"Scheduler worker - dead" is an uncommon event but can occur when:
- A scheduled job throws an exception that was not caught by the worker thread.
- A user manually killed a job (through the Active Transactions module) that was running and the worker did not properly recover. This often happens when killing a running import.
- Additional undocumented causes may also exist.
Dead worker threads, however, will remain dead until the scheduler pool is re-provisioned by an engineer. All dead worker occurrences require corrective action. Monitoring and scraping the stats.do page can identify null occurrences under the Scheduler Workers section. So how do we confirm if scheduler worker threads are dead using stats.do?
How to check if Scheduler is dead
a. When the Scheduler workers are available, they appear like the image below
b. When the Schedulers are dead, the stats.do will appear as in the below image
c. The stats.do page reports less scheduler workers than the actual number listed as the Number of workers
ServiceNow monitoring tool does generate an "Scheduler is dead" Alert when this happens. If no corrective action has been taken and no workers are available, please open a HI incident for Platform Performance Support team requesting to restart the scheduler.
