Re: Barracuda backdoor

[Marsh Ray <marsh () extendedsubset com>]

On 04/28/2011 05:51 AM, Tõnu Samuel wrote:
> On Thu, 2011-04-28 at 11:45 +0100, Benji wrote:
> > Do you actually have any evidence of a backdoor? Or could this just be
> > a remote 'turn-off' switch as such? I'm not saying that one is better
> > than the other, but they are very different features.
>
> I have no idea how this technically is implemented or what they can do
> else. This is clear example of closed source product dangers. Today we
> found some "switch off", tomorrow what?

Tomorrow Barracuda gets pwned and this turns into a cascade failure.

Oh wait, that happened two weeks ago:
   http://www.theregister.co.uk/2011/04/11/barracuda_networks_attack/

> How we can be sure about
> anything? Only thing I am sure now: they kept copy of keys to house you
> bought from them years ago and their used those keys for illegal thing.

Let's be careful though: just because your system stopped working 
doesn't mean it has a backdoor. It could have been implemented as simply 
a periodic "phone home for updates" which received some type of 
"license expired" message. A remote kill switch, for sure, but not 
necessarily the same as a back door.

It raises the question though of how many companies have that particular 
combination of ethics and self-discipline to implement one and not the 
other. It sometimes takes extra work to build a product that performs 
security functions in a customer's network without granting yourself 
unnecessary privilege on that network.

As we saw with RSA SecurID, many admins didn't realize that the vendor 
might be keeping a copy of the keys. Sites with products on their 
networks may want to consider if Barracuda as an external vendor falls 
under the scope of their PCI requirements.

- Marsh

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/