Network Management / (OpenView-NNM) Practitioners Forum

NETMON Problem in NNM

New Member.

NETMON Problem in NNM


This is regarding NETMON problem in our NMS system.

We are having NNM 6.01 running in B1000 workstation with HP-UX 11.00. In this NETMON process often gets stopped automatically.
This had happened for three times in the last two days.

During the course of time we had checked the syslog file, dmesg and top output, but not able to get any relavent information from it.

The output of the ovstatus had been captured during that time and it has been attached for the reference.

Please advice us in this regard.

Awaiting for valuable replies,

Thanks and Regards
Jason Gyure
Frequent Contributor.

Re: NETMON Problem in NNM

1. Run ovtopodump -l and see how many nodes and/or interfaces you are monitoring. You may have too many interfaces in your environment for netmon to keep up.

2. Make sure your NNM station is in DNS if that is what you use on your network. Do a NSLOOKUP on both the hostname of your box and on the ip address of your box. It should return the same info within a couple of seconds. If it takes much longer that, it may indicate a problem with DNS resolution. Properly functioning DNS is vital to the health of NETMON.

3. Run the checkDNS.ovpl script. This script is located in the Support directory of you installation CD. Once you copy the script over to your NNM box, run it.

1)# script
2)# chmod u+x checkDNS.ovpl
3)# ./checkDNS.ovpl

This unsupported script will check to see how long it takes you NNM box to resolve the hostnames of the devices you poll. You should see values of 3-4 seconds at the most. Anything higher than 4-5 indicates a DNS problem as well.

4.) Run netmon -a12 for a few seconds and the do a netmon -M0 (Mzero). This will dump its output into the Netmon.trace file in /var/opt/OV/share/log. Then take a look at that output. If netmon is falling behind in polling you will see negative #'s at the start of this file. Its ok if the negative #'s remain fairly low (under 75-100), but if you have negative #'s that run much higher or if the #'s are negative all the time, netmon has a problem keeping up with polling. That could be due to a number of factors. Keep in mind, every time you run this it appends to the netmon.trace file. You can delete it, then stop and restart netmon and it will recreate it if the file gets to big to work with.

Hope this helps. These are some of the more common problems I have see cause netmon to die.

Edward Lu
Trusted Contributor.

Re: NETMON Problem in NNM

Once I met such a problem like you, when netmon automatically stopped, the SNMP fail at the same time. So pls check the SNMP daemons and use menu 'test TCP/ICMP/SNMP' to test management station node. IF the test fail, that means the configuation of SNMP has problems. Try to reinstall the SNMP product.
Or when netmon stopped, try the command to restart the SNMP:
# snmpd &
# ovstart -c netmon

Best Regards ----:>