Subject: Re: An approach for detachable interfaces.
To: Bill Sommerfeld <email@example.com>
From: Manuel Bouyer <firstname.lastname@example.org>
Date: 11/06/1999 14:54:43
On Fri, Nov 05, 1999 at 03:44:00PM -0500, Bill Sommerfeld wrote:
> I'd be interested in hearing from anyone else besides Jonathan who
> thinks that allowing for hot-swap replacement of a removed, possibly
> failed, interface with a card with a different MAC address is
> fundamentally broken.
Hot-swapping just this way will not work quite good because you've changed
the mac address. You have to flush states in switches and arp tables in
clients. Otherwise you'll have to wait for timeout, which can be quite long.
IMHO hot-swap should keep the same mac address, and maybe a different
mechanims should be used for this: a 'virtual interface' on top of one
or more physical interface; this would allow load balancing as well.
I migth start working on this once the y2k issues are solved here, this
is one of the solution I'm considering to increase the bandwidth of NFS
servers (the other one being gigabit ethernet, but it's much more expansive).
Manuel Bouyer <email@example.com>