How do I find/determine faulty network hardware?

essdeeay · 07-15-2006, 04:11 AM

Hello,

I'm seeing the following error (followed by some kind of dump info) in syslog, sometimes dozens of times per day:

Code:

Jul 14 23:01:06 samba kernel: eth0: Transmit error, Tx status register 82.
Jul 14 23:01:06 samba kernel: Probably a duplex mismatch.  See Documentation/net
working/vortex.txt

So I checked vortext.txt and it says of the error:

"This is a common error which is almost always caused by another host on the same network being in full-duplex mode, while this host is in half-duplex mode. You need to find that other host and make it run in half-duplex mode or fix this host to run in full-duplex mode.

As a last resort, you can force the 3c59x driver into full-duplex mode with

options 3c59x full_duplex=1

but this has to be viewed as a workaround for broken network gear and should only really be used for equipment which cannot autonegotiate."

How would I determine which Windows machine (there are only 8) or hub is responsible for this? slight problem is I'm 100 miles away, so anything I can do remotely would be much preferred (I have full remote access to everything).

Many thanks,
Steve

MensaWater · 07-15-2006, 09:58 AM

Ignore the comment about last resort. Autonegotiate does not work well at all on 10/100 networks (it has be used on 1000 as there is no way to turn it off).

Almost always issue is autonegotiate. Setting everything to Speed 100 and Duplex Full helps. Not familiar with vortex. Have a look at ethtool or mii-tool. ethtool is better but some older cards don't work with it.

ethtool eth0

Will show you current settings of eth0. See if autonegotiate is on and if so turn it off with ethtool. (man ethtool - shows you documentation for this command). Also make sure to set speed to 100 and Duplex to Full.

Ideally all your systems should have 100 Full rather than autonegotiate.