Symptoms:
1. In the RA Oracle EBS Agent log, the following error is present:
'getConnection: retry (ErrorCode 17002) (SQLState 08006) IO Error: Socket read timed out uc4_oae_pk.getStatus 0 timeout : 10 : i : class java.lang.Integer'
2. You may also see the error followed by the following lines:
Internal Error
U00003620 Routine 'UCEX_R' forces trace because of error.
U02000048 TRACE file opened with flags '0000000000000001', max.trace: '9', memory trace: '0', search type: 'S', search string: ''.
U00003449 Output to the TRACE file is finished.
SQL Exception
3. Forced trace is generated where the following errors are present:
java.sql.SQLRecoverableException: Closed Connection
at oracle.jdbc.driver.PhysicalConnection.needLine(PhysicalConnection.java:5416)
at oracle.jdbc.driver.OracleStatement.closeOrCache(OracleStatement.java:1585)
at oracle.jdbc.driver.OracleStatement.close(OracleStatement.java:1570)
at oracle.jdbc.driver.OracleStatementWrapper.close(OracleStatementWrapper.java:94)
at oracle.jdbc.driver.OraclePreparedStatementWrapper.close(OraclePreparedStatementWrapper.java:82)
at oracle.jdbc.driver.OracleCallableStatementWrapper.close(OracleCallableStatementWrapper.java:1444)
at com.uc4.ra.jdbc.DBCall.closeResources(DBCall.java:837)
at com.uc4.ra.jdbc.DBCall.runFunction(DBCall.java:658)
at com.uc4.ra.jdbc.DBCall.runFunction(DBCall.java:570)
at com.uc4.oebs.agent.RunJob$StatusThread.run(RunJob.java:244)
4.RA Oracle OEBS agent can either go down or not due to this
Release : 12.1, 12.2, 12.3
Component : AUTOMATION ENGINE, RA Oracle EBS agent
If you see the following error in the OEBS agent log file, 'getConnection: retry (ErrorCode 17002) (SQLState 08006) IO Error: Socket read timed out uc4_oae_pk.getStatus 0 timeout : 10 : i : class java.lang.Integer : ' followed by the forced trace with java.sql.SQLRecoverableException, that usually means that the connection between agent and the database was made successfully (most likely, the agent did not go down), but because of the issue with network/firewall the socket/data was empty and eventually it timed-out because no data was available.
Firewall or network issues can kill the connection between an agent and DB after a certain amount of idle time. Please check with your network team if there was an issue with network/firewall at the time of the issue.