Oracle RDS: ORA-39006 and ORA-39014 at the beginning of impdp with a very large dump-file

Question

Hi,
we have different ORACLE-RDS instances and all impdp-sessions with small and middle sized dump-files worked ok.
With one large RDS instance we are getting following errors just a view seconds after starting the impdb:

```
Exception in Data Pump job
ORA-39006: Interner Fehler
ORA-39029: Worker 2 mit Prozessname "DW00" vorzeitig beendet
ORA-31671: Worker-Prozess DW00 enthält eine undefinierte Ausnahme.

ORA-39014: Ein oder mehrere Worker wurden vorzeitig beendet.
ORA-39029: Worker 1 mit Prozessname "DW00" vorzeitig beendet
ORA-31671: Worker-Prozess DW00 enthält eine undefinierte Ausnahme.

ORA-39014: Ein oder mehrere Worker wurden vorzeitig beendet.
```
I took both ways to test:
* impdp-client (via windows-server and parfile) 
* DBMS_DATAPUMP-Calls via sqlplus/SQLDeveloper

Result: Same error

The size of the dump-file is huge = about 2.2TB

The dump-file is created from an SE2-Database (PDB) but with NLS German/Germany.
The dump file is a FULL-Export dump - but I defined a **schema-import** via a parfile. 
I know, that AWS-Oracle-RDS is always installed in AMERICAN/AMERICA - so I wrote an logon-Trigger to alter the session and to set the correct NLS-Config.

I hope you can help me.
Thanx a lot
wohu

re:Post Agent · Answer

The errors you're encountering (ORA-39006, ORA-39014, ORA-39029, and ORA-31671) during the import of a large dump file (2.2TB) into an Oracle RDS instance are likely related to the size of the dump file and potential resource constraints.

Here are some suggestions to troubleshoot and potentially resolve the issue:

1. Check available storage: Ensure that your RDS instance has sufficient free storage space to accommodate the large dump file and the data being imported.

2. Increase instance resources: Consider scaling up your RDS instance to provide more CPU and memory resources for handling the large import job.

3. Use parallel processing: If not already doing so, try using the PARALLEL parameter in your import job to utilize multiple workers. This can help distribute the workload.

4. Import in smaller chunks: Instead of importing the entire 2.2TB dump file at once, try breaking it down into smaller, more manageable pieces. You can use the SCHEMAS or TABLES parameters to import specific schemas or tables at a time.

5. Check for compatibility issues: Ensure that the source database (SE2-PDB) and the target RDS instance are compatible in terms of Oracle versions and features.

6. Monitor the import process: Use the DBA_DATAPUMP_JOBS and V$SESSION_LONGOPS views to monitor the progress of your Data Pump jobs and identify any specific issues.

7. Review exception handling: Implement exception handling in your DBMS_DATAPUMP calls to capture more detailed error information. This can help pinpoint the exact cause of the failures.

8. Verify file permissions: Ensure that the user performing the import has the necessary permissions to read the dump file and write to the target schemas.

If these steps don't resolve the issue, you may need to open a support case with AWS to investigate further, as there could be underlying RDS-specific factors contributing to the problem.  
**Sources**  
[Resolve DMBS DATA PUMP API errors for RDS for Oracle DB instances | AWS re:Post](https://repost.aws/knowledge-center/rds-oracle-dmbs-datapump-errors)  
[ORA-39001: invalid argument value while importing data dump file in aws rds for oracle. | AWS re:Post](https://repost.aws/questions/QUVJ7X0Go2RPqdfyrmLcHHJQ/ora-39001-invalid-argument-value-while-importing-data-dump-file-in-aws-rds-for-oracle)

Oracle RDS: ORA-39006 and ORA-39014 at the beginning of impdp with a very large dump-file

Add your answer

Relevant content