Joseph Kowalski writes:
> James Carlson wrote:
> > Yes, vanity naming solves the problem pretty neatly.  No, vanity
> > naming doesn't exist yet and this case isn't dependent on it.  The
> > best we can do is to warn users that such a name change is *in the
> > general case* infeasible.
> >   
> Gee, I must have slept through something.  What is "vanity naming" and
> where can I get some?  (Pointers that is...)

Sorry; that was too terse.

It's a reference to "Clearview Nemo unification and vanity naming"
(PSARC 2006/499; also known as "UV"), which will provide the ability
to rename network interfaces.

One of the most important usage cases for this feature (despite the
"vanity naming" title for the project) is to be able to keep the name
of an interface the same.  Say, for example, you pull out a "qfe" card
and replace with a "ce" card.  UV will allow you to rename the new ce0
interface back to qfe0, so that all those configuration files that
refer to qfe0 won't have to change.

It's an aid for DR and zone migration, among other things.

> Remember the days when all you needed was a major and minor number, and
> then we could call it anything you wanted?   8^)

You still can, but the name (and typically not the number, other than
for Sun Cluster) ends up getting embedded in software that refers to
the node, and thus you can't _change_ the name without hunting down
the affected consumers.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677

Reply via email to