Re: forcedeth generates Invalid MAC address message

On 11/11/2007, Keith Schweikhard <kschweikhard@sbcglobal.net> wrote:

Hello,
I am currently configuring two identical ACER Aspire 5520 laptops.  The
processors are AMD Turion 64 X2 dual core, Nvidia nforce 610M providing the
ethernet connection.  Atheros AR5007EG for the wireless connection.
I am doing a netinstall using Debian Testing (Lenny).  The kernel
configuration is: 2.6.22-2-amd64.  The version of forcedeth is 0.60.  The
first machine configured without problems.  The second machine generates the
following error during boot:

ACPI: PCI Interrupt Link [LMAC] enabled at IRQ 23
ACPI: PCI Interrupt 0000:00:0a.0[A] -> Link [LMAC] -> GSI 23 (level, low) ->
IRQ 23
PCI: Setting latency timer of device 0000:00: 0a.0 to 64
forcedeth: using HIGHDMA
0000:00:0a.0: Invalid Mac address detected: 1d:9b:2a:38:1b:00
Please complain to your hardware vendor. Switching to a random MAC.
eth0: forcedeth.c: subsystem: 01025:0126 bound to 0000:00: 0a.0
.
.
.
udev: renamed network interface eth0 to eth39

This results in udev assigning a different ethernet interface  each time it
boots.  (ie. eth1, eth2, ...... eth39).  This seemingly has two effects.
1. It is impossible to bring the network connection down and back up again.
2. It is impossible to tailor the network connection (dhcp or nothing)
3. The IP address of the machine changes every time it reboots. (Result of a
random MAC address assignment.)
4. It seems to cause the ethernet configuration process to stop.  As a result,
the WAN card never initializes.
I have compared the boot messages generated by both machines and the only
differences are the invalid MAC address message, udev renaming the ethernet
interface, and the machine failing to attempt to initialize the wlan
interface.  I am guessing that the invalid MAC address and the failure to
initialize the wireless are somehow related.  Because this is currently a
dual boot machine I checked the MAC address on the windows side and it
reports back the same MAC address that forcedeth is reporting as being
invalid.  The wireless interface also works properly in windows.  I have done
some searching and have found that the forcedeth module may be incorrectly
identifying/ translating the MAC address.  (Why it works on one machine and
not on the other still confuses me.)   Given the fact that a problem has been
identified in forcedeth, is there a workaround that will prevent the kernel
from giving up on the ethernet initialization so I can at least initialize
the wireless card and obtain a consistent network connection?

--
To UNSUBSCRIBE, email to debian-amd64-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org