apache/hadoop · error · IOException

Got unexpected end tag for ${name}

Error message

Got unexpected end tag for ${name}

What it means

While scanning <SnapshotDiffSection>, ReverseXML read an end-element event whose name was not the section-closing </SnapshotDiffSection>. The section's main loop only accepts <dirDiffEntry>/<fileDiffEntry> start tags and the final section end tag (expectTag with allowEnd=true); any other closing tag aborts the tool. The offending tag name is included in the message.

Source

Thrown at hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/offlineImageViewer/OfflineImageReconstructor.java:1409

      recordSectionLength(SectionName.SNAPSHOT.name());
    }
  }

  private class SnapshotDiffSectionProcessor implements SectionProcessor {
    static final String NAME = "SnapshotDiffSection";

    @Override
    public void process() throws IOException {
      // No header for this section type.
      LOG.debug("Processing SnapshotDiffSection");
      while (true) {
        XMLEvent ev = expectTag("[diff start tag]", true);
        if (ev.isEndElement()) {
          String name = ev.asEndElement().getName().getLocalPart();
          if (name.equals(SNAPSHOT_DIFF_SECTION_NAME)) {
            break;
          }
          throw new IOException("Got unexpected end tag for " + name);
        }
        String tagName = ev.asStartElement().getName().getLocalPart();
        if (tagName.equals(SNAPSHOT_DIFF_SECTION_DIR_DIFF_ENTRY)) {
          processDirDiffEntry();
        } else if (tagName.equals(SNAPSHOT_DIFF_SECTION_FILE_DIFF_ENTRY)) {
          processFileDiffEntry();
        } else {
          throw new IOException("SnapshotDiffSection contained unexpected " +
              "tag " + tagName);
        }
      }
      recordSectionLength(SectionName.SNAPSHOT_DIFF.name());
    }

    private void processDirDiffEntry() throws IOException {
      LOG.debug("Processing dirDiffEntry");
      DiffEntry.Builder headerBld = DiffEntry.newBuilder();
      headerBld.setType(DiffEntry.Type.DIRECTORYDIFF);

View on GitHub (pinned to 2add963021)

Solutions

1. Run xmllint --noout dump.xml to verify well-formedness, then locate the tag named in the message and fix its open/close balance
2. If an entry was deleted, make sure both its start and end tags went with it
3. Re-extract the section from the original dump instead of hand-patching
4. Regenerate the whole XML with a same-version oiv

Example fix

<!-- before: orphaned close tag after a deleted entry -->
<dirDiffEntry>...</dirDiffEntry>
</dirDiff>
</SnapshotDiffSection>
<!-- after -->
<dirDiffEntry>...</dirDiffEntry>
</SnapshotDiffSection>

# well-formedness first, then tag-balance inside SnapshotDiffSection
# xmllint --noout dump.xml
import xml.etree.ElementTree as ET
ET.parse('dump.xml')  # raises ParseError on unbalanced tags

# catch the CLI failure and surface the named tag
if subprocess.call(['hdfs','oiv','-processor','ReverseXML','-i','dump.xml','-o','out.img']) != 0:
    # message names the unexpected end tag; grep the XML for it
    ...

When it happens

Trigger: An extra, duplicated, or orphaned closing tag inside <SnapshotDiffSection> - e.g. a leftover </dirDiff> after a hand-deleted block, or an element removed but its close tag kept; also mis-nested elements from sed/regex surgery on the XML.

Common situations: Scripted edits that delete diff entries by line range; XML re-serialized by tools that re-order or duplicate close tags; merging sections from two dumps.

Related errors

• SnapshotDiffSection contained unexpected tag ${tagName}
• <dirDiffEntry> contained no <inodeId> entry.
• <dirDiffEntry> contained no <count> entry.
• Only read ${actualDiffs + 1} diffs out of ${expectedDiffs}
• Expected to find <childrenSize> in <dirDiff> section.

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/2a5629f1b52f833c. Report an issue: GitHub.