Mark Daniel Reidel escribió:
Just out of curiosity: why not use dspamc as a filter for a
pipe-delivery to something like "[...]exim -oMr spam-scanned -bS", but
instead chose this approach? I fail to see the gaining...
The gaining: with a platform of several hundred domains, with a total of
about 20000 mailboxes, is to mantain a pool of MySQL connections, share
system resources with DSPAM in daemon mode, and not to open a connection
to the MySQL server with the data (it is a separate machine) each time.
It seems that I have to provide something in ClientIdent, that is used
to identify DSPAM itself to the MTA, but I have :
2008-03-05 18:10:31 H=(localhost) [127.0.0.1] sender verify fail for
<[EMAIL PROTECTED]>: Unrouteable address
2008-03-05 18:10:31 H=(localhost) [127.0.0.1] F=<[EMAIL PROTECTED]>
rejected RCPT <[EMAIL PROTECTED]>: Sender verify failed
The message "[EMAIL PROTECTED]" should be enough hint I think. It's
unroutable, of course. But without your actual exim-configuration, it's
hard to tell what's really happening, because I'm afraid you might have
changed some email-addresses in your e-mail, so you don't get
spammed ;o) Let's see your config. Or... chose the pipe-approach -
which should be less overhead anyway.
You are right. ClientIdent and ServerIdent are some fixed items I have
to configure in dspam.conf so the client/server model can work.
BUT, it seems that the value of ClientIdent is also used to identify
DSPAM itself when connecting back to Exim in the indicated port.
Are you worked with client/server configuration in DSPAM ?
P.D.: I've have the "standard" standalone configuration of DSPAM working
now. I'm just trying to improve
Regards,
begin:vcard
fn:Manuel Molina Cuberos
n:Molina Cuberos;Manuel
org;quoted-printable:Informaci=C3=B3n Interactiva, S. L.;Sistemas
adr;dom:;;Plaza de Santa Ana, 7;Madrid;;28012
email;internet:[EMAIL PROTECTED]
title:Administrador de Sistemas
tel;work:914952578
x-mozilla-html:FALSE
url:http://www.informacion-interactiva.com
version:2.1
end:vcard
