I came across PLAIN authentication in a FileMaker Pro plugin once, and that one could also do the LOGIN authentication - so it worked.

In this case, the point was to chech the session log, to see why the authentication failed - and if it is using PLAIN, then change the client behaviour if possible.

Hi,

We're running the las version of Mercury Mail in Windows 2003 Server

When we try to send a message through Mercury with Microsoft

Entourage 2004 using SMTP authentication, it fails with the following

error message:

quote:

---

Authentication failed because Entourage doesn't support any of the available authentication methods.

---

Any ideas on what's the issue and how to fix it?

Thanks.

[quote user="jatubio"]Hi,

We're running the las version of Mercury Mail in Windows 2003 Server

When we try to send a message through Mercury with Microsoft

Entourage 2004 using SMTP authentication, it fails with the following

error message:

quote:

---

Authentication failed because Entourage doesn't support any of the available authentication methods.

---

Any ideas on what's the issue and how to fix it?

Thanks.[/quote]

 David Harris has implemented ESMTP AUTH CRAM-MD5 for WinPMail, Mercury and Mercury/32.  They do a SMTP authentication in accordance with RFCs 2554 and 2195.  Mercury and Mercury/32 also do the LOGIN and the strange MS AUTH=LOGIN.

Now the question is what does Entourage support.  You can use session logging to see exactly what's happening though.

[quote user="jatubio"]Any ideas on what's the issue and how to fix it?[/quote]

Enable session logging in MercuryS.

If Entourage is using AUTH=PLAIN, you need to change to AUTH=LOGIN

MercuryS does not support AUTH=PLAIN as specified by RFC 2554 (http://community.pmail.com/forums/thread/5343.aspx)

[quote user="Peter Strömblad"]

MercuryS does not support AUTH=PLAIN as specified by RFC 2554 (http://community.pmail.com/forums/thread/5343.aspx)

[/quote]

I just want to correct any perception that might arise from Peter's comment that Mercury is in any way non-standards-compliant here... SASL methods such as PLAIN and LOGIN are distinctly optional, and in fact are pretty heavily deprecated by the IETF in any case. At this point, I have not opted to support PLAIN, nor is there any specific standards-based requirement that I should do so. That said, I'll consider it as part of the v4.53 development process; it's a woeful authenticator, but it's no worse than LOGIN, which every copy of Outlook in the world seems to have to use. *sigh*.

Cheers!

-- David --