Last updated on MARCH 30, 2016

Applies to:

Symptoms

ACTUAL BEHAVIOUR:

Customer reported a problem when trying to truncate Endeca collection with odi interface that use standard ikm(IKM SQL to Endeca Server). The problem occur only with larger datastore (about 200k records).
Following are the steps performed:
- Execution 1(OK): execute interface(with truncate collection option enable) for loading data into endeca collection; the interface terminate without error(load about 200k records)
- Execution 2(KO): execute interface(with truncate collection option enable) for loading data into endeca collection; the interface terminate with error after 5 minutes
[ERROR|SimpleAsyncTaskExecutor-99|ODI_SEESION_ID_1354100_1.eidi.com.oracle.endeca.eidi.mdex.writer.BulkloadRecordWriter|com.oracle.endeca.eidi.mdex.writer.RecordWriter.runEventHandlers(RecordWriter.java:404)] Executing a listener failed. Listener class: class com.oracle.endeca.eidi.mdex.service.events._7._6.TruncateCollectionListener . Error: javax.xml.ws.WebServiceException: java.net.SocketTimeoutException: Read timed out
- Execution 3(KO): execute interface(with truncate collection option enable) for loading data into endeca collection; the interface terminate with error after 5 minutes
[ERROR|SimpleAsyncTaskExecutor-100|ODI_SEESION_ID_1355100_1.eidi.interface|sun.reflect.GeneratedMethodAccessor251.invoke(Unknown Source)] Cannot prepare the record writer. Error:javax.xml.ws.WebServiceException: java.net.SocketTimeoutException: Read timed out
- Execution 4(OK): execute interface(with truncate collection option enable) for loading data into endeca collection; the interface terminate without error(load about 200k records)
- and so on

In the second execution the timeout it's caused by truncate operation
In the third execution the timeout it's caused by prepare the record writer operation

Why after 2 execution in error, the truncate operation it's immediate and no error occur?

EXPECTED BEHAVIOUR:

Truncate should complete without errors

Cause