Elasticsearch deactivating and hung

I have a customer that states the RockNSM is good for a while until Elasticsearch and Kibana stop running. I am able to stop and restart Kibana but Elasticsearch is in a deactivating state and my box is hung and have to reboot in order for everything to work correctly. Is there anything I can look at and monitor to find out why this is happening?


If it’s hanging I would say check to see if you sufficient resources. If it’s virtual make sure you haven’t over provisioned ram. Possibly check disk space and folder space. Logs in /var/log/ may lend some help also.

It is virtual. disk space is good, I have 64GB of RAM allocated to the Virtual Machine, 31GB allocated to elasticsearch

By running the top command I see user elastic + consuming 143.9% CPU for java, is that normal or can I get that down?


How many cores does this machine have?

8 CPU 4 Cores per Socket (2 Sockets)
Getting more information back from the customer. We get the Kibana 503 Service not available message and then Kibana says it is waiting for elasticsearch to populate data. Also, when running curl commands, they all work except for indices and shards. What is weird is that when we reboot everything works fine for a while…


Hmmm. I’m gonna have to take some time and look at this one. If possible can to see what /var/log/elasticsearch has in it using tail and the rock version also.

roger that, will do. Also, what is weird is the RockNSM will just lock up for a while and then timeout and come back…

Plugin: elasticsearch at undefined. Request Time out after 30000ms, Kibana status: red