apache/seatunnel · error · DatabendConnectorException

SQL_OPERATION_FAILED

SQL_OPERATION_FAILED

Error message

Failed to convert ResultSet to SeaTunnelRow: ${e.getMessage()}

What it means

Inside convertToSeaTunnelRow, any exception raised while reading ResultSet columns and building the SeaTunnelRow is caught, logged, and rethrown as DatabendConnectorException with code SQL_OPERATION_FAILED and message 'Failed to convert ResultSet to SeaTunnelRow'. It wraps type-conversion failures, column index/type mismatches, or a closed/ exhausted ResultSet.

Source

Thrown at seatunnel-connectors-v2/connector-databend/src/main/java/org/apache/seatunnel/connectors/seatunnel/databend/util/DatabendUtil.java:166

                        log.info(
                                "Field {} ({}) [{}]: {} ({})",
                                i,
                                fieldName,
                                fieldType.getSqlType(),
                                value,
                                value.getClass().getSimpleName());
                    }
                } catch (SQLException e) {
                    log.error("Error getting field {} ({}): {}", i, fieldName, e.getMessage());
                    fields[i] = null;
                }
            }

            SeaTunnelRow row = new SeaTunnelRow(fields);
            return row;
        } catch (Exception e) {
            log.error("Failed to convert ResultSet to SeaTunnelRow: {}", e.getMessage());
            throw new DatabendConnectorException(
                    DatabendConnectorErrorCode.SQL_OPERATION_FAILED,
                    "Failed to convert ResultSet to SeaTunnelRow: " + e.getMessage(),
                    e);
        }
    }

    private static Object getFieldValue(
            ResultSet resultSet, int columnIndex, SeaTunnelDataType<?> fieldType)
            throws SQLException {
        try {
            if (fieldType instanceof BasicType) {
                BasicType basicType = (BasicType) fieldType;
                switch (basicType.getSqlType()) {
                    case STRING:
                        return resultSet.getString(columnIndex);
                    case INT:
                        int intValue = resultSet.getInt(columnIndex);
                        return resultSet.wasNull() ? null : intValue;

View on GitHub (pinned to cf67b549a7)

Solutions

Compare the actual Databend table columns/types with the declared SeaTunnelRowType and re-sync the schema (re-run schema extraction or update the schema config)
Avoid unsupported column types (or cast them to supported types in the query, e.g. TO_VARCHAR(variant_col))
Check the wrapped cause for 'ResultSet closed' or connection errors - if so, fix connection stability rather than the schema
Read the e.getMessage() in the exception/log line to identify the exact column index and conversion that failed

Example fix

// before
SELECT id, attrs FROM my_table -- attrs is VARIANT, unsupported
// after
SELECT id, TO_JSON(attrs) AS attrs_str FROM my_table -- cast to supported string type

Defensive patterns

Strategy: try-catch

Validate before calling

// verify table schema matches declared rowType before running
ResultSetMetaData md = resultSet.getMetaData();
if (md.getColumnCount() != rowType.getFieldNames().length) throw new IllegalStateException("Schema drift detected");

Try / catch

try { row = DatabendUtil.convertToSeaTunnelRow(rs, rowType); } catch (DatabendConnectorException e) {
    log.error("Row conversion failed: {}", e.getCause(), e);
    throw e; // treat as dirty-record or schema-drift failure
}

Prevention

Keep declared SeaTunnelRowType in sync with the live Databend table schema; re-derive schema on table changes
Cast unsupported types (VARIANT, ARRAY) to strings in the SELECT statement
Add dirty-data handling so one bad row does not stop the whole pipeline

When it happens

Trigger: A ResultSet column type not mappable to the declared SeaTunnel type (e.g. unsupported SQL type); reading column i beyond available columns; calling getters after the ResultSet was closed or after next() returned false; null handling mismatch with the declared field type.

Common situations: Schema drift: the Databend table was altered after the job's row type was derived; custom/complex column types (ARRAY, VARIANT) not supported by the converter; connection dropped mid-result-set streaming.

Understand the failure class

Background: "query failed", "%w: SQL error" — wrapped database query errors in Go libraries explained — this error's family across 3 libraries.

Related errors

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/47a93b1b14edd6e5. Report an issue: GitHub.