Re: IDS and SSL

From: Jon (vandiveeat_private)
Date: Fri Mar 22 2002 - 20:47:15 PST

  • Next message: Sverre H. Huseby: "Re: CSS implication"

    Cisco has a new CSS/SSL product 30-90 days out, whereby the SSL card will be
    insertable to the CSS chassis.
    looks pretty cool.....
    
    Jon
    
    ----- Original Message -----
    From: "Dom De Vitto" <Domat_private>
    To: <jlewisat_private>; "'Oliver Petruzel'" <opetruzelat_private>;
    "'zeno'" <bugtraqat_private>; <vuln-devat_private>;
    <bugtraqat_private>; <webappsecat_private>;
    <focus-idsat_private>
    Sent: Friday, March 22, 2002 2:45 AM
    Subject: RE: IDS and SSL
    
    
    > Ditto, for Cisco CSS 11000's
    >
    > They'll give you multi-site loadbalancing too...
    >
    > Dom
    >  |-----Original Message-----
    >  |From: Jason Lewis [mailto:jlewisat_private]
    >  |Sent: Thursday, March 21, 2002 8:17 PM
    >  |To: 'Oliver Petruzel'; 'zeno'; vuln-devat_private;
    >  |bugtraqat_private; webappsecat_private;
    >  |focus-idsat_private
    >  |Subject: RE: IDS and SSL
    >  |
    >  |
    >  |C'mon Ollie, I am doing this now.  Instead of buying
    >  |encryption cards for all my webservers, we threw a couple of
    >  |Alteon iSD SSL accelerators onto our Alteon switches.
    >   ||http://www.nortelnetworks.com/products/01/alteon/isdssl/index.
    >  |html
    >  |
    >  |These offload encryption and allow me to drop a NIDS next to
    >  |the webservers, where all the traffic is un-encrypted.  I
    >  |already had the Alteon infrastructure, and the iSD's won't
    >  |work without them so YMMV.
    >  |
    >  |Granted, eventually we will see congestion, but the
    >  |scalability of the SSL accelerators and the Alteons will make
    >  |that a long range problem.  I think the iSD's an scale to 256
    >  |with the Alteon's distributing the load.  Not to mention I
    >  |save my webserver processing power for serving page not
    >  |encyrption....different discussion though.
    >  |
    >  |Good network design will avoid those traffic problems.  If I
    >  |have that much traffic into one datacenter, it is time to go global.
    >  |
    >  |Now, that isn't an excuse for NIDS.  I like HIDS for the
    >  |drill down on each box.  I think the two can co-exist.  I
    >  |like seeing what is on the wire, not just what made it to each server.
    >  |
    >  |Jason Lewis
    >  |http://www.packetnexus.com
    >  |It's not secure "Because they told me it was secure".
    >  |The people at the other end of the link know less
    >  |about security than you do. And that's scary.
    >  |
    >  |
    >  |//snip
    >  |Nothing short of a big road-block could monitor encrypted
    >  |traffic prior to a host;  it's just not logically possible to
    >  |examine the encrypted traffic without a big roadblock and
    >  |certificate-sharing nightmare.. that is, on the wire
    >  |atleast... with the exception of placing an IDS -ON- a
    >  |VPN...and that still wont help with SSL specifically, and
    >  |that would require SICK amounts of RAM/power to be anything
    >  |close to efficient... SSL PROXY/IDS system? No way... same
    >  |speed/RAM/bandwidth limitations... //snip
    >  |
    >  |
    >
    



    This archive was generated by hypermail 2b30 : Sat Mar 23 2002 - 09:58:45 PST