*BSD News Article 66983


Return to BSD News archive

Newsgroups: comp.unix.bsd.freebsd.misc
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!inquo!in-news.erinet.com!imci5!pull-feed.internetmci.com!news.internetMCI.com!newsfeed.internetmci.com!howland.reston.ans.net!psinntp!psinntp!psinntp!psinntp!jadpc!jdeitch
From: jdeitch@jadpc.jd.com (Jim Deitch)
Subject: Re: ed0: device timeout (help)
Message-ID: <DqED41.Fsn@jadpc.jd.com>
Organization: J. Deitch & Associates, San Diego, CA.
References: <4l695g$sup@river.biddeford.com> <Pine.BSF.3.91.960419105319.12453A-100000@enteract.com> <4ldmsn$16ku@pulp.ucs.ualberta.ca>
Date: Thu, 25 Apr 1996 02:48:01 GMT
Lines: 69

In article <4ldmsn$16ku@pulp.ucs.ualberta.ca>,
Jason George <george@ee.ualberta.ca> wrote:
>Kevin Dulzo (kdulzo@enteract.com) wrote:
>
>
>> On 18 Apr 1996, george p swanton wrote:
>
>
>> [unecessary info cut]
>
>> > ed0: flags 8863<up,br,not,ru,si,mul (abreviated)> MTU 1500
>> >  inet 204.120.68.11 netmask 0xffffff00 broadcast 204.120.68.255
>> >   ether 00:00:C0:0f:26:43
>> > lo0: flags 8009<up,loop,mul> MTU 16384
>> >    inet 127.0.0.1 netmask 0xff000000
>> > add net 224.0.0.0: gateway unix1.hve.com
>> > writeing to routing socket: file exists
>> > add host unix1.hve.com: gateway localhost: file exists
>> > starting routing deamons: routed ed0: device timeout
>> >                                  ^^^^^^^^^^^^^^^^^^^^
>> > 
>> 	It's been awhile so if I'm wrong someone correct me.
>
>> Your device doesn't have any conflicts especially if 'ifconfig' was able 
>> to up the interface (i.e. ed0:...<up,...> netinfo...).  but the way to 
>> verify is to watch your boot sequence and/or use 'dmesg' to verify the 
>> probe results.
>
>> 	But "ed0:devidce timeout" in my experience, assuming the device is 
>> ok at boot, means it cannot access the network (no connection/bad 
>> connection).  If the probe is ok -- check the cables/ports.
>
>
>I have had _many_ device timeouts from _many_ different network cards.
>The only thing these network cards (a selection of Intel, SMC and NE2000
>clones) had in common was that they were at least 18 months old.  Some
>were close to 5 years old.  FreeBSD would give me a timeout error
>everytime, but Linux would work fine _with all of them_.  At first I
>thought it was something to due with the fact that I'd put these old,
>slow cards in a hybrid ISA/PCI-bussed computer and that it was actually
>a bus problem.  Well, Linux proved that wrong.
>
>From the fact that Linux was able to get these cards to work properly, I
>would suggest that there is a small glitch somewhere in the FreeBSD
>low-level drivers supporting the various hardware compnents.  A timing
>problem?  I remember that the Linux crowd has a very difficult time
>originally getting the Intel EE/16 NIC working properly and it turned
>out to be a small timing error in the driver.  This caused network
>operations to be flaky or non-existant.  I currently have an EE/16 that
>works intermittantly with FreeBSD but works fine with Linux (on a
>vanilla ISA bus).
>
>
>Comments?
>
>
>--Jason
>george@ee.ualberta.ca
>jbg@specialty.ab.ca

I agree with the network not connected/termiated properly.  I have
used FreeBSD since it was BSD386 0.0 with original WD cards.  The only
time I saw this was when there was a physical problem with the
network.

Jim
-- 
INTERNET:   jdeitch@jadpc.jd.com
UUCP:	    cg57!jadpc!jdeitch