On February 2, 2010 2:25:50 PM -0800 Rob Tanner <rtan...@linfield.edu>
wrote:
cached (i.e. Is no data treated the same as bad data by upstream bind
servers?
I didn't entirely follow your ramble (paragraphs would have helped),
but it's not BIND or other nameservers that would be the real problem,
it's the applications that depend on name services. For example, if
your link goes down and instead of a DNS lookup which results in an
answer of an MX server that doesn't respond, someone trying to send
you mail would (after cache timeout) get back a non-result DNS answer
and might bounce a mail instead of queueing it for later delivery.
That's perhaps not a good example because actually MTAs should handle
this case as a transient error and queue any mail, but you get my point.
Consider also that folks just browsing your website will get a different
kind of error which might lead them to believe that your site doesn't
even exist. That would definitely be worse than "connection timed out".
Other applications may result in similar types of disconcerting errors
instead of just connection timeouts.
You really do need multiple nameservers, and you absolutely need to make
your DNS zone transfers reliable. I do sympathize with you. Old data is
often worse than no data.
-frank
_______________________________________________
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users
