I've occasionally run into a situation where a client app running CBL stops syncing with the client. What is unusual is that sync does not restart even when the app is killed and restarted or even when the device is restarted. It simply stops syncing in both directions. This leads me to suspect some persistent change on the client or server is preventing sync, even when replication is completely torn down and reconfigured. Since we no longer have persistent replications, it seems like that would no
Could this be an issue where a checkpoint on the server would prevent push and pull replication? For example, are there corresponding sync gateway bookkeeping documents for each client that we can check and possibly delete so they can be recreated, etc.?