apache/druid · error · ResponseException

error closing org.apache.druid.query.aggregation.Serializabl

Error message

error closing org.apache.druid.query.aggregation.SerializablePairLongDoubleComplexColumn

What it means

SerializablePairLongDoubleComplexColumn.close() releases resources via a Closer. If closer.close() throws an IOException (e.g., while closing an underlying channel or file), it is wrapped in a Druid RE with the message 'error closing <class name>'. This signals a resource leak or I/O problem during cleanup of a serialized pair-long-double complex column.

Source

Thrown at processing/src/main/java/org/apache/druid/query/aggregation/SerializablePairLongDoubleComplexColumn.java:86

  public Object getRowValue(int rowNum)
  {
    return serde.deserialize(cellReader.getCell(rowNum));
  }

  @Override
  public int getLength()
  {
    return serializedSize;
  }

  @Override
  public void close()
  {
    try {
      closer.close();
    }
    catch (IOException e) {
      throw new RE(e, "error closing " + getClass().getName());
    }
  }

  public static class Builder
  {
    private final int serializedSize;
    private final AbstractSerializablePairLongObjectDeltaEncodedStagedSerde<?> serde;
    private final CellReader.Builder cellReaderBuilder;

    public Builder(ByteBuffer buffer)
    {
      ByteBuffer masterByteBuffer = buffer.asReadOnlyBuffer().order(ByteOrder.nativeOrder());

      serializedSize = masterByteBuffer.remaining();

      AbstractSerializablePairLongObjectColumnHeader<?> columnHeader =
          AbstractSerializablePairLongObjectColumnHeader.fromBuffer(masterByteBuffer, SerializablePairLongDouble.class);

View on GitHub (pinned to 9b90983fd2)

Solutions

Check that the underlying file/channel is not closed or deleted elsewhere before this column's close() runs
Inspect the wrapped IOException (getCause()) for the real close failure and fix the underlying I/O issue
Ensure the column is closed exactly once via the Closer/Try-style lifecycle (avoid manual close on top of a registered closer)
Verify disk/filesystem health and permissions for the segment directory

Example fix

// before
closer.close(); // IOException propagates as RE
// after
try {
  closer.close();
} catch (IOException e) {
  LOG.warn(e, "error closing %s", getClass().getName()); // tolerate close failures during cleanup
}

Defensive patterns

Strategy: try-catch

Try / catch

try {
  column.close();
} catch (RE e) {
  if (e.getCause() instanceof IOException) {
    LOG.warn(e, "failed to close SerializablePairLongDoubleComplexColumn; possible leaked resource");
  } else {
    throw e;
  }
}

Prevention

Let the Closer own the underlying resources — never close the channel/file manually as well
Close columns before deleting or unmapping their backing segment files
Monitor for IOException causes indicating already-closed channels (double-close) or missing files
Run query cleanup on stable local storage; avoid force-terminating tasks mid-cleanup

When it happens

Trigger: Calling close() when the underlying Closeable held by the Closer fails to close — e.g., an IOException from closing the memory-mapped file or channel backing the serialized column.

Common situations: Underlying segment files already closed or deleted before the column; filesystem issues (NFS staleness, permission changes); double-close bugs where the channel was closed elsewhere and close() surfaces a retry failure; shutdown ordering issues in query/segment cleanup.

Related errors

AI-assisted analysis of apache/druid@9b90983fd2 (2026-09-07). Data as JSON: /api/errors/2e3a47f2ee771e54. Report an issue: GitHub.