Jun, the current level of the code (6.01) is much better at handling
intentional duplicates, although it is not perfect from a user point of
view. At each configuration poll (or netmon startup, and probably
demand-poll) it will move the interface from one box to another,
ending up on the one it is active on. I think I have had some luck,
also, identifying it as an HSRP interface in the seedfile, another
V6 feature.

Before V6, we all just ignored those traps for nodes we knew about,
and tried to discover them in the right order so the one that was up
would get drawn.

Anybody else? I have no special knowledge of cluster configurations.

Cordially,

Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
Detroit


Jun.Chen AT dynegy DOT [email protected] on 01/24/2001 12:55:20 PM

Please respond to IBM NetView Discussion <nv-l AT tkg DOT com>

Sent by:  owner-nv-l AT tkg DOT com


To:   IBM NetView Discussion <nv-l AT tkg DOT com>
cc:
Subject:  [NV-L] Duplicate IP for Cluster




Hi all,

We are running NetView 5.1.1 on Solaris 2.6.
We have a lot of UNIXs and NTs setup as clusters.
NetView keeps reporting duplicate IP addresses on
those cluster servers. How can we let NetView know
those are cluster servers and do not generate snmp traps
to report duplicate IP address?

Thanks.

Jun

_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l