cassandra cluster #802
Description
Hi team:
There are 8 blueflood nodes(3 nodes for ingest and 5 nodes for rollup),and 29 cassandra nodes(9 nodes for ingest and 20 nodes for rollup ).
After running for 24 hour(ingest 2000000metric/minute),I find that 2 canssandra nodes's io are very busy(Disk %busy are nearly 100%),then 1 blueflood node ,rejectedSlotChecks is greater than 0.
Is the load of cassandra cluster equally？If one node is down, Is the cluster disable?