Monitor.jvm WARN message?

Elastic Stack Elasticsearch
1

Hi,
can anybody please explain me the meaning of this WARN message. The problem is, ES cluster is not allocating any shards.

[2016-08-31 11:51:46,993][WARN ][monitor.jvm              ] [sit-master-data-node-0] [gc][young][257][54] duration [4.2s], collections [1]/[4.7s], total [4.2s]/[3.1m], memory [530mb]->[415.7mb]/[5.9gb], all_pools {[young] [128mb]->[996.6kb]/[133.1mb]}{[survivor] [16.6mb]->[16.6mb]/[16.6mb]}{[old] [385.4mb]->[398.1mb]/[5.8gb]}
  1. What measures I have to take so that the situation will not repeat again. I have allocated 6 GB to ES out of 16 GB. Cant allocate more than 6 GB.
  2. Do I need to delete or close old indexes?
  3. How to decide or who decides that shards will be push from new generation to old generation?

br,
Sunil

2

I think,
When I start ES cluster, whole 3.1 GB data in the cluster gets loaded in memory, that's why its RED in the beginning and slowly changes through YELLOW and then GREEN.
Am I correct?

3

That's garbage collection, it's expected.

No, the data is not loaded into memory.