2829748 - Out Of Memory issues in PI/PO system due to high number of messages in HOLD status

You have been experiencing OOM in your PI/PO system, and after analyzing the OOM.hprof file generated, you can see that the OOM occurred due to the following in the Leak Suspects:

This is just a demonstration screenshot, and depending on your system configuration, some information will be different, but it should look similar to this.

If you check the details of this, you will be able to see the stack trace and also the SQL statement that was being executed:

• SAP NetWeaver
• SAP Process Integration

MSXI_InitializationWork, com.sap.bc.krn.perf.PerfTimes.getStampUs, com.sap.sql.services.core.HighResTimeAdapter.currentTimeStamp, Reading data from JDBC result set with connection, SELECT, "MSG_ID","DIRECTION", "SEQUENCE_NBR", "MSG_PROTOCOL", "SEQUENCE_ID", "TO_PARTY_NAME", "TO_SERVICE_NAME", "TRAN_DELV_TIME", "SENT_RECV_TIME", "FROM_PARTY_NAME", "FROM_SERVICE_NAME", "ACTION_NAME", "ACTION_TYPE", "CONN_NAME", "MSG_TYPE", "SCENARIO_ID" FROM "BC_MSG" WHERE "STATUS" =  N'HOLD' AND "MSG_PROTOCOL" =  N'XI' AND "NODE" =, java.util.LinkedList$Entry, OOM, server, node, Process Integration, PI, Process Orchestration, PO, NetWeaver, XI, AEX , KBA , BC-XI-CON-MSG , Messaging System , Problem