ELM Enterprise Manager includes a Performance Alarm out-of-the-box and enabled as one of the default monitor items on Core, System and Performance licenses. This watches for low disk space on all instances (drives) and will be triggered when less than 15% available is detected.
In previous versions of ELM, when the Performance Alarm was triggered, an Alert entry was created. Now that Alerts have been discontinued with ELM 6.5, the alarms create events that are visible in a new Event View named “ELM — Monitor Item Events View”. From here we could view performance issues or assign Notification Methods.
This default threshold in the Performance Alarm Monitor Item is a great starting point for general purposes or across multiple servers with common personalities, but what about those “problem children” we all seem to have? Having a disk free space threshold set at 15% could mean that the alarm is being triggered over and over again and we really don’t want to know about it. For example, a couple of Exchange servers that are always running low on disk space by nature; we know this already, and really don’t worry about it until a lower threshold has been met.
In a case like this we are better off to do a custom setup for a Performance Alarm. To get started we’ll launch the New Monitor Item Wizard and select the Performance Alarm.
In the next step we’re going to change the selected Counter from “% Free Space” to “Free Megabytes” so that we can set a specific value which for this example will be 700MB. We’ll leave the asterisk (*) in place to monitor all instances, or if desired we could add specific instances to monitor. (Monitoring “_Total” is not very helpful when it comes to diskspace across multiple instances when one may be full while another is barely used.)
In the next step we’ll leave enable Event Creation checked and we can then take a look at the Variables that can be included in this event message for more detail.
The list below shows the details of each variable and their description.
- %EVENT% – Event ID, equivalent to the Event Id field in Event Viewer.
- %COMPUTER% – Name of the computer the event was generated on.
- %DATE% – Date the event was created, from the TimeGenerated field.
- %TIME% – Time the event was created, from the TimeGenerated field.
- %TYPE% – Type of the event, I = Informational, W = Warning, E = Error, S = Audit Success, F = Audit Failure, C = Critical, and V = Verbose.
- %LOGNAME% – Name of the event log the event originated from.
- %SOURCE% – The source of the event, equivalent to the Source field in Event Viewer.
- %CATEGORY% – The category of the event, equivalent to the Category field in Event Viewer.
- %USER% – The Username of the account that generated the event.
- %MESSAGE% – Message text of the event. This variable has white space, tabs, and new lines trimmed.
- %VIEWNAME% – Name of the Event View the event originated from.
- %VIEWINDEX% – The unique index of the Event View the event originated from.
- %METHODNAME% – Name of the Notification Method the event originated from.
- %METHODINDEX% – The unique index of the Notification Method the event originated from.
Next we have the option to pick an agent to test the new Performance Alarm Monitor Item on which we’ll skip. The next step is where we can choose an Agent Category to assign the Performance Alarm to. Rather than use this new Performance Alarm across servers in an existing category (where we don’t want it on all of them) there are a couple options.
- We could use the shortcut here to create a new Agent Category, then assign the Performance Alarm to it.
- We could skip assigning to an Agent Category, then come back and assign the Performance Alarm one-by-one to specific Agents.
For our example, we used the “New” shortcut to create a new Agent Category and named it “Exchange Servers”.
Next is to set the schedule for the Performance Alarm to run and we’ll go ahead and change this to every 30 minutes.
And our final step — and this is an important one so don’t just skip through it — is to give the new Performance Alarm a unique name and good description. This is important as you build up a library of these types of Performance Alarms you will want to be able to easily tell them apart without having to look at the properties of each one! (see below)
Now that we’ve established a Performance Alarm specifically for the disk free space on our Exchange Servers we can view any event activity in the “ELM — Monitor Item Events View” shown earlier. To be notified of Performance Alarm issues the best approach is to create a new Event View specific to the Exchange Server category and Performance Alarm events and have our notification be based on that unique criteria. You would not want to assign a notification method to the “ELM — Monitor Item Events View” itself as you would be getting the results of Ping, Service state changes and other less important information being sent.
The fastest way to create this new Event View is to right click on the Performance Alarm event, then choose “Create Event View.” Walk through the steps for the new Event View wizard and you are off and running. (For help on setting up or assigning a Notification Method to an Event View, visit the ELM Help File.)
We hope that you found this article on Setting Up a Low Disk Space Performance Alarm informative and useful and wish you continued success with ELM.