Cluster Backup

Apache ZooKeeper is a highly distributed system and can be configured to provide a high level of resilience on its own. The number of servers in the ZooKeeper ensemble can help survive the loss of several servers and when it is stretched across multiple data centres, we can survive the loss of data centres as well.

In addition, we backup important ZooKeeper data so that it can be restored to a new Apache ZooKeeper cluster in the case of total ZooKeeper failure. The backup will be a tar file of the ZooKeeper snapshot and transaction directories which is all that is required to restore a ZooKeeper ensemble. We backup the entire ZooKeeper data directory. The backup runs every 24 hours and stores it on Amazon S3 storage for a period of 7 days.

If and when you want to restore the configuration to a new Apache ZooKeeper cluster, please contact Instaclustr Support.