[prev in list] [next in list] [prev in thread] [next in thread] 

List:       djbdns
Subject:    Re: Utility to find the source of multiple A records?
From:       Lloyd Zusman <ljz () asfast ! com>
Date:       2004-01-25 4:59:26
Message-ID: m3wu7gbp3l.fsf () asfast ! com
[Download RAW message or body]

John Peacock <jpeacock@rowman.com> writes:

> Lloyd Zusman wrote:
>
>> But given the fact that '.', '&', '=', and '@' records all automatically
>> generate A records, I'm not sure that it's possible to completely avoid
>> multiples.
>
> Only if you include both the name _and_ IP address in multiple lines.
> For example, my domain file contains these lines:
>
> .rlpgbooks.com::a
> .rlpgbooks.com::b
> =a.ns.rlpgbooks.com:12.38.22.1
> +b.ns.rlpgbooks.com:12.151.2.99
> ^99.96/27.2.151.12.in-addr.arpa:b.ns.rlpgbooks.com
>
> In this case, only the '+' and '=' lines generate A records.  The reason
> I don't have '=' lines for both hosts is that I have to do the stupid
> RFC-2317 method (since I don't control the entire Class-C).  An
> equivalent set of lines would be:
>
> .rlpgbooks.com:12.38.22.1:a
> .rlpgbooks.com:12.151.2.99:b
> ^1.22.38.12.in-addr.arpa:a.ns.rlpgbooks.com
> ^99.96/27.2.151.12.in-addr.arpa:b.ns.rlpgbooks.com
>
> where I would be relying on the '.' lines to generate the A record.

Yep.  Now I've done this.  Thanks.


>> Given my own setup, would the multi-datafile patch get rid of the
>> multiples that result from '.' and '@' records?
>
> No, you still need to fix your data lines, no matter how you store the
> individual data lines.  I partition the data lines per domain purely for
> human readable reasons; in all ways it is completely equivalent to just
> cat'ing the files together and feeding that to tinydns-data.
>
> John

I manage that by a series of templates and a script which generates the
zone files, which I only look at for debugging purposes.  They get
cat'ed together in the makefile.


-- 
 Lloyd Zusman
 ljz@asfast.com

[prev in list] [next in list] [prev in thread] [next in thread]