We are running Elasticsearch 1.7.1, Logstash 1.5.3 and Kibana 4.1.1 and are gathering logs data from devices on the network. We are running 3 linux data nodes with as a logstash client and a 4th linux node with Kibana to service http requests. We are not using any replicas and the nodes are all virtuals with 64GB RAM with ES Heap of 30GB. All the collections seem to be running fine and the cluster is green. Most of the time when I go into Firefox and query the Kibana server with an * in the search field for the past 24 hours, the response comes back within 4-5 seconds but sometimes it take 30 seconds or more to refresh the data. I have turned on performance monitoring on the PC side where I am running Firefox and the CPU utilization goes up to 70-80% on the PC side. The other thing is curl statements and the pie chart on a kibana dashboard come back super fast, it seems to be happening when we are rendering any documents on the dashboard under the pie chart. We are relatively new to ELK so I wanted to see if anyone had any ideas on how to troubleshoot this. Thanks for any help or ideas!!!
P.S - I know we are running an older version of ELK but are in in a CM environment so it will take a little time to upgrade if that is truly the issue.