What causes "Unable to close insert after populating table {schemaTable}" in pentaho/pentaho-kettle?

Inspect the cause for the flush/batch error; fix duplicates or constraints.

pentaho/pentaho-kettle · error · KettleException

Unable to close insert after populating table

Error message

Unable to close insert after populating table {schemaTable}

What it means

After batch-inserting seed rows into a repository table, the helper calls database.closeInsert(); a KettleException while closing/flushing the prepared insert is wrapped as 'Unable to close insert after populating table <schemaTable>'. The rows themselves may be in limbo, so the table could be partially populated.

Solutions

Inspect the cause for the flush/batch error; fix duplicates or constraints.
Recreate the repository schema cleanly so seeding starts from empty tables.
Verify connection stability between client and database during creation.
Re-run creation after correcting; the insert closure happens only on failures in the batch pipeline.

Example fix

// before
repo.createRepositorySchema(monitor, false); // fails on existing seeded rows
// after
if (database.getRowCount(TABLE_DATABASE_TYPE) == 0) {
  repo.createRepositorySchema(monitor, false);
}

Defensive patterns

Strategy: validation

Validate before calling

// only seed into empty tables
if (database.getRowCount(schemaTable) == 0) {
  performSeedInserts();
}

Try / catch

try {
  repo.createRepositorySchema(monitor, false);
} catch (KettleException e) {
  log.error("seed insert close failed for " + schemaTable + ": " + e.getCause(), e);
}

Prevention

Ensure stable network connectivity during creation.
Avoid duplicate seed rows by starting from empty tables.
Use JDBC drivers that handle batch flush reliably.
Set generous statement timeouts.

When it happens

Trigger: createRepositorySchema() populates a lookup table (e.g. R_DATABASE_TYPE) with batch inserts, then closeInsert() throws — usually an unflushed batch error, constraint violation at flush, or connection drop.

Common situations: Network drop mid-batch during first-time repository creation; duplicate seed rows against an already-populated table; JDBC driver rejecting the batch on close.

Related errors

AI-assisted analysis of pentaho/pentaho-kettle@f3058517a1 (2026-09-13). Data as JSON: /api/errors/bef4b86ca3f4c46c. Report an issue: GitHub.

Appendix: source

Thrown at engine/src/main/java/org/pentaho/di/repository/kdr/KettleDatabaseRepositoryCreationHelper.java:379

          if ( dryrun ) {
            sql = database.getSQLOutput( null, tablename, table, tableData, null );
            statements.add( sql );
          } else {
            database.setValuesInsert( table, tableData );
            database.insertRow();
          }
        }
      }

      try {
        if ( !dryrun ) {
          database.closeInsert();
        }
        if ( log.isDetailed() ) {
          log.logDetailed( "Populated table " + schemaTable );
        }
      } catch ( KettleException dbe ) {
        throw new KettleException( "Unable to close insert after populating table " + schemaTable, dbe );
      }
    }
    if ( monitor != null ) {
      monitor.worked( 1 );
    }

    // ////////////////////////////////////////////////////////////////////////////////
    //
    // R_NOTE
    //
    // Create table...
    table = new RowMeta();
    tablename = KettleDatabaseRepository.TABLE_R_NOTE;
    schemaTable = databaseMeta.getQuotedSchemaTableCombination( null, tablename );
    if ( monitor != null ) {
      monitor.subTask( "Checking table " + schemaTable );
    }
    table.addValueMeta( new ValueMetaInteger(

View on GitHub (pinned to f3058517a1)