I believe you can trick netmon into handling it for you, if you can
convince it that this is a DHCP node. Here is the bit from the netmon
man page regarding DHCP:
" The netmon daemon supports the Dynamic Host Configuration Program
(DHCP), which enables a TCP/IP host to request an IP address dynamically.
When netmon discovers a new IP address that resolves to the host name of a
currently known host, and the previously discovered IP address does not
resolve
to that host, netmon updates the host with the new IP address. Discovery
must be
turned on to find these address changes because netmon finds these address
changes only during the discovery process."
You might want to hurry things along by issuing a demand poll
(nmdemandpoll),
or else manually adjust the new-node discovery timing for the addresses of
the
TMR/TEC server in xnmsnmpconf, since the change will only be found at
discovery time, and that could be 15 minutes or more. And note that the old
address must fail to resolve.
Cordially,
Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
Detroit
"Austin, Jon (FUSA)" <JonAustin@FirstUSA.com>@tkg.com on 12/18/2000
07:38:06 AM
Please respond to IBM NetView Discussion <nv-l@tkg.com>
Sent by: owner-nv-l@tkg.com
To: "'NV-L@UCSBVM.UCSB.EDU'" <NV-L@UCSBVM.ucsb.edu>
cc:
Subject: [NV-L] re-configuration of a node through automation
I've got a sticky question to pose. I hope some may find it an interesting
problem to solve...
We are operating in an environment where traps are forwarded to Tivoli T/EC
via nvserverd, the built-in non-secure Tivoli
EIF adapter in NetView.
We currently have automation (via ESE.automation ruleset), that we
reconfigure NetView to stop forwarding if the T/EC server
node is lost on the network. This is a good thing.
We are building automation (via ESE.automation ruleset) that will also
signal that the NetView has found the T/EC server down.
BTW, the TMR and T/EC server are on the same box.
We are building a geographic TMR failover solution where the TMR/TEC server
will 'move' from the current system to one in
a different subnet across the country if the TMR/TEC server is lost. This
isn't the scope of the question.
The question is:
Since I can detect with NetView when the TMR/TEC node is gone, and I
know and monitor the node where the
TMR/TEC server will move to, how can I reconfigure NetView objects for the
2
nodes to support automation. We run NetView off
of local hosts file, and we are dynamically moving the TMR/TEC node name in
DNS for the Tivoli Framework parts of the problem.
Can we just get by with changing the host records in /etc/hosts on NetView
and letting it figure out what to do with the
nodes in the NetView DB?
Should we tweak anything in the NetView DB?
Any special concerns needed to make sure NetView tosses a Node_Up trap for
the TMR/TEC server by name to have the
automation re-start trap event forwarding?
Any ideas or thoughts are appreciated.....
Jon C. Austin
ESM Analyst, Enterprise Management
Corporate Services - Technology Operations
First USA Bank (1cc/7)
JonAustin@FirstUSA.com
302-282-3498
_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l
|