Impaired Worker Monitoring
CloudHub continuously verifies the health of the workers running Mule applications.
When a worker is impaired by hardware or software failures, including network unavailability, memory issues, server hardware failures, or file corruption, CloudHub automatically attempts to restart the app on a new worker.
Because CloudHub replaces impaired workers on a fixed schedule, which depends on the underlying infrastructure, you might experience a delay between the time the new worker starts and when it starts sharing the load. As long as the app is running on multiple workers, this delay shouldn’t impact the application.
After the app is completely up and running, CloudHub switches the traffic to the new app instance and deletes the old app instance, ensuring zero downtime.
If the application fails to start after five attempts, its status is Failed or Unrecoverable Runtime Error and you must manually restart the application.
You see a notification indicating that the automatic restart occurred. The application log indicates when an app is restarted automatically and any failures. For example:
Your application was running on an impaired worker instance. To avoid any downtime, the platform has restarted the application.| The automatic worker-restart functionality is different from the automatic application-restart functionality, which is configurable using the Automatically restart application when not responding option on the Settings page for the app. CloudHub automatically attempts to restart impaired workers, regardless of the application-restart setting. |
Scheduled Maintenance Activities
CloudHub checks for these scheduled maintenance activities on the worker infrastructure that can cause potential downtime for the running applications: reboot, stop, start, or replacement of the instances. To ensure zero downtime, CloudHub schedules the app to restart automatically on a new worker before the maintenance occurs.
You can restart your apps manually at your convenience or let CloudHub restart the app automatically.
One month before the scheduled restart, the notification (bell) icon in the top right of both the Applications page and the Settings page for the application turns red to indicate that you have a new notification:
Click the icon to display the notification, which includes the date that CloudHub will automatically restart the worker.
CloudHub displays a reminder notification one week before automatic application restart.