apache/hadoop · error · DiskErrorException

Not a directory: ${dir}

Error message

Not a directory: ${dir}

What it means

After ensuring the directory exists, DiskChecker.checkAccessByFileMethods requires dir.isDirectory(); a path that exists but is a regular file (or a symlink to one) throws DiskErrorException('Not a directory: <dir>'), failing the storage-directory health check.

Source

Thrown at hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/util/DiskChecker.java:156

  private static void checkDirInternal(LocalFileSystem localFS, Path dir,
                                       FsPermission expected)
  throws DiskErrorException, IOException {
    mkdirsWithExistsAndPermissionCheck(localFS, dir, expected);
    checkAccessByFileMethods(localFS.pathToFile(dir));
  }

  /**
   * Checks that the current running process can read, write, and execute the
   * given directory by using methods of the File object.
   * 
   * @param dir File to check
   * @throws DiskErrorException if dir is not readable, not writable, or not
   *   executable
   */
  private static void checkAccessByFileMethods(File dir)
      throws DiskErrorException {
    if (!dir.isDirectory()) {
      throw new DiskErrorException("Not a directory: "
          + dir.toString());
    }

    if (!FileUtil.canRead(dir)) {
      throw new DiskErrorException("Directory is not readable: "
                                   + dir.toString());
    }

    if (!FileUtil.canWrite(dir)) {
      throw new DiskErrorException("Directory is not writable: "
                                   + dir.toString());
    }

    if (!FileUtil.canExecute(dir)) {
      throw new DiskErrorException("Directory is not executable: "
                                   + dir.toString());
    }
  }

View on GitHub (pinned to 2add963021)

Solutions

1. Fix the configuration value to point at a real directory
2. If a file occupies the path, move it away and create the directory
3. Repair symlinks that resolve to regular files

Example fix

# before: yarn.nodemanager.local-dirs = /var/log/hadoop-yarn (a file)
# after: point at a directory
yarn.nodemanager.local-dirs = /var/lib/hadoop-yarn/local

for (String dir : configuredDirs) {
  java.nio.file.Path p = java.nio.file.Paths.get(dir);
  if (java.nio.file.Files.exists(p) && !java.nio.file.Files.isDirectory(p)) {
    throw new IllegalStateException("configured dir is a regular file: " + dir);
  }
}

When it happens

Trigger: A configured data/local/log dir path that is actually a regular file; a symlink in the path resolving to a file; a path built by string concatenation landing on a file.

Common situations: Config values with typos pointing at a jar or log file; admins replacing a directory with a symlink to a file; deployment scripts touching a file where a dir is expected.

Related errors

• Cannot create directory: ${dir}
• Directory is not readable: ${dir}
• Directory is not writable: ${dir}
• Directory is not executable: ${dir}
• Failed to delete ${file}

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/cae60ce9862e0828. Report an issue: GitHub.