Thanks for the response, all. I'm not arguing about the encryption method or 
strength - simply stating that I don't understand what is taking place there. 
It seems that the shared keys are used to authenticate whether or not an OSSEC 
agent is authorized to communicate with the server. That makes sense. My 
question is whether or not the actual payload is encrypted in transit, which 
you all also seem to be pointing to the answer as being: yes


From: [email protected] [mailto:[email protected]] On 
Behalf Of loyd.darby
Sent: Monday, December 20, 2010 4:12 PM
To: [email protected]
Subject: Re: [ossec-list] Securely deploying OSSEC

this is a little dated, but the point is...
 http://www.marktaw.com/technology/HowlongdoesittaketocrackS.html

On 12/20/2010 04:07 PM, Chuck (MdMonk) wrote:
How about saying it's "astronomically improbable." :)

-Chuck (MdMonk)
On Mon, Dec 20, 2010 at 1:58 PM, Erik 
<[email protected]<mailto:[email protected]>> wrote:
Hello,

Technically traffic can be sniffed yes but it would require

1) allot of cpu power and memory
2) heaps (tons of heaps) of patience

to actually "decrypt" the traffic depending on the encryption algoritm used by 
ossec

it is "near to impossible" offcource 90% is not 100%

Op 20/12/2010 21:27, loyd.darby schreef:

The traffic is encrypted but if someone can record the communication, they have 
essentially forever to hack at it until it breaks.
You really don't want all your remote clients connecting to a local server.  
That would be sending way more traffic than actually matters to you.
What I think you want is ossec server preprocessing the events and generating 
alerts, and possibly forwarding only some of those.
You could scp to the remote host and fetch the alerts on a schedule or overlay 
encrypted attachments to email.  If you want to then re-merge and correlate all 
those events, you might look at a limited deployment of OSSIM SIEM.


On 12/20/2010 02:02 PM, dan (ddp) wrote:
On Mon, Dec 20, 2010 at 1:54 PM, Jarred 
White<[email protected]<mailto:[email protected]>>  wrote:

Hello. I'm trying to find a way to remotely deploy OSSEC to some of our
remote sites and have it report back to us on server health/security. There
is no direct connection to the remote network, so any reporting would need
to happen over the Internet since VPN is out of the question.



Naturally I'm not going to send ossec alerts unencrypted via the Internet.
I've thought about writing some scripts that would keep an stunnel up and
running in order to report back to us, but I'm wondering if there is a
better way.



I did see this on the list archives, dated 9/21/06:



Ossec uses blowfish (192 bits) for the agent/server communication channel

and md5+sha1 combined for the integrity verification.



I reviewed a presentation put on by Daniel and while it mentions the use of
pre-shared keys, I'm interested in understanding a little bit more about how
the authentication/security mechanism works. My guess is that the UDP
traffic could be sniffed, but I'm just not sure and with my limited
understanding about how it works, am not anxious to send alerts via the
Internet.



Any thoughts?



Thanks,

Jarred
The traffic between agents and the manager are authenticated and
encrypted. I don't have an understanding of the technologies used to
do this though.






--

R. Loyd Darby, OSSIM-OCSE

Project Manager DOC/NOAA/NMFS

Infrastructure coordinator

Southeast Fisheries Science Center

305-361-4297

Reply via email to