Jobs Fail or Alarms Missing Due to Task Pool Limit

Product version: 11.0.4

Last modified: 04 November 2025

Problem

Jobs may fail, stop updating, or alarms may be missing from the Activities or Dashboard. The MSP Console may also fail to display tenant errors or activity statuses correctly. This behavior indicates the system is overloaded. Logs may contain the following entries:

RejectedExecutionException: Task java.util.concurrent.FutureTask rejected from java.util.concurrent.ThreadPoolExecutor

[Running, pool size = 200, active threads = 200, queued tasks = 200]

Background

When the thread pool limit is reached, queued tasks are rejected, resulting in incomplete synchronization and missing error reports.

Solution

A simple restart of the Director service will clear the queue and temporarily resolve the issue. If the problem occurs frequently, you should permanently increase the thread pool size in the Expert Settings:

  1. Enable Expert mode in your instance by adding “expert” (or “&expert”) to the end of the URL of the Settings page.

  2. Adjust the following parameters:

    system.task.pool.thread.max = 1000

    Note

    By default, the system.task.pool.thread.max parameter is set to 200 on Windows and Linux, and 15 when running on NAS.

  3. Restart the Director service to apply the new configuration.

  4. After the restart, check that job and alarm information are correctly displayed across the tenants.