Can an IPS system do this?

Apr 26, 2007 9 Replies

Hi everybody,



As we all know, if you have a normal firewall that allows certain traffic through to a public server then the firewall doesn't provide any protection for the server on those ports. For example, it doesn't realise that the same external IP address has been hammering away at the server for the past 3 hours trying to guess a valid username and password combination.



Does anyone know of a product that can add extra functionaility to a firewall, or even replace the firewall, so that attacks like this can be automatically caught and the traffic blocked? A cisco engineer I know said that an IPS system is unlikely to be able to pick up this behaviour as suspicious, is he right?



We have a basic budget of 5000 Euros to replace or augment our firewall, specifically to mitigate brute force attacks like this. Current firewall is a Cisco PIX 515E. I was thinking of maybe a Cisco ASA5510 with some add-on module or other, but if it won't help,...



Any help is most appreciated.



Who cares?

But you realize that this is a very very very stupid idea?

This sort of thing?

ftp://shorewall.net/pub/shorewall/contrib/PortsentryHOWTO.txt

Jim Ford

Trust me, you DON'T want any firewall to automatically create new blocking rules.

What do you do if somebody sends spoofed packets at your firewall causing it to automatically block traffic to/from some important server?

Juergen Nieveler

*sigh*

When will people learn that automatic network shunning is a REALLY BAD IDEA? Rate-limiting is a much better way to deal with this kind of problem. If you can't avoid using passwords in the first place.

cu

59cobalt

Instantly lose connectivity?

cu

59-That was easy!-cobalt

Yep, got the message loud and clear... I'll spend the effort instead ensuring the servers and apps are fully patched and tied down as much as possible.

Thanks to all.

Hm... what about an IDS? After all, just because some companies think it's funny to add a shoot-yourself-in-the-foot extension doesn't mean that the idea of detecting suspicious behaviour would be a bad idea.

Yep. An IDS sensor on a mirrored port (or hub port - most companies don`t have THAT fast Internet connections anyway) will at least tell you HOW somebody attacked your machine, which is a great help.

Juergen Nieveler

You could keep your PIX, build an IPS/IDS from some FOSS and be almost as secure as some banks... only difference, banks have managed IPS/ IDS, yours wouldn't be as much....

RedForeman

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required