Troubleshooting OpenJPA 'floods'

Body

An OMNIbus Web GUI server makes OpenJPA calls when it is started as a node in a load-balanced cluster.  It is possible that the number of these calls can be so large as to cause performance issues in Web GUI/TIP/DASH.

If a load in a cluster is experiencing performance issues and there is evidence in the logs of a flood of OpenJPA calls, then there are a couple of things that can be checked and changed.

1.  If this exception is seen:

Exception:com.ibm.db2.jcc.b.SqlException SourceId:com.ibm.ws.rsadapter.jdbc.WSJdbcDatabaseMetaData.getColumns ProbeId:374 Reporter:com.ibm.ws.rsadapter.jdbc.WSJdbcDatabaseMetaData@6bdb59 com.ibm.db2.jcc.b.SqlException: DB2 SQL error: SQLCODE: -443

Then the suggestion in the following technote can be applied to the DB2 instance: http://www-01.ibm.com/support/docview.wss?uid=swg21962585

2.  If this exception is seen in ncw.0.log:

org.apache.openjpa.persistence.ArgumentException: An error occurred while parsing the query filter "SELECT n FROM Node n". Error message: The name "Node" is not a recognized entity or identifier.

Then the suggestion in this technote can be applied to the persistence.xml file in TIP/DASH: https://developer.ibm.com/answers/questions/309378/the-configuration-database-is-unavailable-so-you-c.html#answer-309381