eG Monitoring
 

Measures reported by GfRequestTest

The true test for the efficiency of a web application server, such as the GlassFish Enterprise server, is the speed with which it processes requests to its web applications.

The GfRequestTest test enables administrators to judge the efficiency of GlassFish by monitoring the requests to each web application deployed on it, and proactively alerting them to current or potential bottlenecks in request processing. This way, overloaded applications and the least responsive applications can be isolated. In the process, the test also promptly captures errors (if any) that the server may have experienced during request processing.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Request_count Indicates the number of requests to this application that were processed during the last measurement period. Number This measure is a good indicator of the load on each application. Compare the value of this measure across applications to identify which application is experiencing very high load.
Avg_processing_time Indicates the time taken by this application to process the requests to it. Secs A low value is desired for this measure. A consistent rise in the value of this measure could indicate a processing bottleneck, which in turn may affect application performance.

Compare the value of this measure across applications to know which application is the least responsive to user requests.

Maximum_time Indicates the maximum time taken by this application to process requests. Secs  
Error_count Indicates the number of error responses that were sent by this application during the last measurement period. Number This count includes the following:

  • The number of requests for which responses with response code that is greater than or equal to 400 were sent;
  • The number of requests for which a response could not be sent at all.

Ideally, the value of this measure should be 0.

Error_percent Indicates the percentage of error responses that were sent by this application during the last measurement period. Percent Ideally, the value of this measure should be 0.