Full recovery

With full recovery mode, the data files are removed from the failed over server node and then, during rebalance, the node is populated with new data files (active and replica vBuckets).

From the web console, the Delta Recovery and Full Recovery options display after the server node is failed over. Both recovery methods add the server node back into the cluster during the rebalance operation, however, full recovery removes the node's data prior to the rebalance and delta recovery schedules the node's existing data to be re-used.

Full recovery characteristics:

  • Data files are removed from the server node.
  • Indexes must be rebuilt on the server node that is being re-added.
  • A working set of documents is restored for the vBuckets that are being moved.
  • Use when the data size is smaller than the bucket quotas. In this case, moving data over the network and onto disk (full recovery) may be faster than warming up data files (delta recovery).
Note: With full recovery, the disk is initialized and populated with active and replica buckets. This operation ensures a clean disk, but has overhead in terms of resource use and downtime due to disk re-population.

In the process of failing over a node, performing maintenance, adding the node back into the cluster, and rebalancing, ata files are removed from the failed over server node and then, during rebalance, the node is populated with new data files (active and replica vBuckets).

From the web console, the Delta Recovery and Full Recovery options display after the server node is failed over. Both recovery methods add the server node back into the cluster during the rebalance operation, however, full recovery removes the node's data prior to the rebalance and delta recovery schedules the node's existing data to be re-used.