Monitoring a Microsoft Failover Cluster

With the last build, we have now released a Lua script that can monitor your Microsoft server cluster. The released version will check for operational status in nodes and shared resources.

Kaseya Network Monitor v4.1 Build 7394 or higher
Microsoft Windows Server 2003/2008 Cluster

Setting up this monitor involves setting up an object that represents the cluster. Testing of individual nodes is also possible but in case the node you are testing against would go down, the script would just fail, not being able to read any values.

The IP address entered here is the virtual IP for the cluster. Preferably, you should enter the DNS name. Make sure you have a Windows account associated with the object that has administrative rights on the cluster nodes.

When creating the new monitor, choose Script -> Lua script.

Choose “MSCluster.lua” and check the “No account logon” checkbox. The scripts takes no arguments, so leave that textbox empty. It will be using the object’s default account for authentication, or you can choose another account for the monitor.

For this kind of monitor, you might want to consider lowering the number of failed tests before an alarm is generated. In this example, for testing purposes, we have chosen to only allow one failed test before an alarm is triggered. The default test interval is 60 seconds.

In our test lab, we have a two-node cluster with a few resources. While the monitor is in an OK state, this is what will be displayed. The nodes will be listed with their state in parenthesis, a division bar and then the resources will be listed with their states also.

This is how the monitor could look in an alarm state. The status information shows here that there’s one node (CLU02) down, but also one resource (TestShare) is in an offline state.

To download the script, go to

Leave a Reply

Your email address will not be published.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