IPv6 Dig reverse lookup options roadmap

Edward C Clarkson edcclark at us.ibm.com
Fri Mar 15 20:19:45 UTC 2002

BIND developer folks--

Given the recent changes in IPv6 DNS standards, the current DiG behavior 
for reverse IPv6 lookups isn't (or soon won't be) as useful as it could 
be.  Given that, in our port we are considering the following changes to 
the dig syntax:

|Reverse Lookup Format  |  Current Option  |  Proposed Option 
|bitstring, ip6.arpa    |       -x         |       ?? 
|nibble,    ipt.int     |      -n -x       |  -n -x or -i -x? 
|nibble,    ip6.arpa    |       N/A        |       -x

We decided that the RFC standard (nibble, ip6.arpa) should obviously be 
the easiest to do (via one option instead of two), but weren't sure 
whether it was worth including code to do reverse lookups for an 
experimental label type or not.  We also weren't sure whether we were 
going to leave the -n option as the same letter, since it's not really an 
option to append 'ip6.int' rather than use the nibble format.  If nothing 
else, the doc. for the -n option would be changed.

I noted that the most recent 9.3.0 snapshot has the same behavior as 
9.2.x...given current (RFC 3152) and probable future RFCs, the present dig 
syntax/function isn't exactly ideal, and I would expect that it will be 
changed in the (near?) future.

So my question for the BIND developers:  is there any plan for the 
behavior post-9.2 dig might have?  If so, is it similar to what I have 
here?  If not, are these changes reasonable?  Is a bitstring option 

We would very much like the changes we make to our port of the 9.2 branch 
to match whatever BIND does going forward so we're not caught differing 
from the reference server after our release date.  If anyone has some 
helpful hints/comments on what BIND might (or might -not-) do, I'd 
appreciate the clue!


Edward Clarkson

z/OS Communication Server Development
Email:  edcclark at us.ibm.com

More information about the bind-workers mailing list