Problem

This article explains how to interpret and address the condition that causes the Performance Monitor to return the following message:

Limited performance data available, no check completed.

This message indicates that, at the time the service monitor was run, no performance data was collected from the agent within the "Sustained Average Time Interval". An accurate check could not be run because there was no data against which to perform a check.

The Performance monitor relies on the performance data that is collected by the Platform Performance Gatherer. If the Performance monitor runs more frequently, or checks too little performance data, the monitor may find that there is no performance data to check against.

An example of circumstances that may cause this message to appear follows:

  • The Platform Performance Gatherer runs once every 10 minutes to gather performance information (e.g. CPU, memory and process usage) from the agent.
  • The Performance monitor runs once every 5 minutes with a one-minute "Sustained Average Time Interval". 

In the example above, the Performance monitor will run twice before the Platform Performance Gatherer collects new data from the agent. This happens because the Performance monitor only looks at the past minute of historical CPU performance data to determine the status of the monitor. The monitor is likely to miss any older data that was collected by the Platform Performance Gatherer.

Another example involves checking CPU usage for a specific process. By default, the Uptime Infrastructure Monitor agent collects information about the top 20 processes (based on CPU load) on the system that is being monitored. If the process that you want to check for CPU usage does not fall into the top 20 processes, the Performance monitor will not be able to verify the performance of the process. Instead, it will return the Limited performance data available, no check completed message. 

Solution

Ensure that the Sustained Average Time Interval is greater than the check interval for the Platform Performance Gatherer for the agent against which the service is running. This enables the agent to collect data more often than the Performance monitor checks it and ensures that the Performance monitor always has data from the agent to check against.

For example, change the Sustained Average Time Interval so that:

  • The Platform Performance Gatherer runs once every 10 minutes.
  • The Performance monitor runs once every 5 minutes with a 15-minute Sustained Average Time Interval.

In this case, the Performance monitor runs once every five minutes, will check data from the last 10 minutes and therefore should always find good data. As another example, change the check interval for the Platform Performance Gatherer so that:

  • The Platform Performance Gatherer runs once per minute.
  • The Performance monitor runs once every five minutes with a two-minute Sustained Average Time Interval.

In this case, the performance data from the agent is updated once per minute and the Performance monitor should always have some data to check against, even if the monitor only looks for one minute of data.

  • No labels