4.3 Analyzing events to identify the configuration element that has a problem

From the JP1 events that occurred in an application, analyze the performance of the related system configuration elements, and identify the configuration element that has a problem. Events that occurred in the application are sorted chronologically by JP1/OA and are sorted according to the relationships between the applications. From the sorted events, check the details of the events starting with those that have the highest probability of containing the cause of the problem, and filter the events to be analyzed. Then, analyze the filtered events and the resource performance trends related to these events to identify which configuration elements have a problem.

Procedure

1. In the E2E View window, understand the configuration of the IT base system.

   You can check elements related to the application that was selected as the starting point, as well as the status of occurrence of warnings and errors.

2. To check events that are occurring in the application, click the Event Analysis View button.

   

   Events that occurred in the related application and the application that was selected as the starting point are displayed in chronological order. The most recent events that have a greater probability of affecting the application are displayed in the upper-right corner of the window.

3. In the next window, you can check multiple events that occurred during the last 30 minutes in the starting-point application AJSM-HOST01 and the highly relevant application AJSA-HOST02 most recently around the same time. To analyze these events, click the event button.

   

   Detailed information such as messages and the time when events occurred is displayed.

4. To analyze the resource performance status at the time that an event occurred, select an event to analyze from the displayed event list, and then click the Show Performance button.

   Detailed information for the most recent events is displayed at the top, enabling you to check performance information from the top one by one.

   If you click the Show Performance button, the performance graphs and metric information for virtual machines or hosts (considered virtual machines in this manual) related to the selected events will be displayed. If the related resources are virtual machines, information about the hypervisors on which the virtual machines are running will also be displayed.

5. Check the selected event and virtual machine performance trends.

   In the displayed performance graph, you can see that there were no notable changes in the performance of the virtual machine of AJSM-HOST01 when the event occurred. The second, third, and fourth events are also related to AJSM-HOST01, which indicates that these events are not relevant to the virtual machine performance trends either.

6. Select the event in AJSA-HOST02, and click the Show Performance button to check the virtual machine performance of AJSA-HOST02.

   

   In the performance graph for the virtual machine of AJSA-HOST02, you can see that performance degraded significantly just before the event occurred. From this you can conclude that the virtual machine of AJSA-HOST02 is causing the problem in the starting-point application.

   Tip

   In the Event Analysis View window, you can sometimes identify the cause by checking both the performance graph for virtual machines and the performance graph for hypervisors. If there is no problem related to virtual machines or hypervisors, you can assume that applications are the cause of the problem.

7. To continue with an analysis of the virtual machine of AJSA-HOST02, click the Show E2E View button.

To Page Top