[En-Nut-Discussion] TCP/IP Stack crashes

[Michael Jones]

Hello,

Just to add my ongoing observations on this topic as I've been looking for
the cause of this effect for quite a while: 

I personally believe that this must be a result of some race condition in
the tcp state machine were only the "watched" tread gets trapped in a "wait
for event" (I had already found one some time ago). Here I have mainly
observed this when trying to establish a connection - mostly outgoing in my
case. Interesting enough a second tcp connection will remain functional (as
well as UDP and ICMP) which again would indicate problems in the state
machine. Sadly this problem has been haunting me for quite a while and I
have not been able to locate the cause. 

Yet saying that, I will have to intensify my efforts on this as one customer
of ours is "not happy" about the side effects :-)

So let's say: the race is on - who finds the bug may keep it and the honor!


Michael


-----Original Message-----
From: en-nut-discussion-bounces at egnite.de
[mailto:en-nut-discussion-bounces at egnite.de] On Behalf Of Harald Kipp
Sent: Monday, January 16, 2006 8:21 PM
To: Ethernut User Chat (English)
Subject: Re: [En-Nut-Discussion] TCP/IP Stack crashes

Not that easy. On Windows the winpcap library allows you to assemble
raw packets, so you can simulate many hosts.

Harald

At 17:04 13.01.2006 +0100, you wrote:
> > Together with a customer we will set up another long term stress test
> > shortly. I'll try to add some broadcast showers.
>Do you have an idea how to setup such stress test environment at home?

_______________________________________________
En-Nut-Discussion mailing list
En-Nut-Discussion at egnite.de
http://www.egnite.de/mailman/listinfo.cgi/en-nut-discussion