apache/hadoop · error · IOException

Failure parsing JSON document: ${je.getMessage()}

Error message

Failure parsing JSON document: ${je.getMessage()}

What it means

JSONHistoryViewerPrinter assembles the job view into an org.json JSONObject and streams it out; any JSONException raised while building (problematic puts) or writing the document is wrapped as IOException("Failure parsing JSON document: <je.getMessage()>", je). The underlying org.json message names the actual problem.

Source

Thrown at hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/jobhistory/JSONHistoryViewerPrinter.java:80

   * machine-readable JSON format.
   * @param ps the {@link PrintStream} to print to
   * @throws IOException when a problem occurs while printing
   */
  @Override
  public void print(PrintStream ps) throws IOException {
    json = new JSONObject();

    Writer writer = null;
    try {
      printJobDetails();
      printTaskSummary();
      printTasks();

      writer = new OutputStreamWriter(ps, StandardCharsets.UTF_8);
      json.write(writer);
      writer.flush();
    } catch (JSONException je) {
      throw new IOException("Failure parsing JSON document: " + je.getMessage(),
          je);
    } finally {
      if (writer != null) {
        writer.close();
      }
    }
  }

  private void printJobDetails() throws JSONException {
    json.put("hadoopJob", job.getJobId().toString());
    json.put("user", job.getUsername());
    json.put("jobName", job.getJobname());
    json.put("jobConf", job.getJobConfPath());
    json.put("submittedAt", job.getSubmitTime());
    json.put("launchedAt", job.getLaunchTime());
    json.put("finishedAt", job.getFinishTime());
    json.put("status", ((job.getJobStatus() == null) ?
        "Incomplete" :job.getJobStatus()));

View on GitHub (pinned to 2add963021)

Solutions

1. Read the wrapped JSONException message to identify the offending field/operation.
2. View the same file in the default human format ('human' printer) which tolerates the same data - if it prints, the issue is JSON-specific.
3. Retry the JSON view on a history from a normally-completed job; abnormal (killed/failed-early) jobs can carry nulls that trip the printer.
4. If a specific field consistently fails, patch the printer to null-guard that put and report it upstream with the JSONException message.

Example fix

# before
hadoop job -history -format json <path>/job_*.jhist 2>&1 | head
# IOException: Failure parsing JSON document: ...

# after: same file, tolerant printer
hadoop job -history <path>/job_*.jhist | head

catch (IOException e) when printing JSON history: if message starts with "Failure parsing JSON document", retry the same file with the human-format printer (new HumanReadableHistoryViewerPrinter path via format "human") before giving up.

When it happens

Trigger: Running HistoryViewer with format "json" on a history whose parsed job data triggers a JSONException - e.g. null/odd values put into the JSON object for abnormal jobs (killed jobs with unset finish times), or an I/O failure while the JSON writer flushes to the output stream.

Common situations: JSON output on incomplete/abnormal job histories; piping JSON viewer output into a closed/full output stream; automation scraping JSON for many jobs hitting one malformed record.

Related errors

• Incompatible event log version: ${version}
• Event schema string not parsed since its null
• Unknown mode: ${mode}
• Unable to initialize History Viewer
• MapReduce JobHistory WebApp Address does not contain a valid

AI-assisted analysis of apache/hadoop@2add963021 (2026-08-22). Data as JSON: /api/errors/97ff6892dfefee42. Report an issue: GitHub.