apache/hadoop · error · IOException

unknown DiffEntry type {}

Error message

unknown DiffEntry type {}

What it means

While dumping the SNAPSHOT_DIFF section to XML, a DiffEntry message carried a type other than FILEDIFF or DIRECTORYDIFF. Protobuf round-trips unknown enum numbers, so this happens when the image contains a DiffEntry type added in a Hadoop newer than this oiv build's generated protobufs — or, less likely, bit corruption decoding to an out-of-range type value.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/offlineImageViewer/PBImageXmlWriter.java:750

  }

  private void dumpSnapshotDiffSection(InputStream in) throws IOException {
    out.print("<" + SNAPSHOT_DIFF_SECTION_NAME + ">");
    while (true) {
      SnapshotDiffSection.DiffEntry e = SnapshotDiffSection.DiffEntry
          .parseDelimitedFrom(in);
      if (e == null) {
        break;
      }
      switch (e.getType()) {
      case FILEDIFF:
        out.print("<" + SNAPSHOT_DIFF_SECTION_FILE_DIFF_ENTRY + ">");
        break;
      case DIRECTORYDIFF:
        out.print("<" + SNAPSHOT_DIFF_SECTION_DIR_DIFF_ENTRY + ">");
        break;
      default:
        throw new IOException("unknown DiffEntry type " + e.getType());
      }
      o(SNAPSHOT_DIFF_SECTION_INODE_ID, e.getInodeId());
      o(SNAPSHOT_DIFF_SECTION_COUNT, e.getNumOfDiff());
      switch (e.getType()) {
      case FILEDIFF: {
        for (int i = 0; i < e.getNumOfDiff(); ++i) {
          out.print("<" + SNAPSHOT_DIFF_SECTION_FILE_DIFF + ">");
          SnapshotDiffSection.FileDiff f = SnapshotDiffSection.FileDiff
              .parseDelimitedFrom(in);
          o(SNAPSHOT_DIFF_SECTION_SNAPSHOT_ID, f.getSnapshotId())
              .o(SNAPSHOT_DIFF_SECTION_SIZE, f.getFileSize())
              .o(SECTION_NAME, f.getName().toStringUtf8());
          INodeSection.INodeFile snapshotCopy = f.getSnapshotCopy();
          if (snapshotCopy != null) {
            out.print("<" + SNAPSHOT_DIFF_SECTION_SNAPSHOT_COPY + ">");
            dumpINodeFile(snapshotCopy);
            out.print("</" + SNAPSHOT_DIFF_SECTION_SNAPSHOT_COPY + ">\n");
          }

View on GitHub (pinned to 2add963021)

Solutions

1. Use an oiv build at least as new as the cluster that took the snapshots
2. Verify the image md5 to rule out in-section corruption
3. If versions cannot be aligned and snapshots are not needed, file a HADOOP JIRA with the type value shown in the message

try {
  new PBImageXmlWriter(conf, out).visit(file);
} catch (IOException e) {
  if (e.getMessage() != null && e.getMessage().contains("unknown DiffEntry type")) {
    // newer protobuf enum value than this build knows: align tool version
    LOG.error("snapshot diff type unknown to this oiv build: {}", e.getMessage());
  }
  throw e;
}

When it happens

Trigger: `hdfs oiv -p XML` on an image from a newer Hadoop that introduced a new snapshot diff entry kind; corrupted bytes inside the snapshot diff section.

Common situations: Snapshotted namespaces read with old tooling after a cluster upgrade.

Related errors

• Unrecognized section {}
• Unsupported file version {}
• Unrecognized section {s.getName()}
• Cannot process fslayout version {imageVersion}
• Unrecognized section {}

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/f20749b88c72e556. Report an issue: GitHub.