I’m trying to understand the implication of a node going down.
If some node goes down in a cluster and we failover, then every docs not yet replicated will be lost.
If later we manage to recover this node, maybe it simply froze and want to check if there is some data we can recover, can we query the data in that node alone?
I understand there is the delta recovery, but that seem to be automated and we may want to perform the recovered docs are not conflicting with docs created in the mean time.