MailScanner / Clam .. (grrrr)

[Rick Cooper]

 > -----Original Message-----
 > From: mailscanner-bounces at lists.mailscanner.info 
 > [mailto:mailscanner-bounces at lists.mailscanner.info] On 
 > Behalf Of Rick Cooper
 > Sent: Friday, June 15, 2007 7:34 AM
 > To: 'MailScanner discussion'
 > Subject: RE: MailScanner / Clam .. (grrrr)
 > 
 >  
 > 
 >  > -----Original Message-----
 >  > From: mailscanner-bounces at lists.mailscanner.info 
 >  > [mailto:mailscanner-bounces at lists.mailscanner.info] On 
 >  > Behalf Of Randal, Phil
 >  > Sent: Friday, June 15, 2007 7:07 AM
 >  > To: MailScanner discussion
 >  > Subject: RE: MailScanner / Clam .. (grrrr)
 >  > 
 >  > OK, on my test (completely unloaded) CentOS 5 box 
 >  > MailScanner starts all
 >  > its children thus (to nearest 10 seconds, I hate excessive 
 >  > precision):
 >  > 
 >  >   without clamavmodule:  40 seconds
 >  > 
 >  >   with clamavmodule:    190 seconds
 >  > 
 >  > Cheers,
 >  > 
 >  > Phil
 >  > 
 > 
 > I downloaded
 > http://freshmeat.net/redir/clamav/38302/url_tgz/clamav-0.91rc
 > 1.tar.gz and
 > the difference in load time is huge. From nearly 30 seconds 
 > to about 2
 > seconds. I am timing how long it takes clamd to create and 
 > listen to it's
 > socket, which it does after it loads it's DBs (of course). 
 > This should
 > translate directly to MailScanner and ClamAVModule, as well 
 > as clamscan
 > (since the DB load time per file is the slowest part of 
 > clamscan). Given
 > your times it sounds like you must be running about 5 children?
 > 
 > Rick
 > 
 > 
 > --

I put that clamscan theory to the test and found:
(Scanning same eicar.rar file rounding to nearest second)

Version 0.90.3 : average time 23 seconds
Version 0.91rc1: average time 3 seconds

So I would think 0.91rc1 would be a worth using

Rick


--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.