apache/seatunnel · info

Skip analyze, approximateRowCntStatement

Error message

Skip analyze, approximateRowCntStatement: {}

What it means

Info/warn log in YashanDbDialect.approximateRowCntStatement indicating the ANALYZE TABLE step for row-count estimation was skipped because the table config has skipAnalyze=true. Splitting then relies on a plain COUNT(*) query instead of optimizer statistics. It is not an error; the job continues.

Solutions

1. No action needed if skipping is intentional; the COUNT-based row estimate still runs.
2. Remove skipAnalyze=true from the table options if you want faster/more accurate split estimation.
3. If ANALYZE is undesired globally, tune split size explicitly (query-based splitting config) instead.

Example fix

// before
partition_column-analyze=true (analyze enabled)
// after (skip analyze)
table-path = "db.tbl"
skip_analyze = true

// ensure skipAnalyze is intentional
boolean skip = tableOptions.getSkipAnalyze();
if (!skip) log.info("ANALYZE will run for {}", tablePath);

When it happens

Trigger: JDBC source read from YashanDB where the table options set skipAnalyze=true, so `analyze table ... compute statistics for table` is not executed before the row-count query.

Common situations: Users disabling analyze to avoid locking or stats overhead on large/production tables during split enumeration; ops teams restricting ANALYZE privileges.

Related errors

• Column not found in table
• COMMON-17
• COMMON-19
• Invalid YashanDB 2-byte vector buffer length:
• No result returned after running query

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/800468f5fd141435. Report an issue: GitHub.

Appendix: source

Thrown at seatunnel-connectors-v2/connector-jdbc/src/main/java/org/apache/seatunnel/connectors/seatunnel/jdbc/internal/dialect/yashandb/YashanDbDialect.java:233

            }
        }

        if (useTableStats) {
            TablePath tablePath = table.getTablePath();
            String rowCountQuery =
                    String.format(
                            "select NUM_ROWS from all_tables where OWNER = '%s' AND TABLE_NAME = '%s' ",
                            tablePath.getSchemaName(), tablePath.getTableName());
            try (Statement stmt = connection.createStatement()) {
                String analyzeTable =
                        String.format(
                                "analyze table %s compute statistics for table",
                                tableIdentifier(tablePath));
                if (!table.getSkipAnalyze()) {
                    log.info("Split Chunk, approximateRowCntStatement: {}", analyzeTable);
                    stmt.execute(analyzeTable);
                } else {
                    log.warn("Skip analyze, approximateRowCntStatement: {}", analyzeTable);
                }
                log.info("Split Chunk, approximateRowCntStatement: {}", rowCountQuery);
                try (ResultSet rs = stmt.executeQuery(rowCountQuery)) {
                    if (!rs.next()) {
                        throw new SQLException(
                                String.format(
                                        "No result returned after running query [%s]",
                                        rowCountQuery));
                    }
                    return rs.getLong(1);
                }
            }
        }
        return SQLUtils.countForSubquery(connection, query);
    }

    @Override
    public Object queryNextChunkMax(

View on GitHub (pinned to cf67b549a7)