From: Tom Kovar (no email)
Date: Mon Jan 01 2007 - 13:06:03 EST
Yeah, Magnus, very much right. But not only that: LOGIN apparently has
to be the **FIRST** method proposed by the server - if not, M$ does
neither find nor recognise it.
Tack så mycket,
--- Tom
-----Original Message-----
From:
[mailto:] On Behalf Of Magnus Bäck
Sent: Monday, January 01, 2007 7:00 PM
To:
Subject: Re: SASL vs. M$ Outlook and Outlook Express
On Monday, January 01, 2007 at 14:32 CET,
Rene van Hoek <> wrote:
> I did an telnet to your machine and that seems ok:
>
> Leto:/Volumes renevanhoek$ telnet mail.kovarovi.org 25
> Trying 194.212.102.169...
> Connected to bimbo.kovarovi.org.
> Escape character is '^]'.
> 220 mail.kovarovi.org ESMTP Postfix
> EHLO test.a8.nl
> 250-mail.kovarovi.org
> 250-PIPELINING
> 250-SIZE 10240000
> 250-VRFY
> 250-ETRN
> 250-AUTH PLAIN
> 250-AUTH=PLAIN
> 250-ENHANCEDSTATUSCODES
> 250-8BITMIME
> 250 DSN
No, that's not okay. The Microsoft-style LOGIN mechanism is missing.
More recent Microsoft clients may support the PLAIN mechansim as well,
but since people may be running older software I'd say it's a
requirement to provide both PLAIN and LOGIN.
While the OP is fixing LOGIN, I suggest he fixes support for CRAM-MD5
and DIGEST-MD5 as well so that clients won't be forced to send passwords
in the clear.
[...]
-- Magnus Bäck
|
|
|