Default Monitoring of CloudTest Environment

Document created by Chris Sommerstad Employee on Jul 22, 2017
Version 1Show Document
  • View in full screen mode

SOASTA CloudTest provides default monitoring of the resources listed below for both Single-Server and Multi-Server environments:

For Maestro instances:

  • CPU
  • Memory
  • Network I/O
  • JVM heap usage
  • JVM threads

For Results Service instances:

  • All of the above
  • Web server threads
  • Number of database connections

For CloudTest System Monitor, navigate to Central list > Library > System Monitors > Single-Server Environment > CloudTest System Monitor. 

Data from default monitors is displayed in the new System Monitor Summary Dashboard as well as in the Monitor Container.

The System Monitor Summary dashboard shows a point-in-time snapshot of resources in the manner of the Windows System Monitor. The screenshot on the right shows the Monitor Summary for a multi-server environment.

Monitor Data Automatic Purge of Aged Data

Monitor data can be summarized on a periodic basis and the detail data can be optionally removed after some period of time. This aggregation is done per Monitor. Note that if a monitor server has more than one host, such as when specified by a Monitoring Server Group, then host level information is not stored with these aggregates.

There are two settings to support this functionality that can be found in the Central list > Monitor, lower panel > General tab:

Aggregation Duration

This is the period for which monitor sample data should be aggregated. The default is every 10 seconds. Use a higher number if the trend analysis would not find 10 seconds to be significant. Using a period that is less than the monitor sample rate will provide no benefit. 

Detail Data Retention Period

Specifies how far back to retain detail monitor sample data. Defaults to –1 (do not purge detail data). Setting this to 0 will not store any detail data. Setting to a positive number (e. g. 30) will retain detail data for that number of minutes.

Aggregate Tables for Monitors and New Aggregate Monitor Charts

Aggregate tables for monitors present both aggregate and detail information using powerful filtering options. Aggregate information results from Monitoring Server Groups, while detail information results from specific Monitoring Servers.

Aggregate filtering can produce valuable comparisons and contrasts to detail charts when used on the same dashboard.

Select either a Single Resource from Result or Single Resource from Monitor when adding a widget, and then select a defined monitor and resource on it. Finally, to filter for aggregate data (single- or multi-select from Groups for aggregate data), while for detail data use Servers or defined hosts.

An Aggregate monitor chart, CPU Percentage, shows average CPU usage for the aggregated period.

When defining a Single Resource for Result you can select aggregated data (All Monitoring Server Groups) or detail data (All Monitoring Servers or hosts).

Selecting host-level detail (by selecting two hosts as shown below) produces a very different chart than the aggregate chart above. For related monitor settings, refer to the General tab settings for Aggregate Duration and Detail Data Retention Period to set a time period.