pentaho/pentaho-kettle · error · KettleException

Unable to save job entry of type job to the repository with…

Error message

Unable to save job entry of type job to the repository with id_job=

What it means

JobEntryJob.saveRep() wraps any KettleDatabaseException raised while writing the job entry's attributes (directories, filename, parameter arrays, pass_all_parameters) into the repository with this message, including the id_job. It indicates the repository write failed so the job entry metadata was not persisted.

Solutions

  1. Read the chained KettleDatabaseException for the concrete SQL error and fix the underlying DB issue (space, locks, constraints)
  2. Verify the repository schema matches the PDI version and run the repository upgrade/repair tool if needed
  3. Shorten or remove oversized parameter values before saving
  4. Retry the save after confirming the repository connection is healthy

Example fix

// before
jobEntry.saveRep(rep, idJob);
// after
try {
  jobEntry.saveRep(rep, idJob);
} catch (KettleException e) {
  logError("Save failed: " + e.getCause(), e);
  throw e; // surface so caller can re-run against a repaired repository
}
Defensive patterns

Strategy: try-catch

Validate before calling

// before saveRep
if (rep == null || rep.getConnection() == null) throw new IllegalStateException("Repository not connected");
for (String v : parameterValues) {
  if (v != null && v.length() > MAX_PARAM_LENGTH) throw new IllegalArgumentException("parameter_value too long");
}

Try / catch

try {
  entry.saveRep(rep, idJob);
} catch (KettleException e) {
  Throwable cause = e.getCause();
  if (cause instanceof KettleDatabaseException) {
    // retry after checking constraints/locks, or surface to user for repair
  }
  throw e;
}

Prevention

When it happens

Trigger: Calling saveRep(rep, id_job) when the repository connection drops mid-save, a parameter value violates a column constraint (too long, null), the R_JOBENTRY_ATTRIBUTE insert fails, or the DB is read-only/locked.

Common situations: Repository DB read-only or out of space; parameter_value exceeds column length; concurrent saves causing lock timeouts; schema mismatch after PDI upgrade.

Related errors

AI-assisted analysis of pentaho/pentaho-kettle@f3058517a1 (2026-09-13). Data as JSON: /api/errors/f042d98bb13f5250. Report an issue: GitHub.

Appendix: source

Thrown at engine/src/main/java/org/pentaho/di/job/entries/job/JobEntryJob.java:607

        for ( int i = 0; i < arguments.length; i++ ) {
          rep.saveJobEntryAttribute( id_job, getObjectId(), i, "argument", arguments[i] );
        }
      }

      // save the parameters...
      if ( parameters != null ) {
        for ( int i = 0; i < parameters.length; i++ ) {
          rep.saveJobEntryAttribute( id_job, getObjectId(), i, "parameter_name", parameters[i] );
          rep.saveJobEntryAttribute( id_job, getObjectId(), i, "parameter_stream_name", Const.NVL(
            parameterFieldNames[i], "" ) );
          rep.saveJobEntryAttribute( id_job, getObjectId(), i, "parameter_value", Const.NVL(
            parameterValues[i], "" ) );
        }
      }

      rep.saveJobEntryAttribute( id_job, getObjectId(), "pass_all_parameters", passingAllParameters );
    } catch ( KettleDatabaseException dbe ) {
      throw new KettleException(
        "Unable to save job entry of type job to the repository with id_job=" + id_job, dbe );
    }
  }

  @Override
  public Result execute( Result result, int nr ) throws KettleException {
    result.setEntryNr( nr );

    LogChannelFileWriter logChannelFileWriter = null;

    LogLevel jobLogLevel = parentJob.getLogLevel();
    //Set Embedded NamedCluter MetatStore Provider Key so that it can be passed to VFS
    if ( parentJobMeta.getNamedClusterEmbedManager() != null ) {
      parentJobMeta.getNamedClusterEmbedManager()
        .passEmbeddedMetastoreKey( this, parentJobMeta.getEmbeddedMetastoreProviderKey() );
    }

    if ( setLogfile ) {

View on GitHub (pinned to f3058517a1)