Network Monitoring With OpsMgr 2007 R2?

Twice last week at our user group event on OCS 2007 R2, I was asked about monitoring the network using Operations Manager 2007.  Well, you can. 

Check out the Jalasoft offering.  I had a demo of it last year at TechEd and it looked very nice.  They do something similar to the HP blade management pack.  You install an application either on your management server or on a dedicate machine depending on the load.  This communicates with your routers, switches and firewalls.  The app then feeds all the data into OpsMgr so you can monitor your network infrastructure health and performance using OpsMgr.

There’s another trick you can do on _small_ networks.  System Center Essentials includes a basic SNMP based management pack for network monitoring.  It assumes there’s only a small number of network devices so it’s not efficient CPU-wise when that number grows.  You can pull out that MP and install it on OpsMgr.  That’s totally unsupported by MS.

Then you can do what I’m doing with OpsMgr 2007 R2 (cross platform extensions).  Our network guys set up a SLES VM on our Hyper-V cluster.  They’re used to the traditional tools like SYSLOG and Cactii.  That’s fine with me; I’m not a network guy (I just request changes and want to know things are OK at a high level) so I want them to work efficiently.  We’re doing a few things to extend this with OpsMgr 2007 R2:

  • We’re running synthetic transactions to test port 80 connections on a number of service providers on the Internet.  If one goes down we can assume an issue with the service provider (you wouldn’t believe how often web hosting providers go down!).  If more than one goes down it might be a large issue on the Internet or a routing issue.  If they all go down then we have a connection issue.
  • I have deployed an Cross Platforms Extensions agent to the SLES Linux VM.  We’re using the log file template to check the SYSLOG file for specific strings that indicate issues.
  • I’m checking the status of the httpd and the syslog-ng daemons/services using the UNIX/Linux Services template.

I’ve bundled all of those and the SLES box into a group.  The network guys are delegated rights to that group and are getting notifications for alerts from entities in the group.  That gives us network monitoring via OpsMgr 2007 R2 with minimum spend.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.