RE: [tcpm] New I-D (draft-mahesh-persist-timeout-00.txt)

At 05:35 a.m. 14/02/2007, Anantha Ramaiah \(ananth\) wrote:

>Yep, on a related note, some TCP stacks shrink windows, window shrinking
>is used as a convenience tool in some cases :-) We probably need to deal
>with such cases as well, since the robustness principle allows it.

The amount of data queued for transmission, is another factor worth 
considering in the heuristics, as for DoS attacks there will usually 
be a large amount of queued data (send buffer full, or close to).

Also, if the corresponding socket has been close()ed by the 
application running at the TCP sender, the application will usually 
have no control on the socket anymore (unless the SO_LINGER (?) 
socket option was set?). This may be worth noting in the doc, too.

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

_______________________________________________
tcpm mailing list
tcpm@ietf.org
https://www1.ietf.org/mailman/listinfo/tcpm