apache/iceberg · error · ParquetDecodingException

could not read page in col %s

Error message

could not read page in col %s

What it means

Wrapping error in PageIterator.initDataReader: values.initFromPage threw an IOException while binding the page's values reader to the current page; it is rethrown as a ParquetDecodingException naming the column (%s is the column descriptor/path). The cause is usually a truncated or corrupt data page in the Parquet file.

Source

Thrown at parquet/src/main/java/org/apache/iceberg/parquet/PageIterator.java:270

                + " as the dictionary was missing for encoding "
                + dataEncoding);
      }
      this.values =
          dataEncoding.getDictionaryBasedValuesReader(desc, ValuesType.VALUES, dictionary);
    } else {
      this.values = dataEncoding.getValuesReader(desc, ValuesType.VALUES);
    }

    //    if (dataEncoding.usesDictionary() && converter.hasDictionarySupport()) {
    //      bindToDictionary(dictionary);
    //    } else {
    //      bind(path.getType());
    //    }

    try {
      values.initFromPage(valueCount, in);
    } catch (IOException e) {
      throw new ParquetDecodingException("could not read page in col " + desc, e);
    }

    if (CorruptDeltaByteArrays.requiresSequentialReads(writerVersion, dataEncoding)
        && previousReader instanceof RequiresPreviousReader) {
      // previous reader can only be set if reading sequentially
      ((RequiresPreviousReader) values).setPreviousReader(previousReader);
    }
  }

  @Override
  protected void initDefinitionLevelsReader(
      DataPageV1 dataPageV1, ColumnDescriptor desc, ByteBufferInputStream in, int triplesCount)
      throws IOException {
    ValuesReader dlReader =
        dataPageV1.getDlEncoding().getValuesReader(desc, ValuesType.DEFINITION_LEVEL);
    this.definitionLevels = new ValuesReaderIntIterator(dlReader);
    dlReader.initFromPage(triplesCount, in);
  }

View on GitHub (pinned to 86d9c8fc54)

Solutions

1. Check storage/network stability and retry the read
2. Validate the file integrity (size, checksums, parquet-tools)
3. Re-copy or regenerate the corrupted file

Example fix

// before
TableScan scan = table.newScan(); // fails on truncated file in storage
// after
// re-upload the data file, then retry the scan

FSDataInputStream in = fs.open(path); long fileLen = fs.getFileStatus(path).getLen(); Preconditions.checkArgument(fileLen >= footerOffset, "File truncated: %s", path);

try {
  readPages();
} catch (ParquetDecodingException e) {
  if (e.getCause() instanceof IOException) { /* retry read or re-fetch file from storage */ }
  else throw e;
}

When it happens

Trigger: Calling initDataReader on a page whose underlying input stream throws IOException during initFromPage (e.g. truncated page data, read errors from storage).

Common situations: Truncated files; network/storage failures mid-read (S3, HDFS); bit-packed/RLE data inconsistent with the page header.

Understand the failure class

Background: "failed to read file", EACCES, ENOENT and "could not read <path>" errors: when a program can't read a file from disk — this error's family across 49 libraries.

Related errors

• Error reading mini block.
• Failed to read binary data
• could not read page " + valueCount + " in col " + desc
• Failed to read a byte
• Non-supported bytesWidth: " + bytesWidth

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/41dfa69888055e4c. Report an issue: GitHub.