Re: [Exim] Virus scanning architecture

Top Pagina
Delete this message
Reply to this message
Auteur: Dr Andrew C Aitchison
Datum:  
Aan: Elena Blanco
CC: exim-users
Onderwerp: Re: [Exim] Virus scanning architecture
On Wed, 3 Jan 2001, Elena Blanco wrote:

> control, how do I get the virus-scanning box to send the message back
> to the last server that appears in the received headers so that that
> host is responsible for spooling and delivering the clean message?


I suppose you could pass the mail to the virus-scanner as batch smtp,
and then inject it into the appropriate machine in the first
"Received: from" line. I have not experience in this area.

However, while I can see that you would want the last host to take the
load off your back, I wouldn't bother. I would just accept that by
providing this service I had become a smart host, and process the
mail as normal.

For incoming mail passing it back to the last server would be fine,
but sooner or later one of your clients will pass outgoing mail
through you. (Sooner I'd have thought - since the mail servers are
unlikely to determined whether or not it is local until it has been
checked for virii).

I can see PC users wishing to use your service as their smtp host,
but since they don't typical run mail servers you can't get the mail back
to them. Is there a good reason other than load not to use this
opportunity to stop random Oxford PCs spreading virii ?

For Oxford -> Oxford mail, would it not be more efficient to pass the mail
on than passing it back ? If both ends use the same header technique,
that might even save you from having to check the mail twice.

Another idea would be that mail servers who wish to use your service
make you their MX thus saving a hop for all incoming mail.
Cambridge do a similar thing for spam blocking.

Will users get confused if they get messages telling them about virii
from a system that is neither theirs nor their correspondents ?

-- 
Dr. Andrew C. Aitchison        Computer Officer, DPMMS, Cambridge
A.C.Aitchison@???    http://www.dpmms.cam.ac.uk/~werdna