There are 'netmon -a' commands that cause netmon to dump to
netmon.trace a list of what has been offloaded to what MLMs for status
polling. The various '-a' options are documented somewhere now - the
manpage for netmon maybe? Also remember that you can assign nodes to
the MLMs beyond what they discover. I believe that if a node is
assigned to the MLM for status, that includes all of the interfaces on
that node as they are known to the Netview database. You can see that
in the netmon.trace when you do that dump. And when the MLM sends a
status change trap (which is configured in trapd.conf to logonly or
don't log or display, I forget which), it triggers a 'fake'
netmon-type status change trap, so I would expect normal propogation
and correlation to take place regardless of whether the information
came from the MLM or netmon.
You should plan on using the Smartset Editor to modify the scope of
the MLMs to cover the subnets you need them to cover. Special
smartsets are created for you for that purpose when you demandpoll an
MLM. You should also plan on having Netview do the discovery rather
than relying on the MLMs.
Cordially,
Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
Detroit
*Jane Curry <jane.curry@skills-1st.co.uk>*
Sent by: owner-nv-l@lists.us.ibm.com
06/21/2004 04:00 PM
Please respond to
nv-l
To
nv-l@lists.us.ibm.com
cc
Subject
Re: [nv-l] MLM Questions
I think I tested this one in the past (though it may have changed...).
From memory, MLM doesn't discover or poll interfaces that are not in
the same subnet as his "primary" interface. Can't remember how he
decided which interface to use if he had several, but my memory is that
it really only worked on one interface.
If you do further testing, please report the outcome back here!!
Cheers,
Jane
James Shanks wrote:
>
> I don't mean to be rude, Scott, but if you are testing, then whom
> would be better able to answer those questions than you? Nothing like
> testing the code yourself to become the expert. Both those
> scenarios are a bit off the norm for me and for anyone else I know.
> Even the MLM code jockey here said he wasn't sure about either one,
> and the second is really about netmon. But our best guess on both is
> negative. We could be wrong, however, so go ahead, test!
>
>
> James Shanks
> Level 3 Support for Tivoli NetView for UNIX and Windows
> Tivoli Software / IBM Software Group
>
>
> *"Bursik, Scott {PBSG}" <Scott.Bursik@pbsg.com>*
> Sent by: owner-nv-l@lists.us.ibm.com
>
> 06/21/2004 11:22 AM
> Please respond to
> nv-l
>
>
>
> To
> "Nv-L (nv-l@lists.us.ibm.com)" <nv-l@lists.us.ibm.com>
> cc
>
> Subject
> [nv-l] MLM Questions
>
>
>
>
>
>
>
>
>
> AIX 4.3.3 NetView 7.1.3 FP2
>
>
>
> I have several MLM's that I am testing with. One is AIX and the
others are
> NT. I have a couple of questions about MLM functionality that I
didn't see
> covered in the manual although I may have overlooked it.
>
> 1. If the MLM has multiple interfaces how will the MLM manage its
> networks?
> Will it discover only one subnet and if so how do you determine which
> interface the MLM will bind to? Or will it discover all subnets it has
> interfaces for?
>
> 2. If an MLM is polling a subnet and a node in that subnet has 2
> interfaces, one on the network with the MLM and another that isn't
managed
> by the MLM but NetView manages it, will NetView correlate the mlm
detected
> down event and a NetView Interface Down event into a Node Down event?
>
> Thank You,
>
> Scott Bursik
> PBSG
>
>
>
--
Tivoli Certified Consultant & Instructor
Skills 1st Limited, 2 Cedar Chase, Taplow, Bucks, SL6 0EU, UK
Tel: +44 (0)1628 782565
Copyright (c) 2004 Jane Curry <jane.curry@skills-1st.co.uk>. All
rights reserved.