[lfs-support] udev problem

Casey Daniels mailinglist at cd.kcfam.net
Mon Oct 21 04:43:21 PDT 2013

On 10/20/2013 05:13 PM, Bruce Dubbs wrote:
> Casey Daniels wrote:
>> On 10/20/2013 11:25 AM, Bruce Dubbs wrote:
>>> Are these lines split for email or are they that way in the file?  I
>>> think you need to escape some newlines. For example,
>>> # net device e1000e
>>> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", \
>>> ATTR{address}=="XX:XX:XX:XX:XX:XX", ATTR{dev_id}=="0x0", \
>>> ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
>>> You can also try:
>>> udevadm test --action=ADD /sys/class/net/<device>
>>>       -- Bruce
>> Found out something interesting.  As long as I don't try to assign
>> anything except the onboard NIC to eth0, it will let me do as I please.
>> I can even name the onboard NIC to something other than eth0, but
>> anytime I try to name something eth0 (besides the onboard nic) udev just
>> does what ever it wants to do.
> Just a thought.  I believe a system sets up to preset values and then
> renames what's required when it processes the rule.  If something is
> already named to the value requested, then the rename fails.
> What are the values in /sys/class/net/ with no rules?  Then what if you
> just try to change one of the problem interfaces to say, eth1, does that
> work?
> One thing to try is to skip the udev sections in the boot sequence
> completely and then run udevd with --debug after boot.  Another thing to
> try is setting udev.log-priority=debug on the kernel command line.
>     -- Bruce
I finally got tired of fighting it, and just moved the cables.


More information about the lfs-support mailing list