Henrik K wrote: > On Wed, Apr 08, 2009 at 02:43:00PM -0700, Bill Landry wrote: >> Michael Scheidell wrote: >>> and it looks like sane security test #2 and 3 did fail if I don't do >>> this in amavisd.conf: >>> (uncomment out the qr'^MAIL'. >>> >>> so, 'can be slow'. how slow is it? and is bill landry wrong saying I >>> need bypass-decode_parts=1? >>> is this something fixed in 2.6.2? >> Not required, but if you don't use "bypass-decode_parts = 1", then you >> will want to set "$keep_decoded_original_re = 1" so that clamd can also >> scan the entire message and not just all of the decoded parts separately. > > I was just thinking about this same stuff. I use bounce killer too, but I > want scanners to scan _only_ the full email and not all the parts which > creates unnecessary load. > > Mark, would it be possible to have some option to only do the stuff that > bounce killer needs and not write unnecessary parts to disk?
Amavisd-New already does this, just set "bypass-decode_parts = 1" in your amavisd.conf file, and your good to go (works with amavisd-new 2.5.1-pre1 and newer). When this is set, amavisd does *not* decode messages and split them into parts, it simply passes the full message to clamd for scanning. Bill ------------------------------------------------------------------------------ This SF.net email is sponsored by: High Quality Requirements in a Collaborative Environment. Download a free trial of Rational Requirements Composer Now! http://p.sf.net/sfu/www-ibm-com _______________________________________________ AMaViS-user mailing list AMaViS-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/amavis-user AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3 AMaViS-HowTos:http://www.amavis.org/howto/