Subject: Initial RTO too big in TCP? No PSH flag in FINs?
To: 'tech-net@netbsd.org' <tech-net@netbsd.org>
From: Lipinski, Michal <Michal.Lipinski@intel.com>
List: tech-net
Date: 04/10/2001 13:31:54
Hi, 
I've noticed that the initial RTO (retransmission timeout) in TCP is set to
6 seconds in NetBSD 1.x. Whereas two documents RFC-1122 and RFC-2988, state
that RTO should be initialized to 3 seconds. Why is NetBSD different?

According to RFC-2525, FIN segments should include the PSH flag to ensure
expedited delivery of any pending data at the receiver. NetBSD does not
include PSH flag in FIN segments. Instead it always passes the data to
application without delay. But what if the peer is not a BSD system? 

What are the reasons for NetBSD to behave differently? Should the TCP/IP
implementation in NetBSD be improved?
 
Thanks in advance for any comments,
Michal Lipinski