prestodb/presto · error · SQLException

Error fetching results

Error message

Error fetching results

What it means

ResultSet.next() advances the row cursor by pulling the next row from the query results stream. Any RuntimeException raised while fetching (transport failure, protocol error, server-side failure surfacing through the client) is wrapped into a SQLException with message "Error fetching results" and the original cause attached — unless the cause is already a SQLException, which is rethrown as-is. It means results could not be retrieved from the Presto coordinator, not that the result set was exhausted.

Source

Thrown at presto-jdbc/src/main/java/com/facebook/presto/jdbc/PrestoResultSet.java:157

    @Override
    public boolean next()
            throws SQLException
    {
        checkOpen();
        try {
            if (!results.hasNext()) {
                row.set(null);
                return false;
            }
            row.set(results.next());
            return true;
        }
        catch (RuntimeException e) {
            if (e.getCause() instanceof SQLException) {
                throw (SQLException) e.getCause();
            }
            throw new SQLException("Error fetching results", e);
        }
    }

    @Override
    public void close()
            throws SQLException
    {
        closed.set(true);
        client.close();
    }

    @Override
    public boolean wasNull()
            throws SQLException
    {
        return wasNull.get();
    }

View on GitHub (pinned to 55bb57d202)

Solutions

1. Inspect the cause chain (getCause()) for the real transport or server error and address it
2. Retry the whole query (re-execute the statement) with a fresh ResultSet; result sets are not resumable
3. Reduce result size (LIMIT, pagination, or server paging) to shorten streaming time
4. Check coordinator health/load-balancer idle timeouts and increase them if large gaps between pages occur

Example fix

// before
while (rs.next()) { ... } // whole loop dies on one transient failure
// after
try {
    while (rs.next()) { ... }
}
catch (SQLException e) {
    e.getCause().printStackTrace();
    rs = stmt.executeQuery(sql); // re-run the query and start over
}

boolean done = false;
for (int attempt = 0; attempt < 3 && !done; attempt++) {
    try (ResultSet rs = stmt.executeQuery(sql)) {
        while (rs.next()) { /* consume rows */ }
        done = true;
    }
    catch (SQLException e) {
        if (e.getMessage().equals("Error fetching results") && attempt < 2) continue; // re-run whole query
        throw e;
    }
}

When it happens

Trigger: Calling next() when the underlying HTTP connection to the coordinator drops mid-stream, the server returns a query error/failure while streaming pages, or the results client throws an unchecked exception (e.g. deserialization failure of a results page).

Common situations: Long-running queries whose pages take longer than coordinator limits; coordinator restarts or load balancer idle timeouts during result streaming; network partitions; query killed server-side mid-iteration while the client loops over rows.

Related errors

• Error fetching version from server
• Result set type must be TYPE_FORWARD_ONLY
• Result set concurrency must be CONCUR_READ_ONLY
• Invalid date from server:
• Expected column to be a timestamp type but is

AI-assisted analysis of prestodb/presto@55bb57d202 (2026-09-04). Data as JSON: /api/errors/b0769fcafac8b517. Report an issue: GitHub.