Re: [exim] Hotmail

Top Page
Delete this message
Reply to this message
Author: Ian Eiloart
Date:  
To: Mike Cardwell, Exim Mailing List
Subject: Re: [exim] Hotmail


--On 28 January 2008 16:26:57 +0000 Mike Cardwell
<exim-users@???> wrote:

> Ian Eiloart wrote:
>
>>> The "headers_remove = User-Agent" line is something entirely different
>>> that tends to help. You'll find many references on the web to hotmail
>>> blocking certain messages that contain Thunderbird in the User-Agent
>>> header, but allowing through messages that are exactly the same, but
>>> without the User-Agent header. I tested this myself a while back and it
>>> was true.
>>
>> Why would an MUA add a "User-Agent" header? It's an HTTP or net-news
>> header, not a mail header.
>>     <http://www.iana.org/assignments/message-headers/perm-headers.html>
>> Arguably, Microsoft are doing the right thing by punishing clients for
>> using non-standard headers.

>
> Non-standard headers? You can add *any* arbitrarily named header you
> want to an email. At least Thunderbird and Mutt both use "User-Agent".
> I've not tested other MUAs. Microsoft aren't, "doing the right thing,"
> or anything even close to sensible by scoring so harshly on this header.


RFC2822 does say that you can add other headers, but not arbitrarily named
- the name can't clash with a registered name. However, that doesn't mean
that every imaginable header is "Standard". Indeed rfc2076 lists several
headers as "not internet standard". I'd suggest that a "standard" header is
one that's registered with IANA according to rfc3864:
<http://tools.ietf.org/html/rfc3864>. I guess, for clarity, I should have
used the phrase "registered".

I suspect that Microsoft are adding spam points to messages with
non-standard headers, which would explain why some messages are acceptable
when they don't contain a user-agent header.

>> Perhaps the sensible thing to do is to replace the User-Agent header
>> with an X-mailer: header
>>
>> Or, perhaps someone should register user-agent as a mail header.
>> <http://tools.ietf.org/html/rfc3864> says how, and says that part of the
>> point is "encouraging convergence of header field name usage across
>> multiple applications and protocols"
>
> Or perhaps Microsoft should drop SmartScreen and use a decent filtering
> technology.
>
> Mike




--
Ian Eiloart
IT Services, University of Sussex
x3148