On Thu, Apr 04, 2013 at 09:07:00AM -0400, Tanstaafl wrote:
> On 2013-04-04 5:13 AM, Alan McKinnon <alan.mckin...@gmail.com> wrote:
> > I gets so bad that people are starting to make shit up to be worried
> > about, instead of just reading the simple document that is right in
> > front of their eyes that already fully and completely answers the
> > question at hand....
> 
> But Alan, haven't you read the recent (past couple of DAYS) emails in 
> this very thread from people who followed the current 'simple document' 
> you refer to and it did not work as advertised? I don't think these 
> people are making this stuff up - are you saying you do?
> 
> Not to mention the fact that this final/current seemingly complete 
> document was way, way too late for the many people who ended up with 
> totally broken systems, and *that* is what caused all of the 'hysteria 
> and mob-think' you so condescendingly speak of.
> 
> It is these reports that is causing me all kinds of fear/trepidation at 
> this seemingly simple/benign update (as you seem to believe it is).
> 
> So, again, I would really, really like a very simple answer as to the 
> *best* way to retain the old way that is least likely to cause problems 
> down the road (ie, if/when udev is subsumed by systemd). Currently I 
> count 3 different ways.
 
You are right, there are several different ways to disable udev's
predictable network interface names:

1) add net.ifnames=0 to the kernel command line in your boot loader
configuration.

2) use any of the methods listed on the upstream wiki [1] under "I don't
like this, how do I disable this?"

None of those should cause problems later.

William

[1]
http://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames

Attachment: pgpEMxf4Me5Ht.pgp
Description: PGP signature

Reply via email to