Hi All
I'm currently going through the process of discovering a relatively large
network of approximately 1000 routers. I am using a phased approach using
seedfiles.
The seedfiles contain the IP address for an IBM 2216 router and then the IP
addresses for 50-100 IBM 2212 routers connected to the 2216. There is one
IP address per router and these addresses are pingable. The seedfile also
contains a wildcard IP address range for an unreachable network to force
the discovery to be limited to the nodes listed in the seedfile.
Once I have created the seedfile, I update the netmon configuration and
restart the netmon daemon.
It can take quite a while (30 minutes) before netmon starts to discover any
of the devices listed in the seedfile.
The system is a uniprocessor RS/6000 F50 with 1GB RAM. RAM and CPU are
over-utilised in my opinion so there are moves afoot to upgrade. However
there should be sufficient resources to perform the discovery - and the
system does eventually get around to it!
There are currently 9000+ objects in the database and the ovwdb cache is
set to 18000.
When I run netmon -a xx to get a /usr/OV/log/netmon.trace output, it shows
only a few hosts in the PingWait and SNMPWait queues with no apparent
problems.
Software: AIX 4.3.3 ; Framework 3.6.2 ; NetView 6.0.
Any ideas what could be causing the delay, or where I could look to
determine if anything else could be causing the problem?
Regards,
Danny Williams
IBM Global Services
E-mail: danny@uk.ibm.com
P.S. Please cc any responses to my e-mail address
|