Enabling bounces

[Julian Field]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



Kai Schaetzl wrote:
> Julian Field wrote on Thu, 24 May 2007 16:06:03 +0100:
>
>   
>> Because it is necessary. If you want to bounce spam, you have to jump 
>> through some hoops.
>>     
>
> I'm not questioning the hoops, I'm wondering if I really understand it right that 
> *two* rulesets are necessary.
>   
Yes.
> For a given message, spam actions has to evaluate to 
>   
>> a result including "bounce", and enable spam bounce has to evaluate to 
>> "yes". The ruleset is necessary because I say it is :-)
>>     
>
> But I don't see how *two* rulesets are supposed to work or are necessary.
> I think it's not clear what my confusion is about:
>
> spam.actions.rules:
> From:   IP  bounce
> FromOrTo:   default     store notify header "X-Spam-Status: Yes"
>
> *and*
>
> bounce.rules;
> From:   IP  yes
> From: default no
>
> is that really what I have to do if I wanted to bounce?
>   
Yes.
>   
>> MailScanner doesn't use the From: address in the headers at all, it only 
>> uses the envelope sender address.
>>     
>
> That's why I wanted to boost spamminess of the messages with Spamassassin which 
> would then enable MailScanner to bounce the messages by detecting them as spam 
> (for sure), but with high scoring spam excluded it's very unreliable as most 
> messages would score high probably even without boosting them up.
>   
A little Custom Function could set '$message->{ishigh} = 0;' if the IP 
address matches, to stop it being treated as high-scoring spam, so you 
could then bounce it.
> I think I could also just forward all messages to a specific address, as the 
> forward action is available for spam and non-spam. But the problem still is the 
> detection. Can I create a rule that detects messages from <> *and* IP? I know 
> that I can "and" rules, but can I detect an empty sender?
>   
I seem to remember you can set the address to /^$/ and it will treat 
that as an empty address. That is a regular expression that will match 
an empty address.
So you can say
From: /^$/ and From: IP yes
in a ruleset and that should do what you want.

>   
>>> Any ideas how I can do it with MailScanner (as obviously what I had in mind 
>>> won't work).
>>>   
>>>       
>> You could do it with a Custom Function tied to "Enable Spam Bounce" 
>> probably. I would need to think a bit about exactly what you would want 
>> to do. You want to spot spam coming from postmaster at hisdomain (in the 
>> From: header) and switch off its $message->{ishigh} flag so it gets 
>> treated as normal scoring spam, at which point you can bounce it. It's 
>> just a problem of working out which config option to attach the Custom 
>> Function to.
>>     
>
> Frankly, it's then probably easier to compile milter-regex as I think it can use 
> header parts as well. I hoped I could do it somehow with MailScanner as I just 
> need the blockage for this one thing.
>
> Kai
>
>   

Jules

- -- 
Julian Field MEng CITP
www.MailScanner.info
Buy the MailScanner book at www.MailScanner.info/store

MailScanner customisation, or any advanced system administration help?
Contact me at Jules at Jules.FM

PGP footprint: EE81 D763 3DB0 0BFD E1DC 7222 11F6 5947 1415 B654
For all your IT requirements visit www.transtec.co.uk



-----BEGIN PGP SIGNATURE-----
Version: PGP Desktop 9.6.1 (Build 1012)
Charset: ISO-8859-1

wj8DBQFGVb4PEfZZRxQVtlQRAmW5AJ41Z6ZH9M5aMJDlZ7ghkYTYGQEBJQCg6oiO
uad6qK3lhlnpiF+7dfXE7MQ=
=eg63
-----END PGP SIGNATURE-----

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
For all your IT requirements visit www.transtec.co.uk