apache/iceberg · error · RuntimeIOException

Failed to create Parquet input file for %s

Error message

Failed to create Parquet input file for %s

What it means

ParquetIO.file(InputFile) adapts Iceberg InputFiles to Parquet InputFiles. For HadoopInputFile instances it re-derives the Parquet HadoopInputFile from the file status; if that filesystem status lookup throws IOException, it is rethrown as RuntimeIOException with this message.

Source

Thrown at parquet/src/main/java/org/apache/iceberg/parquet/ParquetIO.java:62

import org.apache.parquet.io.OutputFile;
import org.apache.parquet.io.ParquetFileRange;
import org.apache.parquet.io.PositionOutputStream;
import org.apache.parquet.io.SeekableInputStream;

/** Methods in this class translate from the IO API to Parquet's IO API. */
class ParquetIO {
  private ParquetIO() {}

  static InputFile file(org.apache.iceberg.io.InputFile file) {
    // TODO: use reflection to avoid depending on classes from iceberg-hadoop
    // TODO: use reflection to avoid depending on classes from hadoop
    if (file instanceof HadoopInputFile) {
      HadoopInputFile hfile = (HadoopInputFile) file;
      try {
        return org.apache.parquet.hadoop.util.HadoopInputFile.fromStatus(
            hfile.getStat(), hfile.getConf());
      } catch (IOException e) {
        throw new RuntimeIOException(
            e, "Failed to create Parquet input file for %s", file.location());
      }
    }
    return new ParquetInputFile(file);
  }

  static OutputFile file(org.apache.iceberg.io.OutputFile file) {
    if (file instanceof HadoopOutputFile) {
      HadoopOutputFile hfile = (HadoopOutputFile) file;
      try {
        return org.apache.parquet.hadoop.util.HadoopOutputFile.fromPath(
            hfile.getPath(), hfile.getConf());
      } catch (IOException e) {
        throw new RuntimeIOException(
            e, "Failed to create Parquet output file for %s", file.location());
      }
    }
    return new ParquetOutputFile(file);

View on GitHub (pinned to 86d9c8fc54)

Solutions

1. Check the RuntimeIOException cause for the underlying Hadoop IOException
2. Verify the file path exists and is accessible (permissions, FS credentials)
3. Test filesystem connectivity (HDFS NameNode, S3 endpoint) and retry transient failures
4. Re-plan the scan if the file was deleted or moved

if (!fs.exists(path)) { throw new FileNotFoundException(path.toString()); }

try {
  ParquetReader reader = ParquetIO.file(inputFile).build();
} catch (RuntimeIOException e) {
  logger.error("Input file unusable: {}", inputFile.location(), e.getCause());
  throw e;
}

When it happens

Trigger: Calling ParquetIO.file(InputFile) with a HadoopInputFile whose underlying FileSystem getFileStatus/fromStatus call fails (file missing, FS unreachable, permissions).

Common situations: Deleted or renamed file between planning and read; HDFS/S3 connectivity failures; permission errors on the Hadoop filesystem; misconfigured Hadoop FS credentials.

Understand the failure class

Background: "open() failed", "failed to open file", "cannot create file" — what a file open error means and how to fix it — this error's family across 42 libraries.

Related errors

• Failed to open input stream for file: %s
• Failed to create Parquet output file for %s
• Failed to read from input stream
• Failed to read bytes from stream
• Error reading mini block.

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/28256496d77f009f. Report an issue: GitHub.