response-Flag

Steven Wilmot steven at data-utilities.co.uk
Sat Apr 10 00:52:40 UTC 2010


See inline replies below 

-----Original Message-----
From: Doug Barton [mailto:dougb at dougbarton.us] 
Sent: 09 April 2010 8:50 PM
To: Steven Wilmot
Cc: bind-users at lists.isc.org
Subject: Re: CNAME Issue - Whether to use CNAME-data or Response-Flag

>> When I try to resolve mail.wilmot.me.uk against my local resolver (which
happens to be BIND 9.6.2-P1 atm) I get the expected result:

>> host mail.wilmot.me.uk
>> mail.wilmot.me.uk is an alias for wilmot.me.uk.mail.aaisp.net.uk.
>> wilmot.me.uk.mail.aaisp.net.uk has address 81.187.30.19
wilmot.me.uk.mail.aaisp.net.uk has IPv6 address
2001:8b0:0:30:230:48ff:fedb:25dc

>> Looking at the response from the local resolver with dig confirms that it
is as expected.

Good - this is the result that I expect to receive

>> Querying your authoritative server also produces the expected result:
>> (data removed for clarity)

>> So I'm not sure exactly what problem you're reporting.

Doug, thanks for the comments.

The "problem" would seem to be an issue with the way that "Microsoft DNS"
server interprets the result.

I already have this as an open Microsoft support incident, but am trying (by
asking on the BIND support forums in the hope of finding some DNS-experts)
to find some "official information to reinforce my opinion that "BIND is
operating 'CORRECTLY'", and to find the exact RFC text (or otherwise) that
basically says the follwing:

* Here is the spec, here is confirmination that the "original record from
primary-dns.co.uk" is valid, and here is an RFC saying how this should be
interpreted.

So, I'm really asking for assistance in troubleshooting, rather than
reporting a problem.

It is no good me just saying to Microsoft "I think you are wrong", but if I
can find a "SPECIFIC" problem and come up with a statement like "Here is
EXACTLY where the Microsoft-implemention does NOT adhere to section XXX of
RFC yyy" then I'm more likely to get them to implement a hotfix.

---

Regards, 

Steven Wilmot 
Director 
Data Utilities Ltd 




More information about the bind-users mailing list