Re: [offtopic] Are 8-bit characters completely illegal in a raw message?

[Enrico Scholz <en...@sigma-chemnitz.de>]

vitas1@itera.ru writes:

> So, as I've found in RFC's all header fields in message should
> be encoded to 7-bit data.

s/should/must/


> In addition my SMTP server does *not* support 8-bit MIME for
> incoming e-mail.

That's very uncommon and lot of mail will be probably rejected
due to this.


> The question is in subject - or am I missed some *legal* usage
> of 8-bit characters (maybe some kind of comments, optional
> fields, etc)?

Without extensions (RFC 1652), 8-bit in SMTP body is illegal.


> Of course, the goal is to write the rule for SA that will
> trigger on 8-bit symbols in raw ("raw"="what I've seen in
> tcpdump output for this message") message.

fwiw, sendmail has

| O EightBitMode=strict

rule. See

        https://www.cvg.de/people/ensc/sendmail-8hdr.txt

for a rule to check headers.



Enrico

Re: [offtopic] Are 8-bit characters completely illegal in a raw message?

[Enrico Scholz <en...@sigma-chemnitz.de>]

Matus UHLAR - fantomas <uh...@fantomas.sk> writes:

>> > In addition my SMTP server does *not* support 8-bit MIME for
>> > incoming e-mail.
>>
>> That's very uncommon and lot of mail will be probably rejected
>> due to this.
>
> are there known problems with mailers that can send/receive
> 8-bit but can't encode to QP or base64?

exim causes such problems; it is not MIME aware and won't be able
to make such a conversion. So it will either violate RFC 2821
(which is done already by exim when generating bounces, btw), or
it will abort delivery when it does not see an '250-8BITMIME'
EHLO response.


Enrico

Re: [offtopic] Are 8-bit characters completely illegal in a raw message?

[Matus UHLAR - fantomas <uh...@fantomas.sk>]

> vitas1@itera.ru writes:
> > In addition my SMTP server does *not* support 8-bit MIME for
> > incoming e-mail.

On 01.04.08 10:52, Enrico Scholz wrote:
> That's very uncommon and lot of mail will be probably rejected
> due to this.

are there known problems with mailers that can send/receive 8-bit but can't
encode to QP or base64?
-- 
Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Fucking windows! Bring Bill Gates! (Southpark the movie)

Re: Re: [offtopic] Are 8-bit characters completely illegal in a raw message?

[vi...@itera.ru]

Thanks for all answers.

Matus UHLAR - fantomas <uh...@fantomas.sk> wrote on 01.04.2008 18:11:03:

> > vitas1@itera.ru writes:
> > > In addition my SMTP server does *not* support 8-bit MIME for
> > > incoming e-mail.
> 
> On 01.04.08 10:52, Enrico Scholz wrote:
> > That's very uncommon and lot of mail will be probably rejected
> > due to this.
> 
> are there known problems with mailers that can send/receive 8-bit but 
can't
> encode to QP or base64?

My configuration is - Exim as a front-end (filtering) server for incoming 
e-mail and Lotus Domino as a back-end.
Both don't support 8BITMIME. And I haven't any problems with 8-bit bodies 
- Exim just passes them to back-end an then they are delivered to users in 
correct/readable format. 

The only problem is that such e-mail is usually spam :-)

Well I've found another interesting thing (sorry for offtopic again :-). 

RFC2045 defines a term "preamble". Usually it looks like a phrase "This is 
a multipart message..." in raw message and MIME-compatible software should 
just ignore it. 

According to RFC2045 and RFC2822 (as I understand them properly) preamble 
MUST consist of US-ASCII characters.
But I see many 8-bit preambles in *ham* e-mail. I think this is because of 
badly localized software at sending side. So *be careful* if you want to 
score 8-bit characters - you can get some false positives!

Thanks again to all.

Vitas.