Optimal major and minor page faults

Hello All,

I have a 12 node cluster with 64GB RAM and 31GB heap memory. Often late i am having issues with Elasticsearch performance, i happen to observe that there are major and minor page faults as below.

ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {} && sleep 60 && ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {}
MINFL MAJFL
185689007 298915
MINFL MAJFL
185717299 298918

ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {} && sleep 60 && ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {}
MINFL MAJFL
283503608 2427945
MINFL MAJFL
283544632 2428401

ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {} && sleep 60 && ps -ef | grep -E "[e]lastic.*java" | awk '{print $2}' | xargs -I {} ps -o min_flt,maj_flt {}
MINFL MAJFL
277297674 1118228
MINFL MAJFL
277341921 1118338

and so on.

Could this indicate a problem? CPU, Memory, IOWAIT utilizations are pretty much very low values. Increasing the memory beyond 64GB would help?

What is the full output from the cluster stats API?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.