My /etc/default/mpathd is all default:

FAILURE_DETECTION_TIME=10000
FAILBACK=yes
TRACK_INTERFACES_ONLY_WITH_GROUPS=yes


Failback I'm ok with, even though my network folks say it's a bad thing.
I monitor the failovers, so I know when they happen......a single
flopping connection I don't mind, and it provides more resilience in
that it might allow ce3 to come back before ce4 fails.
I'd like to see a "last man standing" config, though.

The hostname files are setup as follows:

/etc/hostname.ce3:
x.x.x.7 netmask + broadcast + group multiIP up addif x.x.x.8 deprecated
-failover netmask + broadcast + up

/etc/hostname.ce4:
x.x.x.9 netmask + broadcast + group multiIP up addif x.x.x.10 deprecated
-failover netmask + broadcast + up



-----Original Message-----
From: Hindle, Greg [mailto:Greg.Hindle AT constellation DOT com] 
Sent: May 16, 2005 10:51 AM
To: Paul Keating; veritas-bu AT mailman.eng.auburn DOT edu
Subject: RE: [Veritas-bu] Throughput problem any ideas?


What settings are you using for IPMP?


Greg Hindle