[35933] in bugtraq

home help back first fref pref prev next nref lref last post

Re[2]: Aladdin response regarding eSafe

daemon@ATHENA.MIT.EDU (3APA3A)
Mon Aug 2 15:22:19 2004

Date: Mon, 2 Aug 2004 02:17:56 +0400
From: 3APA3A <3APA3A@SECURITY.NNOV.RU>
Reply-To: 3APA3A <3APA3A@SECURITY.NNOV.RU>
Message-ID: <642792442.20040802021756@SECURITY.NNOV.RU>
To: Aleksandar Milivojevic <amilivojevic@pbl.ca>
Cc: bugtraq@securityfocus.com
In-Reply-To: <410A5601.1070407@pbl.ca>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Aleksandar Milivojevic,



--Friday, July 30, 2004, 6:06:57 PM, you wrote to bugtraq@securityfocus.com:

>> engines I tested (KAV, ClamAV and others) are file-oriented. It makes
>> it impossible to code good antiviral protection for proxy server with
>> this engines.

AM> Hm.  What about option of sending one byte of data to the client every
AM> minute (with configurable limit that not more than xx% of file can be
AM> transffered before scanning, just in case you stummble accross site that
AM> is actually that slow ;-) ), instead of just feeding him up to 80% of
AM> the file in advance of file being scanned?  For those that prefer a bit
AM> more security over interactivity.  This would prevent client from timing
AM> out, 99.99% (number from the back of my head) of files would take less

There  is a difference between inability to create good protection (both
strong  and  convenient)  and  inability to protect. Of cause, there are
ways  to  implement  protection:  you can check file after each 100KB of
data  (anyway  only first X KB of files are usually checked by antiviral
software)  you  can  send  one  byte  per minute, you can send fake HTML
headers  one  in a minute (they will be ignored), or you can simply show
HTML  page with your own progress bar for large files, you can recognize
file  type and act depending on it. Any of this is "hack", because there
is no standardised way.

-- 
~/ZARAZA
You know my name - look up my number (Beatles)


home help back first fref pref prev next nref lref last post