[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

BOUNCE rtl@rtlinux.org: Approval required: (fwd)



>From owner-rtl Tue Jul 31 02:38:06 2001
Received: from colo.asti-usa.com (IDENT:root@colo.asti-usa.com [205.252.89.99])
	by hq.fsmlabs.com (8.11.2/8.11.2) with ESMTP id f6V8c5729700
	for <rtl@fsmlabs.com>; Tue, 31 Jul 2001 02:38:06 -0600
Received: from lineo.com (hal.uk.zentropix.com [212.74.13.151])
	by colo.asti-usa.com (8.9.3/8.9.3) with ESMTP id EAA14706;
	Tue, 31 Jul 2001 04:46:58 -0400
Sender: ians@colo.asti-usa.com
Message-ID: <3B666F85.1FBBAFFE@lineo.com>
Date: Tue, 31 Jul 2001 09:42:45 +0100
From: Ian Soanes <ians@lineo.com>
Organization: Lineo UK
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-rthal3 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: rtl@fsmlabs.com
CC: "Estabridis, Janet P" <EstabridisJP@navair.navy.mil>
Subject: Re: [rtl] Off Topic - Ethernet releasing port addresses
References: <200107310745.f6V7jUC15256@kanga.hofr.at>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Der Herr Hofrat wrote:
> 
> > I'm using RTLV2.3 and kernel 2.2.14.  I will most likely upgrade
> > everything
> > to the newer versions, but not until the test system is accepted by the
> > customer and that could be 3 months from now.
> >
> > I have miniRTL on 8 embedded machines 2 other Linux PC's and 3 Windows
> > machines on a local network.  I have both real-time modules running on
> > my
> > linux PC's and user space applications.  Specifically, the user space
> > applications send information to the Windows machines via ethernet
> > socket
> > code.  I have a set of defined port numbers that the information uses.
> >
> > My problem is that sometimes when I know I have closed the code out and
> > I
> > restart it I get the message
> >
> > Unable to establish a port Connection:  Address already in use.
> >
> > And sometimes I have to re-boot a linux computer to get rid of this
> > message
> 
> this is happening because one side is not closing the socket cleanly so the
> other side has to listen for the handshake to timeout that is as far as I
> know 240s by default - so for this time the address is realy in use . you
> should never need to reboot though - if you check with netstat you will see
> that the port is in "FIN WAIT" state for quite a while if the other end hangs
> up - but latest after timeout*2 ( 8minutes) you should be able to restart
> everything . and at worst if that does not happen do a network stop/start
> - no need to reboot.
> 
> hofrat
> 

Try the socket option SO_REUSEADDR..

----- End of forwarded message from owner-rtl@fsmlabs.com -----