Finding 100's of INFO message in system.log
RANGE_SLICE messages were dropped in last 5000 ms: 0 internal and 5 cross node. Mean internal dropped latency: 0 ms and Mean cross-node dropped latency: 17248 ms
How This can be fixed and whats the cause
@adityajain22_141041 it means that the nodes are overloaded and cannot keep up with range reads. You need to either throttle the traffic to the cluster or add more nodes to increase the capacity of the cluster. Cheers!
Unfortunately, the warning indicates that the replicas have a latency of over 17 seconds for range reads. It can also happen if your application is doing a large (expensive) read request. Cheers!
5 People are following this question.
