This chapter covered a broad range of subjects related to the Apache Cassandra nodetool utility. The commands covered here will help you to scale your cluster horizontally, as well as to remove and replace failed nodes. If you support a cluster that has nodes deployed in public or private cloud instances, you will find that it is often much faster to replace a failed node, than to try and resurrect it. To that end, you will find yourself using these tools and techniques often.
We discussed things such as taking snapshots, and incremental backups, as well as the maintenance associated with cleaning them up. It's important to remember that snapshots and backups taken are only valid for the data stored on a particular node. If the number of nodes exceeds your replication factor, you will likely have to maintain separate sets of off-cluster backup files for each node...