| 1 |
On Tue, Sep 19, 2006 at 07:09:00AM +0200, Pawel Kraszewski wrote: |
| 2 |
> Dnia sobota, 16 wrze?nia 2006 09:28, felix@×××××××.com napisa?: |
| 3 |
> |
| 4 |
> > # Onboard Gigabit |
| 5 |
> > KERNEL=="eth*", SYSFS{address}=="00:E0:81:2A:58:C6", NAME="eth0" |
| 6 |
> |
| 7 |
> |
| 8 |
> At my EM64T computer I have this '10-eth.rules': |
| 9 |
> |
| 10 |
> KERNEL=="eth*", SYSFS{address}=="00:11:22:33:44:55", NAME="eth0" |
| 11 |
> KERNEL=="eth*", SYSFS{address}=="55:44:33:22:11:00", NAME="eth1" |
| 12 |
> |
| 13 |
> It looks exactly like your (well, MACs are fake. However, in real MACs i have |
| 14 |
> letters lowercase, which is NOT what Juergen suggested) - and it works. I had |
| 15 |
> once "wandering" interfaces - 2 cards were randomly assigned to eth0 and |
| 16 |
> eth1 - which was strange, because they are 2 diff cards and modules were |
| 17 |
> loaded always in the same sequence. |
| 18 |
> |
| 19 |
> I am using sys-fs/udev-100-r2 |
| 20 |
|
| 21 |
Same version here. I have another system, x86 instead of ~amd64, with |
| 22 |
four PCI cards with network interfaces on them among other things, and |
| 23 |
it always assigns ports in a consistent order, and it has |
| 24 |
10-net.rules. |
| 25 |
|
| 26 |
Maybe I will try changing the name to 10-eth, altho I don't see what |
| 27 |
difference the name makes. |
| 28 |
|
| 29 |
I am reluctant to use completely different names, like net* instead of |
| 30 |
eth*, because so many scripts know the interface names. I may have to |
| 31 |
try that, but it seems strange that your EM64T obeys it, so does my |
| 32 |
x86, but my ~amd64 doesn't. |
| 33 |
|
| 34 |
-- |
| 35 |
... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._. |
| 36 |
Felix Finch: scarecrow repairman & rocket surgeon / felix@×××××××.com |
| 37 |
GPG = E987 4493 C860 246C 3B1E 6477 7838 76E9 182E 8151 ITAR license #4933 |
| 38 |
I've found a solution to Fermat's Last Theorem but I see I've run out of room o |
| 39 |
-- |
| 40 |
gentoo-amd64@g.o mailing list |
Archives