Bug#792627: linux-image-4.0.0-2-686-pae: linux-image-4.0 does not bind ethernet device

[Hans]

Am Freitag, 29. Januar 2016, 16:00:52 schrieben Sie:
> Am 30.09.2015 um 23:33 schrieb Michael Biebl:
> > On Thu, 16 Jul 2015 23:59:33 +0200 Hans <hans.ullrich at loop.de> wrote:
> >> Hi all,
> >> 
> >> there is another hint, I dioscovered in my kernel logs.
> >> 
> >> -------------
> >> Jul 16 21:51:55 localhost kernel: [ 1277.742162] atl1c 0000:01:00.0:
> >> version 1.0.1.1-NAPI
> >> Jul 16 21:51:55 localhost kernel: [ 1277.775065] atl1c 0000:01:00.0
> >> enp1s0:
> >> renamed from eth0
> >> 
> >> ----------------
> >> 
> >> However, there is NO enp1s0 in my
> >> 
> >> /etc/udev/rules.d/70-persistent-net.rules
> >> 
> >> and I saw no reason, why it should be renamed from eth0 to enp1s0.
> >> 
> >> Of course, in /etc/network/interfaces there is only "eth0" and NO enp1s0.
> >> I believe, the reason for this bug is this strange renaming behaviour, as
> >> if I do an "ifconfig -a" it shows me "enp1s0" instead of "eth0".
> >> 
> >> And as "enp1s0" is not defined in /etc/network/interfaces, I get the
> >> delay.
> >> 
> >> I could prove:
> >> 
> >> NO renaming and "eth0" = working correctly
> >> renaming "eth0" to "enp1s0" = no network, delay at boot and "ifconfig -a"
> >> shows NO eth0 but "enp1s0".
> > 
> > Is that a VM, if so, what kind?
> > Do you have a /etc/udev/rules.d/70-persistent-net.rules? If so, please
> > attach it along with the output of "ip a s"have notr 
> 
> Hans, could you provide the requested information?
> 
> Michael

Hi Michael,

I am sorry that I can not help at the moment, as in the meantime my notebook 
is in repair (the graphics card is defective). However, the last thing, which 
fixes this for me, was to add a parameter into the grub commandline (something 
like "if.names=0" or similar). Sorry, I can not check this, because my 
notebook is at the repair. On the other hand, this solution worked for me as a 
workaround for now. It does not explain, why persistent-net.rules is not used.

But I can confirm, I have /etc/udev/70-persistent-net.rules active here and 
also checked the MAC-address in it.

Maybe, there is a hint: By default debian installs the package "macchanger", 
which might interfere with the MAC-addresses. Just an idea. I have not proven 
it yet.

As soon, as I got the notebook back, I can / will tell more. 

Best regards

Hans