uCosminexus Application Server, Operation, Monitoring, and Linkage Guide
This subsection describes the two types of targets whose statistics are to be monitored and for which you can set a threshold value.
- Full garbage collection frequency
Monitors the frequency of full garbage collection. The full garbage collection can detect frequently occurring abnormal conditions. Specify the frequency of full garbage collection as the threshold value.
You can monitor the frequency of full garbage collection in the execution environments of the J2EE applications and batch applications.
The message ID reported as a Management event is KDJE53850-W.
- Number of entire pending requests for Web containers
Monitors the number of entire pending requests for Web containers. Before the number of entire pending requests for Web containers exceeds the request queue size, it can be detected that the request queue space has decreased. For a threshold value, specify the ratio of the number of entire pending requests for Web containers to the pending request queue size of the number of entire pending requests for Web containers.
You can monitor the number of entire pending requests for Web containers in the execution environment of J2EE applications.
The message IDs reported as a Management event are KDJE53862-W (High Threshold value), and KDJE53863-I (Low Threshold value).
- Number of pending requests for Web containers
Monitors the number of pending requests for Web containers. Before the number of pending requests for Web containers exceeds the request queue size, it can be detected that the request queue space has decreased. For a threshold value, specify the ratio of the number of pending requests for Web containers to the pending request queue size of the number of pending requests for Web containers.
You can monitor the number of pending requests for Web containers in the execution environment of J2EE applications.
The message IDs reported as a Management event are KDJE53864-W (High Threshold value), and KDJE53865-I (Low Threshold value).
- Number of entire pending requests for Web applications
Monitors the number of entire pending requests for Web applications. Before the number of entire pending requests for Web applications exceeds the request queue size, it can be detected that the request queue space has decreased. For a threshold value, specify the ratio of the number of entire pending requests for Web applications to the pending request queue size of the number of entire pending requests for Web applications.
You can monitor the number of entire pending requests for Web applications in the execution environment of J2EE applications.
The message IDs reported as a Management event are KDJE53866-W (High Threshold value), and KDJE53867-I (Low Threshold value).
- Number of pending requests for Web applications
Monitors the number of pending requests for Web applications. Before the number of pending requests for Web applications exceeds the request queue size, it can be detected that the request queue space has decreased. For a threshold value, specify the ratio of the number of pending requests for Web applications to the pending request queue size of the number of pending requests for Web applications.
You can monitor the number of pending requests for Web applications in the execution environment of J2EE applications.
The message IDs reported as a Management event are KDJE53868-W (High Threshold value), and KDJE53869-I (Low Threshold value).
- Number of pending requests in URL group
Monitors the number of requests present in the pending requests queue of a URL group. This type of monitoring can detect the reduced number of free slots in the request queue, before the number of pending requests for URL groups exceeds the size of the request queue. For the threshold value, specify the ratio of number of pending requests for URL groups to the pending request queue size for URL groups.
The number of pending requests in a URL group can be monitored with the execution environment of J2EE applications.
The message IDs reported as a Management event are KDJE53860-W (High Threshold value), and KDJE53861-I (Low Threshold value).
Use the resource depletion monitoring functionality to set up and monitor a threshold value for the number of requests in the default pending queue for Web containers and the pending queue for Web applications. For the resource depletion monitoring functionality, see 4.3 Resource depletion monitoring functionality and output of resource depletion monitoring information.
All Rights Reserved. Copyright (C) 2013, Hitachi, Ltd.