[126003] in North American Network Operators' Group
Re: Mail Submission Protocol
daemon@ATHENA.MIT.EDU (Tony Finch)
Wed Apr 28 03:20:14 2010
From: Tony Finch <dot@dotat.at>
To: Jeroen van Aart <jeroen@mompl.net>
In-Reply-To: <4BD7879D.3090208@mompl.net>
Date: Wed, 28 Apr 2010 08:19:38 +0100
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Happily Microsoft have fixed their smtps stupidity, so you only need
to support it on the server if you need to support users running old
versions of Outlook etc. There was never anything particularly wrong
with smtps, apart from a dogma in the IETF that it is architecturally
wrong. The consensus now is that it was wrong to rescind the port
allocation, because that completely failed to stop people (er,
Microsoft) from deploying smtps, and just led to interop problems.
Tony (on his iPod).
--
f.anthony.n.finch <dot@dotat.at> http://dotat.at/
On 28 Apr 2010, at 01:55, Jeroen van Aart <jeroen@mompl.net> wrote:
> Raoul Bhatia [IPAX] wrote:
> > i recently had the problem that an lotus notes server insisted on
>> sending emails to one of our clients via port 465. so having
>> mandatory
>> authentication there actually broke delivery for an exchange sender.
>
> Leave it "broken" for the other end that is. Only way to force them
> to fix it.
>
> The only acceptable, and standard, way to submit email these days is
> using port 587 with TLS. And if you have users with broken clients,
> they can use webmail behind https. I am against facilitating (and
> thus perpetuating the existence of) old broken clients by making
> available port 465.
>
> Regards,
> Jeroen
>
> --
> http://goldmark.org/jeff/stupid-disclaimers/
>
