apache/hadoop · error · FileNotFoundException

"File/Directory does not exist: " + iip.getPath()

Error message

"File/Directory does not exist: " + iip.getPath()

What it means

getStoragePolicy resolves the path (DirOp.READ_LINK), checks READ permission, then requires a non-null last INode; nothing at the path throws FileNotFoundException with the resolved path. This is the standard missing-path guard for the storage-policy query used by 'hdfs storagepolicies -getStoragePolicy'.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/FSDirAttrOp.java:213

    return fsd.getAuditFileInfo(iip);
  }

  static BlockStoragePolicy[] getStoragePolicies(BlockManager bm)
      throws IOException {
    return bm.getStoragePolicies();
  }

  static BlockStoragePolicy getStoragePolicy(FSDirectory fsd,
      FSPermissionChecker pc, BlockManager bm, String path) throws IOException {
    fsd.readLock();
    try {
      final INodesInPath iip = fsd.resolvePath(pc, path, DirOp.READ_LINK);
      if (fsd.isPermissionEnabled()) {
        fsd.checkPathAccess(pc, iip, FsAction.READ);
      }
      INode inode = iip.getLastINode();
      if (inode == null) {
        throw new FileNotFoundException("File/Directory does not exist: "
            + iip.getPath());
      }
      return bm.getStoragePolicy(inode.getStoragePolicyID());
    } finally {
      fsd.readUnlock();
    }
  }

  static long getPreferredBlockSize(FSDirectory fsd, FSPermissionChecker pc,
      String src) throws IOException {
    fsd.readLock();
    try {
      final INodesInPath iip = fsd.resolvePath(pc, src, DirOp.READ_LINK);
      return INodeFile.valueOf(iip.getLastINode(), iip.getPath())
          .getPreferredBlockSize();
    } finally {
      fsd.readUnlock();
    }

View on GitHub (pinned to 2add963021)

Solutions

  1. Guard with fs.exists(path) and accept it is racy - still handle FileNotFoundException in the call.
  2. For audit/reporting tools, catch FileNotFoundException per path and report/skip rather than abort the run.
  3. Fix the manifest or upstream job if the entries were expected to exist.

Example fix

// before
BlockStoragePolicy p = ((DistributedFileSystem) fs).getStoragePolicy(path);

// after
if (!fs.exists(path)) {
  LOG.warn("skip policy check, path missing: {}", path);
  return;
}
BlockStoragePolicy p = ((DistributedFileSystem) fs).getStoragePolicy(path);
Defensive patterns

Strategy: validation

Validate before calling

if (!fs.exists(path)) {
  LOG.warn("cannot query storage policy, path missing: {}", path);
  return null;
}
return ((DistributedFileSystem) fs).getStoragePolicy(path);

Try / catch

try {
  return dfs.getStoragePolicy(path);
} catch (FileNotFoundException e) {
  return null; // audit tools: record and continue with the next path
}

Prevention

When it happens

Trigger: DistributedFileSystem.getStoragePolicy(path) on a path that does not exist: deleted between listing and lookup, typo, wrong partition/mount path, or a manifest entry that was never created.

Common situations: Policy audit scripts iterating stale inventories; files deleted by retention jobs mid-audit; path construction bugs in generated manifests.

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/5a64827b04bb7007. Report an issue: GitHub.