Cassandra cleanup on several servers at once -
we have big cassandra cluster 18 servers (on 1 server near 5t data )
http://docs.datastax.com/en/cassandra/2.0/cassandra/operations/ops_add_node_to_cluster_t.html - have added new nodes following documentation .
after have added new servers, began process of cleaning data (nodetool cleanup)
in documentation advise: after new nodes running, run nodetool cleanup on each of existing nodes remove keys no longer belonging nodes. wait cleanup complete on 1 node before doing next)
but cleanup 1 server takes near 2 - 3 days in our case. question can start cleaning @ once on multiple servers, 2 or 3 ...
or may lead data loss ?
some more info .
we use cassandra 2.0.13 vnodes . keep files in blons in cassandra .
replication factor = 3
cleanup doesn't involve other nodes safe run in parallel. however, may want run on 1 @ once reduce performance impact since cleanup may use lots of disk i/o.
Comments
Post a Comment