apache/hadoop · error · IOException

Unable to initialize History Viewer

Error message

Unable to initialize History Viewer

What it means

HistoryViewer requires the history file's base name to look like a job history name: it splits on '_' and requires at least 2 segments (job_<id>...). When the name has no '_' at all (jobDetails.length < 2), it prints "Ignore unrecognized file: <name>" to stderr and throws IOException("Unable to initialize History Viewer"). The errorIndex points at the throw inside the name check (HistoryViewer.java:85).

Source

Thrown at hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/jobhistory/HistoryViewer.java:85

   * Constructs the HistoryViewer object.
   * @param historyFile the fully qualified Path of the History File
   * @param conf the Configuration file
   * @param printAll toggle to print all status to only killed/failed status
   * @param format the output format to use
   * @throws IOException when there is a problem parsing the history file
   */
  public HistoryViewer(String historyFile, Configuration conf, boolean printAll,
                       String format) throws IOException {
    String errorMsg = "Unable to initialize History Viewer";
    try {
      Path jobFile = new Path(historyFile);
      fs = jobFile.getFileSystem(conf);
      String[] jobDetails =
        jobFile.getName().split("_");
      if (jobDetails.length < 2) {
        // NOT a valid name
        System.err.println("Ignore unrecognized file: " + jobFile.getName());
        throw new IOException(errorMsg);
      }
      JobHistoryParser parser = new JobHistoryParser(fs, jobFile);
      job = parser.parse();
      String scheme = WebAppUtils.getHttpSchemePrefix(fs.getConf());
      if (HUMAN_FORMAT.equalsIgnoreCase(format)) {
        jhvp = new HumanReadableHistoryViewerPrinter(job, printAll, scheme);
      } else if (JSON_FORMAT.equalsIgnoreCase(format)) {
        jhvp = new JSONHistoryViewerPrinter(job, printAll, scheme);
      } else {
        System.err.println("Invalid format specified: " + format);
        throw new IllegalArgumentException(errorMsg);
      }
    } catch(IOException e) {
      throw new IOException(errorMsg, e);
    }
  }

  /**

View on GitHub (pinned to 2add963021)

Solutions

1. Pass the real history file: job_<jobid>[-...].jhist from the Yarn done/intermediatedone/ history directories.
2. Validate the name before constructing the viewer: require a '_' in the file name.
3. Check stderr - the preceding 'Ignore unrecognized file' line names the offending file.
4. List the history directory and copy the exact .jhist name.

Example fix

// before
new HistoryViewer("/user/me/Logs/jobconf.xml", conf, true);

// after
new HistoryViewer("/tmp/hadoop-yarn/staging/history/done/2026/08/22/000000/"
    + "job_1690000000000_0001-1690000000000-user-word+count-1-0-user-StreamJob.jhist",
    conf, true);

String name = Path.getPathWithoutSchemeAndAuthority(historyPath).getName();
if (name.split("_").length < 2) {
  throw new IllegalArgumentException("Not a job history file name (expected job_* pattern): " + name);
}

When it happens

Trigger: Constructing HistoryViewer (or running 'mapreduce job -history') with a path whose file name contains no underscore - e.g. 'summary.txt', 'jobconf', a directory-ish name, or a mistyped/garbled path.

Common situations: Passing the job conf/xml or a log file instead of the .jhist; shell scripts globbing the wrong directory; typos in the history path; tools that rename history files dropping the job_<submitter> pattern.

Related errors

• Incompatible event log version: ${version}
• Event schema string not parsed since its null
• Unknown mode: ${mode}
• Failure parsing JSON document: ${je.getMessage()}
• MapReduce JobHistory WebApp Address does not contain a valid

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/3663836a7afeac79. Report an issue: GitHub.