You might be seeing the effects of https://issues.apache.org/jira/browse/CASSANDRA-9136 which was just fixed in 2.0.15.
The connection thrashing from the above would definitely cause some hiccups. On Thu, May 14, 2015 at 3:10 PM, Christopher Kelly <[email protected] > wrote: > Hi all, > > We're running a 24 node cluster and we've noticed that whenever we either > update an existing table or add a new table, the cluster's load increases > immensely during and after for about 1 minute, triggering a lot of read and > write timeouts. Has anyone seen this behavior before or have any ideas as > to why schema updates are such an intensive operation? We are running on > 2.0.10. > > Thanks, > Chris > -- ----------------- Nate McCall Austin, TX @zznate Co-Founder & Sr. Technical Consultant Apache Cassandra Consulting http://www.thelastpickle.com
