Re: IIS SMTP server in a redundant formation?

Tech-Archive recommends: Fix windows errors by optimizing your registry

Ok. I'll give you that. Though if something needs to be disabled,
there by indicating that it is enabled by default, I think it is safe to assume that the very large majority of systems will have it enabled, thus flying in the face of the spirit of the RFC.

It is incorrect to say that using the OS resolver library means that
the MX algorithm is not being followed. They are two different
components. If the resolver returns an RRset to the SMTP sender --
*regardless of whether it fetches the RRset from its internal cache or
passes it through from a remote server* -- the SMTP sender MUST
randomize the MX records in that RRset. And that is what happens in
the real world, enough to call it a rule.

Client-side caches break RFCs and should not be treated as standard components of an SMTP delivery architecture.

Possibly. Still a reality verses dream issue.

Please provide evidence that the use of resolver caches has disrupted
the operation of the worldwide SMTP architecture to the point that the
MX randomization algorithm is a "dream"!

Furthermore, we are talking in the OP's case about someone with
complete control over the SMTP sender. It is self-evident that the
resolver cache is not a fixed part of the configuration under
discussion.

--Sandy




------------------------------------
Sanford Whiteman, Chief Technologist
Broadleaf Systems, a division of
Cypress Integrated Systems, Inc.
------------------------------------
.