apache/hadoop · error · IOException

Output file not at zero offset.

Error message

Output file not at zero offset.

What it means

Error "Output file not at zero offset." thrown in apache/hadoop.

Source

Thrown at hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/file/tfile/BCFile.java:291

        }
      }
    }

    /**
     * Constructor
     * 
     * @param fout
     *          FS output stream.
     * @param compressionName
     *          Name of the compression algorithm, which will be used for all
     *          data blocks.
     * @throws IOException
     * @see Compression#getSupportedAlgorithms
     */
    public Writer(FSDataOutputStream fout, String compressionName,
        Configuration conf) throws IOException {
      if (fout.getPos() != 0) {
        throw new IOException("Output file not at zero offset.");
      }

      this.out = fout;
      this.conf = conf;
      dataIndex = new DataIndex(compressionName);
      metaIndex = new MetaIndex();
      fsOutputBuffer = new BytesWritable();
      Magic.write(fout);
    }

    /**
     * Close the BCFile Writer. Attempting to use the Writer after calling
     * <code>close</code> is not allowed and may lead to undetermined results.
     */
    @Override
    public void close() throws IOException {
      if (closed == true) {
        return;

View on GitHub (pinned to 2add963021)

Solutions

1. Open the output stream for a new/truncated file so the BCFile writer starts at offset 0.

When it happens

Trigger: Thrown at hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/file/tfile/BCFile.java:291 when the library encounters an invalid state.

Common situations: Occurs when a BCFile writer is created on an output stream whose position is not zero. Open a fresh stream or seek to offset 0 before creating the writer.

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/fb8f00a2d32db810. Report an issue: GitHub.