Analyzing events in the Event Viewer
The in Pega Predictive Diagnostic Cloud (PDC) provides a wide array of information about the functioning of your system. Interpreting this information correctly is the key to resolving issues in your application. By analyzing the context of the events, you can better understand the underlying reasons for certain issues in your system and resolve them.
The exact details that you need to check depend on the specific issue. The following example illustrate how you can investigate a common event – PEGA0001.
Analyzing PEGA0001 - HTTP interaction time exceeds limit
Pega Platform™ sends the PEGA0001 alert when the elapsed time for an HTTP interaction time exceeds the threshold setting for a single server interaction, which consists of the following actions:
- fetching data
- rendering the UI
- sending data to the end-user
To analyze the diagnostic data:
- Expand the PEGA0001 event in the by clicking the record.
- On the Execution summary tab, you can see which request your application processed during the time of the event, what were the activity and last step before the event.
- On the Clipboard tab, you can identify the stream class and page class of the content that your application generated at the time of the event.
- On the Performance tab, you can see information for different types of processes that your application performed for specific activities. Here, you can identify which parts of the processing took longer than others. For example, the application generated the UI very quickly, but fetching the information from the database took a very long time which resulted in exceeding the time threshold configured for this operation.
- On the Request context tab, you can identify which events appeared in the same session as the event and that might be connected to the issue. The tab displays events that have the same node requestor and interaction ID. For example, if an OPS0024 event is present, it could indicate an insufficient memory issue.
- Examine the session of the specific user, which resulted in your application generating the alert:
Filter the by the user tied to the event, then filter the results by the requestor to display a single session for that user. Examine the results to find other events that appeared during the session.
Based on this information, you can pinpoint probable causes for the issues. For example, if an OPS0024 event is present in the same user session as the PEGA0001 event, and the OPS0024 event originates from a single node, it could indicate lack of memory for that node. In such a case, restarting the node should fix the issue connected to the PEGA0001 event.
Learn more about PDC through the following articles: