Re: distributed attack, high or not

["Steven M. Bellovin" <smb () research att com>]

In message <20020131025142.A12260 () monet titania net>, "Joseph T. Klein" writes:
> I define it as random because the traffic rise could be seen
> coming in from multiple providers and looked to be the same
> percent from all sources (separate routers with separate
> interfaces to separate ASNs in separate geographic locations).
> The traffic was inbound and not backsplash from randomized
> source addresses.
>
> It looks to me like a infection with someone turning a control
> knob. Is this common or a precusor of a bad thing?
It's a classic DDoS attack, aimed at you.  Someone has lots of zombie 
machines out there; at some point, they sent a command packet to all of 
them, saying "bombard such-and-such an IP address for 3600 seconds".

Common?  It happens frequently to someone.  Precursor?  Entirely 
possible, though there's no way to know for sure.  But it can be very 
bad -- see http://news.zdnet.co.uk/story/0,,t269-s2103098,00.html
for what happened to a British ISP.

                --Steve Bellovin, http://www.research.att.com/~smb
                Full text of "Firewalls" book now at http://www.wilyhacker.com