System Events in Pega Predictive Diagnostic Cloud
System events are records in Pega Predictive Diagnostic Cloud (PDC) created based on alerts, unhandled exceptions, and other incidents in your system. For each event, PDC compiles a wide range of contextual data to help you prevent, diagnose, and remediate issues with your application. After creating the events, PDC stores them in a database for 14 days, after which the system deletes them automatically.
PDC supports the following types of events:
- Events based on Pega Platform™ performance and security alerts
- Events based on errors and exceptions
- Events created by PDC as an additional insight
Events based on Pega Platform performance and security alerts
PDC creates these events when an application reports a performance or security alert. These alerts are predefined and occur under specific conditions, for example, when a specific activity exceeds a preconfigured time limit, or a component is unresponsive.
Pega Platform generates the following performance and security alerts:
- PEGA performance alerts have a predefined structure and contain information about the performance of the important functionalities of your system. Your application generates these alerts when a specific condition exceeds a preconfigured threshold. For example, you receive PEGA0001 alert when an HTTP interaction exceeds the threshold setting. PEGA alerts always have the same structure and contain the information necessary to describe the functionality of the system at the time of the alert.
- SECU security alerts also have a predefined structure as well and contain information about important security issues. Your application generates these alerts when a Pega Platform server is at risk, for example, when someone attempts to hijack a user session.
- ROBO robot manager alerts inform you about performance issues with your robots, for example, when a robot loses connectivity.
Events based on errors and exceptions
PDC creates these events when an application sends an error or exception that interrupts the operation of your application. PDC classifies the exceptions as events and provides context useful to diagnose and remediate the underlying issue. Exceptions can occur for a wide range of reasons, for example, when a node unexpectedly runs out of memory.
PDC creates the following events based on errors and exceptions:
- OPS operations events are issues that your applications encounter at run time that interrupt normal operation. For example, when an application cannot allocate an object in the Java Metaspace because the Metaspace has run out of memory, Pega Platform generates a java.lang.OutOfMemoryError exception, for which PDC creates an OPS0025 event.
- DBMS database events are issues with databases that your applications access. For example, when a database query stops in idle-in-transaction state, Pega Platform generates an exception that PDC identifies as a DBMS0010 event.
- SMTP email service events are issues connected to sending or receiving emails. For example, when there are connection issues with the SMTP server, PDC creates a SMTP0001 event.
- INTG integration events are issues connected to the integrations with third-party systems. For example, when your application copies files using File Transfer Protocol (FTP) and is unsuccessful, PDC creates an FTP0002 event.
PDC also captures all unhandled exceptions that Pega Platform does not recognize. After detection, PDC displays these exceptions in the , using the exception name as an event title. If your application generates a large number of unhandled exceptions, it might indicate a severe issue.
Events created by PDC as an additional insight
PDC creates these events based on the activities and changes in your system that PDC detects. PDC provides additional details from system analysis that helps you to identify the root cause.
PDC creates the following additional insight events:
- PAL insight events indicate that a specific activity in your system took longer than usual to complete. For example, when database queries take 30% of the request processing time, PDC creates a PAL0005 event.
- GC garbage collection events inform you about major and minor garbage collection activities that Java Virtual Machine (JVM) monitoring completes.
You can open each event that PDC presents in the to inspect additional details which helps you to identify the root cause of the event. You can access the following information:
- , , and , which comprise a record of activities performed by the user before the event occurred.
- , which is connected to the event. PDC displays only those clipboards that do not contain Personal Identifiable Information (PII).
- data, which contains timers which record the duration of specific activities, and counters which record how many times your application performed specific operations. With this information, you can identify which activities were performed the longest by the user.
- , which is a record of every subroutine that was active but not completed when the exception occurred. Every Java exception that PDC records contains a stack trace.
Learn more about PDC through the following articles: