Intel Driver e1000e-0.5.11.2... compiles and loads fine, but can't setup ethX...?

Hey all,

I am using debian stable with a 2.6.26-1-amd64 kernel.

Since I added a

02:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network
Connection

the 'stock' e1000 kernel driver isn't able to detect this card.

After getting the appropriate kernel headers and fetching the lates intel
oss driver (e1000e-0.5.11.2.tar.gz) from their sourceforge site I was very
pleased to see it *seems* to work well after insmodding the driver....

[ 197.277725] 0000:02:00.0: eth2: (PCI Express:2.5GB/s:Width x1)
ffff81012e8ec218
[ 197.277732] 0000:02:00.0: eth2: Intel(R) PRO/1000 Network Connection
[ 197.277739] 0000:02:00.0: eth2: MAC: 4, PHY: 8, PBA No: e46981-003
softcell-3:/usr/src/e1000e-0.5.11.2/src#

but, despite everything looks good, I actually *don't* have a device eth2.

neither ethtool

softcell-3:/usr/src/e1000e-0.5.11.2/src# ethtool eth2
Settings for eth2:
Cannot get device settings: No such device
Cannot get wake-on-lan settings: No such device
Cannot get message level: No such device
Cannot get link status: No such device
No data available
softcell-3:/usr/src/e1000e-

nor ifconfig know anything about eth2....

ifconfig eth2 up
eth2: ERROR while getting interface flags: no such device


Weired... has anyone got a workaround?

BTW: Yes, I already got two onboard devices (2x forcedeath) and yes, I did
pull out e1000 before and yes I did also do a hard reset and a cold reboot
with only e1000e beeing loaded and no e1000 w/o e at all... Bummer, that
didn't change anything.

Thanks a lot!

Cheers,

Uli

Grant wrote:

>>but, despite everything looks good, I actually *don't* have a device eth2.
>
> Heh! I been caught by this behaviour too Check the udev files, and
> look carefully at the logs, you'll probably find a 'helpful' script
> changed the new NIC to eth2 -- just need to empty the offending file to
> fix...

Errr... I already 'radomly' tried eth1 to 5... but I'll take a look for
that, but don't think that's it... also I hope that's it..;-)

Cheers,

Uli

(E-Mail Removed) wrote:
> I am using debian stable with a 2.6.26-1-amd64 kernel.

> Since I added a
> 02:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network
> Connection

> but, despite everything looks good, I actually *don't* have a device eth2.
> neither ethtool

> ifconfig eth2 up
> eth2: ERROR while getting interface flags: no such device


Is it a udev problem? Check /etc/udev/rules.d/70-persistent-net.rules for
a mention of your new card. If it's not there, copy the file somewhere
safe, run /lib/udev/write_net_rules (as root), and recheck the file. Once
it's there, try rebooting (yes, uck!) or otherwise persuading udev to
notice the device and its newly associated rule.

Chris

Chris Davies schrieb:

>> ifconfig eth2 up
>> eth2: ERROR while getting interface flags: no such device
>
>
> Is it a udev problem? Check /etc/udev/rules.d/70-persistent-net.rules for
> a mention of your new card. If it's not there, copy the file somewhere
> safe, run /lib/udev/write_net_rules (as root), and recheck the file.

What? I thought udev (without any mean ubuntu-like helper scripts) would
do everything itself...

> Once
> it's there, try rebooting (yes, uck!)

Rebooting....? I remember the times I worked for an ISP, ten years ago
or so, where I was poked with 'the stick of shame' for rebooting Linux..;-)


> or otherwise persuading udev to
> notice the device and its newly associated rule.

Could be a problem to restart it when the system is up, I see...

Thanks

Uli

Chris Davies wrote:

> Is it a udev problem? Check /etc/udev/rules.d/70-persistent-net.rules for
> a mention of your new card. If it's not there, copy the file somewhere
> safe, run /lib/udev/write_net_rules (as root), and recheck the file. Once
> it's there, try rebooting (yes, uck!) or otherwise persuading udev to
> notice the device and its newly associated rule.

Cool. That was the problem...:

# PCI device 0x8086:0x10d3 (e1000e)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*",
ATTR{address}=="00:1b:21:2e:d3:e1", ATTR{type}=="1", KERNEL=="eth*",
NAME="eth1"

the Problem was, eth1 was already my second nforce ethernet port... don't
know what drives udev to state duplicate entries in this list...

Changed it to eth2 and now everything (at least) seems to work well.

Thanks a lot for that hint!

Cheers,

Uli

Ulrich Gierschner <(E-Mail Removed)> wrote:
> Rebooting....? I remember the times I worked for an ISP, ten years ago
> or so, where I was poked with 'the stick of shame' for rebooting Linux..;-)

Quite right too, but I can't see how to restart something like udev
without breaking everything else in the process. Perhaps someone more
clever than me can make a "no reboot required" suggestion?

CHris