[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[ale] eth numbering change
- Subject: [ale] eth numbering change
- From: agcarver+ale at acarver.net (Alex Carver)
- Date: Thu, 9 Feb 2017 09:19:45 -0800
- In-reply-to: <[email protected]>
- References: <CAFc0yP3L7=mA3YrW_OgvxZc5_i1dbQ5kS3T7P54FCVojocriVQ@mail.gmail.com> <[email protected]> <CAFc0yP0koNJNBQTJ3jKzV=pH5++Qwiq=bxDdJ6tWSCt0cybcjA@mail.gmail.com> <[email protected]>
On 2017-02-09 09:15, Phil Turmel wrote:
> On 02/08/2017 05:02 PM, Brian Stanaland wrote:
>> There were no udev rules in place. We set that up so the NICs were named
>> the way they wanted. It's possible that the motherboards were from
>> different production runs. The wired order of the slots isn't something
>> we would specify.
>>
>> That answers the question for me.
>
> Well, not necessarily. The wired order is often the most significant
> factor, but the kernel is free to probe asynchronously nowadays. The
> slightest hiccup in the normal "eth0" driver could delay it long enough
> for the "eth1" device to initialize first, flipping the "natural" order.
>
> This is also true for hard drives. And all of this is precisely the
> reason udev persistent name rules were created, and blkid label and uuid
> names are used.
>
> Don't use ethX names with modern kernels. Period.
>
> Phil
I wouldn't say never. I had one installation that steadfastly refused
to boot properly with UUIDs for the disk. It was a very recent kernel,
too, which made it much more confusing. I had to force it back to using
/dev/sdAN references to get things working.