Eike Lantzsch writes:
> must not try to steer". But I'd say that in this case the change of an
> interface name from one release to the other is at worst a slight
> inconvenience which I easily accept for the better big picture of predictable
Even when I had to use machine with multiple interfaces, I had NO
predictability problem, only with VM images I had to bash a bit udev
scripts, because the same image started more than once saw different
virtual interfaces. Those using a cloud would benefit of this...
It's ten years or more than I run my custom configurations and
scripts, and I have to go to edit all of them with the machine unable
to use the network, and this could be more than a "slight"
inconvenience, but I got worse...
> Naming interfaces according to the place of the hardware on the bus are
> debateable - especially for the technically unaware
> never set up her/his own hardware,
Gentleman, when you buy a rack mounted thing you do not set up your
HW. When you receive a laptop from your employer, you do not set up
your HW.
This change gives nothing more than before. Unless it names the
interfaces in a way that you understand from the NIC name that it's
the one corresponding to leftmost plug on the second row on the rear
of the box, this naming adds NOTHING except the need to edit legacy
scripts.
At least the developers could share these fine mushrooms :)
--
/\ ___ Ubuntu: ancient
/___/\_|_|\_|__|___Gian Uberto Lauri_____ African word
//--\| | \| | Integralista GNUslamico meaning "I can
\/ coltivatore diretto di software not install
già sistemista a tempo (altrui) perso... Debian"
Warning: gnome-config-daemon considered more dangerous than GOTO
