Hey everyone,
First of all, I'm really getting a kick out of DSPAM. It's getting
progressively better as training goes on (approximately 1,000 messages
have been processed thus far), and should prove to be a highly effective
replacement for my aging MailScanner/SpamAssassin implementation of yore.
At this point, I am using it with Postfix 2.3.3 on RHEL5 in a
MySQL-driven virtual environment (given the shortcomings of virtually
every virtual mail system I saw for Postfix, I went ahead and crafted
one myself). The WebUI is for the most part working properly, running
under suexec as the dspam:dspam user with authentication provided by
mod_auth_mysql against my mail system database. The dspam daemon itself
is running as root:mail, and in order to allow things to work properly
the dspam user was added to the mail group as a supplementary group.
At this time, the quarantine is indeed trapping messages, however, when
I attempt to view them the message field is entirely blank. Running
stack traces on the Apache child responsible for processing the request
is showing the message is indeed being read in to the dspam.cgi script,
however, it is never being output into the resulting data sent to the
end user. The strace output can be found here:
http://pastebin.org/24577
For the sake of safety, my e-mail address was replaced with
[EMAIL PROTECTED] in the output above. The dspam.cgi app is able to read
my user's quarantine file, but once it has opened, read, and closed the
mbox, no data is ever output to the browser. I'm pretty sure the issue
affecting this is directly related to the fact that messages in the
quarantine cannot be released, nor can they be individually deleted (you
have to delete them all at once). Because of these issues, the
quarantine is pretty much useless to me at this point in time.
Let me know your thoughts - I get the feeling this is something simple
I'm overlooking, and would really appreciate any help you could give
me. Thanks in advance!
Regards,
Campbell Krueger
