Cassandra understands racks based on the configured snitch and the rack
assigned to each node (for example in cassandra-rackdc.properties if using
GossipingPropertyFileSnitch). If you have racks configured, to perform a
"rack-aware" repair you would simply need to run repair on only one rack.
Note that there is no real concept of a "rack-aware" repair, it's just a
side effect of the replication strategy and topology choices you make.​

Reply via email to