[Mimedefang] ANNOUCEMENT: Net::validMX v2.2.0 now available.

[Jeff Rife]

On 2 Jun 2006 at 9:19, Kevin A. McGrail wrote:

> > IMO, any domain that publishes *any* bogus MX record should be rejected.
> 
> However, it is a constant in the code that can be changed easily and
> v2.3.0 will have the option overridable on a call-by-call basis.  So
> we can agree to disagree and the module will be suitable for either
> point of view. 

A "gotcha" on this for MIMEDefang users is that depending on their 
config, they need to make sure they call "check_valid_mx" after any 
tests at the right place compared to tests for known "good" private IP 
blocks.

In other words, if you call:
  check_valid_mx(email=>'test at domain.com',flag_intranets=>'false')

before you check to see if the e-mail is coming from one of your 
private IPs, you might flag it as a "bad MX", when in reality a private 
IP address for an MX might be OK at that point.

I probably screwed up what I'm trying to explain here, but hopefully 
people will get the idea.


--
Jeff Rife |  
          | http://www.nabs.net/Cartoons/FoxTrot/TransporterError.jpg