Our current logic will only connect to the MongoDB instance on startup and if the connection is lost the mapper will continue without it which means the endpoints won't be updatable.
We can implement listeners on the DB which triggers a reconnect when an error is detected. The reconnect retries should be staggered.
Our current logic will only connect to the
MongoDB
instance on startup and if the connection is lost the mapper will continue without it which means the endpoints won't be updatable. We can implement listeners on the DB which triggers a reconnect when an error is detected. The reconnect retries should be staggered.The logic would probably sit here
Connection listener docs
https://jembiprojects.jira.com/browse/OHIE-384