UNIX hosts and MX "best practice"

[Kevin Darcy]

Barry Margolin wrote:

> In article <38923A98.4F126B1B at daimlerchrysler.com>,
> Kevin Darcy  <kcd at daimlerchrysler.com> wrote:
> >Newer sendmails use ANY queries to avoid this performance problem.
>
> I've heard this said many times, but I don't understand how it can work
> reliably.  How do they deal with the fact that a caching server responds to
> ANY queries only with the data it happens to have cached, so the response
> might not be complete?

Further investigation reveals that sendmail only uses ANY queries for
*some* purposes (canonicalizing host names), but if the ANY query fails to turn up
any MX records, it'll still do an explicit MX query when it comes time to actually
make delivery (assuming, of course, that it's configured to use MX-based routing
at all).

So I guess it's still worthwhile to generate those matching MX records...


- Kevin