FQDNs in masters-list (was: Help: Secondary for...)

[Andreas S. Oesterhelt]

Hello

Brad Knowles writes:
> 	It's one thing to understand the desires of the users and the 
> people who are ignorant of the technical details, it's quite another 
> to actually change the protocol itself (and the programs that serve 
> it) to accommodate those desires, without causing irreparable harm.

Maybe it's just my ignorance of the technical details, but I fail to
see how 
 a) Allowing masters to be specified by name and
 b) having the slave do, if necessary, lookups for the master's
    address before zone transfers, 
could possibly require a change in the protocol or standard.

As for the BIND software, I would (naively) expect changes to be
confined to minor extensions to parsing the named.conf file, the
dns_c_slave_zone data structure and adding the lookup.

Rest assured that I'm quite aware of the implications of changing
the protocol of such a fundamental service as DNS and I'm not
proposing that.

In fact, my last mail wasn't even a proposal but the question to
you experts, whether such a proposal would make sense.

Best regards,
--Andreas