What causes "%s not found: %s" in apache/hadoop?

Use fs.exists(path) when you only need existence.

apache/hadoop · error · FileNotFoundException

%s not found: %s

Error message

%s not found: %s

What it means

GoogleHadoopFileSystem.getFileStatus throws FileNotFoundException("File|Directory not found: <path>") when getGcsFs().getFileInfo(gcsPath) reports the item as non-existent. This follows standard Hadoop FileSystem semantics: stat on a missing path raises FileNotFoundException, prefixed with whether the path would have been a directory or file.

Source

Thrown at hadoop-cloud-storage-project/hadoop-gcp/src/main/java/org/apache/hadoop/fs/gs/GoogleHadoopFileSystem.java:654

        return true;
      },
      String.format("mkdirs(%s)", hadoopPath));
  }

  @Override
  public FileStatus getFileStatus(final Path path) throws IOException {
    return runOperation(
      GcsStatistics.INVOCATION_GET_FILE_STATUS,
      () -> {
        checkArgument(path != null, "path must not be null");

        checkOpen();

        URI gcsPath = getGcsPath(path);
        FileInfo fileInfo = getGcsFs().getFileInfo(gcsPath);
        if (!fileInfo.exists()) {
          throw new FileNotFoundException(
                  String.format(
                          "%s not found: %s", fileInfo.isDirectory() ? "Directory" : "File", path));
        }
        String userName = getUgiUserName();
        return getFileStatus(fileInfo, userName);
      },
      String.format("getFileStatus(%s)", path));
  }

  /**
   * Returns home directory of the current user.
   *
   * <p>Note: This directory is only used for Hadoop purposes. It is not the same as a user's OS
   * home directory.
   */
  @Override
  public Path getHomeDirectory() {
    Path result = new Path(fsRoot, "user/" + System.getProperty("user.name"));

View on GitHub (pinned to 2add963021)

Solutions

Use fs.exists(path) when you only need existence.
Catch FileNotFoundException as the idiomatic 'missing' signal rather than pre-checking in concurrent environments.
Verify the path/bucket prefix construction if the object is expected to exist.

Example fix

// before
FileStatus st = fs.getFileStatus(new Path("gs://bucket/missing")); // FileNotFoundException

// after
if (fs.exists(path)) {
  FileStatus st = fs.getFileStatus(path);
} else {
  // handle missing path
}

Defensive patterns

Strategy: try-catch

Validate before calling

if (fs.exists(path)) {
  FileStatus st = fs.getFileStatus(path);
} else {
  // handle missing path without exception
}

Try / catch

try {
  FileStatus st = fs.getFileStatus(path);
} catch (FileNotFoundException e) {
  // idiomatic Hadoop 'missing' signal - do not wrap in generic error handling
}

Prevention

Use fs.exists() for pure existence checks; reserve getFileStatus for when you need metadata.
In concurrent environments, catch FileNotFoundException instead of pre-checking to avoid TOCTOU.

When it happens

Trigger: fs.getFileStatus(new Path("gs://b/missing")) for any absent object/directory; also TOCTOU where the object is deleted between a prior check and this call.

Common situations: Using getFileStatus for existence checks instead of fs.exists(); races with concurrent deleting/renaming jobs; wrong bucket or root prefix in constructed paths; missing GCS directory placeholders for 'empty' directories.

Understand the failure class

Background: "File not found" and ENOENT errors: why libraries can't find a file that should exist — this error's family across 50 libraries.

Related errors

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/0f7bc2d3deb890b6. Report an issue: GitHub.