On 11/16/2012 01:47 PM, Per-Erik Persson wrote:
On 11/15/2012 10:24 PM, John Hardin wrote:
On Thu, 15 Nov 2012, Per-Erik Persson wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Is there a way to add spamassin rules without editing the config and
reloading the process?
To be more specific, I can set up a RBL of my own and add suspicious
servers found in the header, no problems to do that.
You can also set up your own local URIBL as well. It's all just DNS.
Yes that was my intention since it is so simple if you already have to
tools to manage a large scale network.
But if I want to insert a rule to shoot down a malicious link to a
phishingsite listed in the body of the email I need to do the following
steps.
The caveat with a private URIBL is it's only on the domain name. If
you want to poison a specific page on an otherwise acceptable host a
local URIBL won't help you.
Excactly, I was hoping for a solution to that problem. A way to use an
URIBL to match evil httplinks in the body.
Or if that is not possible I will have to go with the other solution:
Write a fairly simple interface to build the ruleset and insert it into
cvs(or whatever) and on the serverside poll the cvs server for changes,
testdrive them and after that reload amavis, on failure back down to the
latest working change.
You may want to look into ClamAV's sigtool.
