Endpoint decommission is a process of permanent endpoint instance shut down. It can happen for various reasons, such as:
- Dividing endpoint responsibility among other endpoints
- Scaling down endpoint instances
- Other (like removing certain functionality from the system)
When decommissioning endpoints, several aspects should be considered:
- Routing should be adjusted to make sure that no new messages will be delivered to that endpoint.
- All messages should be processed from the endpoint's queue.
- Timeouts should be rerouted to other endpoint instances.
- If any errors are discovered in the ServicePulse/ServiceControl tools, redirect functionality can be used to reroute messages to another endpoint instance.