With the release of Kaseya Network Monitor version 4.1, we now have two built-in ways of monitoring your VMware infrastructure. We support versions 4.1 and 5.0 of ESX/ESXi from build 7345.
With the Lua script that comes with the install package, you can easily setup monitoring of the most common health counters’ operational status. The counters that are checked are the following:
- Storage volumes
- Storage Extent
- SATA ports
- Power supply
The script checks for any abnormalities in the operational status for each counter and will set an alarm if the status is not in an OK state.
To start monitoring, first you need to create an object choosing VMware as the operating system and then choose version (ESX 4.1, ESXi 4.1 or ESXi 5.0).
When you choose Vmware as the template, you will see the following options for choosing monitors. The new options for version 4.1 is the built-in Lua script for health and the VMware performance monitor.
VMware Health Check
For the health check Lua script, choose “WBEM_ESXi_Health_.lua” script. In most cases, you can leave “Secure” and “Ignore not found” as True, which is their default values. Also make sure you select a logon account (Use object default account is default) and check “No account logon.”
For the display results of this monitor, you will just see “Test OK” in the status windows as long as the operational status for the tested hardware is OK.
If any of the performed tests would fail, the details of what part that failed will be shown in the status field and will be included in the alarm event.
VMware Performance Monitor
When setting up this montor, it will automatically check what performance counters that are available for your version of ESX/ESXi. For detailed information about the counters, refer to your VMware version’s documentation.
In this example, the counter ”rescpu.active5.latest” was chosen. This counter will show the CPU’s last 5 minute’s average activity in percent from VMware.
The “Data Type” in this case should be chosen as “CPU Utilization.” Please refer to the KNM Documentation for creating your own data types.
In the “Comparison Options”, you could for example set that this monitor should pass if the current average CPU load during 5 minutes is less than 50% or trigger an alarm if over 50%.
Click on ”Save” and this is how the monitor will look like in KNM.
If we then create a report from the collected data, in this case on a ESXi 4.1 system with some more load, it could look something like this.
Below is how the monitoring could be configured for an ESXi 5.0 system. We have CPU load during the last 5 minutes, memory average usage, network average usage plus the health check Lua script configured for this object.