Couchbase server promoting replica document
On http://www.couchbase.com/couchbase-server/architecture, it states
If a server crashes, Couchbase Server detects the failure, promotes replica documents on another server to active status, and updates the cluster map to reflect the new cluster topology
I currently have 2 nodes in a cluster, with a view called byUuid.
The bucket is configured with 1 replica, index = true.
When I stop the server on any one node, I intermittently get result code 24, Connection Failure. Is this the right behaviour ?