At 21:55 08/09/2006, Jim Shankland wrote:

Travis Hassloch <travis.hassloch@rackspace.com> writes:

...

The part where it becomes a DoS is when they tie up all the listeners on a socket (e.g. apache), and nothing happens for several minutes until their connections time out. Whether intentional or not, it does have a negative effect.

Ah, that makes sense. I was assuming a deliberate attack, which is not actually implicit in the term "DoS". A deliberate denial of service is not made easier by shrinking the window. But an implementation that advertises a 0 window in lieu of sending FIN or RST can certainly deny service inadvertently by tying up resources that should have been freed.

FYI, this issue was raised at the IETF TCPM WG mailing-list a month ago or so. The OP argued to reduce the amount of time for which a peer could advertise a 0 window. However, the problem is that if the goalis to perform a DoS attack, the attacker could advertise a 1-byte window (or ay other small window). Or he could advertise a 0-window for some time (less than the "threshold" the OP proposed), then increase the window to, say, one segment, and then go back to advertising a 0 window. The OP had suggested seeing this behaviour tying up all system resources, hence leading to the attacked system to not be able to service legitimate systems. There seemed to be agreement as at the TCPM WG that yu should handle these scenarios at the application layer. Kindest regards, -- Fernando Gont e-mail: fernando@gont.com.ar || fgont@acm.org PGP Fingerprint: 7809 84F5 322E 45C7 F1C9 3945 96EE A9EF D076 FFF1