Subject: fxp in IBM Netfinity 4000R not useable (device timeout, no link)
To: None <port-i386@netbsd.org>
From: Timo Schoeler <timo.schoeler@riscworks.net>
List: current-users
Date: 03/08/2006 16:09:49
hi,

i have a bunch of IBM Netfinity 4000R servers (each dual PIII-700, a 
Gigabyte of ECC SDRAM, and dual intel PRO/100 NICs on board) sitting 
here waiting to go on duty running NetBSD.

unfortunately, i discovered that when trying fetch the install sets from 
ftp a device timeout for the ethernet interface occurs. even worse: the 
switch (managed Nortel 450) says 'no link', as does the LEDs on the 
machines back.

i gave Fedora Core 4 a shot, and it runs fine on that machine.

as *all* other machines here run NetBSD, it'd be a shame not to run it 
on the 4000Rs, also.

so, is this 'error' known (no PR which really matches the scenario 
here)? what can i do to help fixing it (specific information on the 
device itself)?

dmesg output for the NIC from Fedora Core 4:

e100: Intel(R) PRO/100 Network Driver, 3.3.6-k2-NAPI
e100: Copyright(c) 1999-2004 Intel Corporation
e100: eth0: e100_probe: addr 0xfebff000, irq 9, MAC addr 00:06:29:DE:7C:29
e100: eth1: e100_probe: addr 0xfebfe000, irq 10, MAC addr 00:06:29:DE:7C:28

regards,

timo