Dealing with Internal Failures

The default behaviour for the X Engine is to fail open in any case where an internal failure in a rule occurs. For most situations this will be appropriate.

However, for complex online environments where the rule set is well established and known to be solid, there may be a case for altering this to fail closed (automatically recover) from a failure. This behaviour can be altered by changing the configuration for the X Engine.

Should a failure occur, an email could be automatically sent to one or more recipients alerting them to the fact that it has happened. The email will contain a failure notification and a Java stack trace of the error that was detected. This later approach makes it easy to determine precisely what caused the error and allows the recipients of the notification to quickly forward it to technical support for further investigation, if required.

Using the email feature requires an SMTP server to be defined for the server.

You can also use the Fail Safe Point rule to encapsulate code that is considered at risk of failing, but for which you do not wish the entire X Engine to terminate.

If you have access to trace data you also have access to system failure information in the server status. More details on this are found in the Trace Data section, Tracing the result of a system failure.