Maintenance Mode provides a simplified and pain-free solution for stopping all Orchestrator activity and shutting down IIS when needed, such as undertaking maintenance or upgrading your Orchestrator instance. This is useful for all installations, but especially for large deployments with numerous active schedules and many jobs processing at all times.
Maintenance Mode is only available to the
hostorganization, and is accessible only via Orchestrator's API. For details, see the Orchestrator API Reference.
To use Maintenance Mode:
- Enable Pending Maintenance mode using the
/api/Maintenance/Startcall and selecting Draining as the phase parameter. During this time:
- Robot communication with Orchestrator is still active (i.e. logs are uploaded, queue items are added);
- No new jobs can be created;
Stopcommand is sent to all running jobs;
- All schedules are paused;
- The Orchestrator interface is unavailable to everyone during maintenance.
If any jobs do not complete, or are too lengthy to wait, you have the ability to send a
Killcommand for all remaining jobs using the
When all running jobs have stopped, enable Maintenance Mode using the
/api/Maintenance/Startcall and selecting Suspended as the phase parameter. During this time all Robot to Orchestrator communication is stopped.
You can now shut down IIS and proceed with any maintenance or update actions needed. Maintenance Mode will persist through any stop, restart, or upgrade operation until manually switched off.
Restart IIS and Orchestrator when all maintenance actions have been completed. Turn off Maintenance Mode using the
/api/Maintenance/Endendpoint. As outcomes:
- All job executions scheduled to run during the maintenance mode are skipped;
- A maintenance log is created for each tenant. The log summarizes all stopped/killed jobs and skipped trigger executions.
This log is only accessible at the tenant level using the
api/Maintenance/Getcall and contains only those job and schedule details specific to that tenant.
Updated about a month ago