I'm not sure if this is a bug or a missing feature. With PouchDB, if I am listening to the _changes feed and there is a timeout Pouch restart the connection passing the 'since' parameter. cloudant-follow doesn't do it with the _db_updates feed so every time there is a timeout the changes handler receive all databases.
Thanks for raising this issue. You're right. It looks like we don't retry the /_db_updates query using a since parameter. Hoping to get a fix out this week.
I'm not sure if this is a bug or a missing feature. With PouchDB, if I am listening to the _changes feed and there is a timeout Pouch restart the connection passing the 'since' parameter. cloudant-follow doesn't do it with the _db_updates feed so every time there is a timeout the changes handler receive all databases.