Re: Correct value for EHLO when submitting mail

Hector Santos <[email protected]> Wed, 29 Jan 2014 23:46:09 -0500
Newsgroups gmane.ietf.rfc822
Organization Santronics Software, Inc.
Message-ID <[email protected]>
On 1/29/2014 4:23 PM, Patrick Ben Koetter wrote:

>> What is the current best practice?
>
> Dunno if this is common best practice, but here's my rationale:
>
> In MTA to MTA communication on port 25 it makes sense to check for reasonable
> HELO hostnames:
>
> - not your own
> - not from a private network you can't connect back (bounce)
>
> In MUA to MTA communication on port 587 it doesn't make sense to check for or
> enforce HELO hostnames:
>
> - the underlying OS often does not return a valid HELO hostname (not a FQDN)
> - the identity sent during AUTH creates the trust required to accept and
>    transport the message
>
> p@rick

I agree with the relaxation. Its what our MSA does.

Port 587 does raise the (authentication/authorization) bar. It does 
make sense, just not reliable in practice to enforce as it always was 
the case with port 25.   Port 587 assumes updated compliant ESMTP AUTH 
clients and even if it follows RFC 5821 with using a square bracketed 
IP literal, it could still be wrong (private machine IP vs public NAT 
ip address).

-- 
HLS


_______________________________________________
ietf-822 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-822