This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Windows Servers SNMP stop responding to NPM.

I am hoping someone out there can help me with this one support siad they have never heard anything like this.

I recently installed Application Performance Monitor and Netflow Performance Monitor to NPM. The website performance became extremely unreliable and would often return errors, such as request timed out. These were Orion errors and not IIS standard response page. So I added a second NIC to the NPM server reconfigured the Website portion to be on the second NIC. Changed the DNS records. Now my routers and switch are reporting into everything just fine. (the original IP address is still active and works just fine. ) How ever none of my servers a reporting into NPM any longer. The server can still be pinged just fine from NPM, but if you try and Poll the server or list resources it reports that it is down. The servers show that they are in an up state but the NIC has the blue question mark. If I go and add the second IP address to the SNMP settings it reports in just fine than, however it will not report to the primary IP address, that net flow and SNMP on all the other devices is using.

HAs any one experienced this type of problem before? If so were you able to fix it? I have about 150 servers that are doing this. I was able to Update the GPO's for the 2003 and 2008 servers with the new address just fine and it applied to them. The 2000 servers however were not as compliant.

  • redpig...

    Is there any particular reason you're using two NIC's? Load balancing? Have you tried to rescan the 1st IP address (on the 1st NIC) after several seconds? Does it then show up in NPM? Also, do you have the NIC bonded together? Let's see if we can get this fixed. Otherwise, please open a support ticket.

    Thanks...

    Philip

  • I had added the second NIC to try to improve performance for the NPM website. The NIC's are not bonded. i was trying to primarily use one NIC for the Netflow and the other for servers to respond with SNMP. However no matter what i am doing I am haveing a serious problem with the website timing out on me and basic performance beig really bad, taking excessivelty long to send notifications etc...

  • I don't uderstand why you added another NIC.  We have between 8K-10K elements per poller and don't even put a dent in a single NIC.  Maybe it's something with your DB.  Tell us about that setup.

  • The server is a VMware guest on ESX3.5 update 3. It has been configured with 4 CPU's, 4096MB Memory and is connected to a SQL 2000 for the DB. The Primary DB is at 10gb, the Log file is at 3gb and the 4 file groups are scattered between 500mb and 3 gb.

  • Is the DB on a seperate server from Orion?
    Is the DB on a VMware guest?
    What is the RAID setup on the DB?

  • The DB is on a physical server that has the DB's and log files hosted on SAN drives and the log files and db's are on separate LUN's. Believe it is vraid 5( the SAN is an HP EVA 4400). Also I took the second NIC Completey out of the equation this morning by disabling the second NIC, now the ORION NPM web site is completely timing out. I ran the Orion configuration to update the Website also.

  • It has been fixed, I had to remove some applets from the Summary page that had been on there. Once that was removed the site started to respond better than it ever has. I had support help me optimize also.