Targeting elasticsearch


We run a small cluster, 2 x master, 4 x data, 2 x ingest in addition to a number of logstash servers. Can someone confirm that we are targeting the correct elasticsearch nodes from kibana and beats.

From kibana (elasticsearch.hosts in kibana.yml) we only target the data nodes (not master or ingest)

From the various beats agents and logstash we only target the ingest nodes.

Is that best practice, or is there a more efficient methodology?


That's a pretty safe approach :slight_smile:

Thanks for your quick response Mark.

I assume if we were to add other types of data nodes .... data_hot, data_warm etc then we would add those to our elasticsearch.hosts target group in kibana.yml ?

Yeah, or you could use the ingest nodes, or setup (a) client node(s) for them if you want.

Also, 2 masters is not ideal (I missed that). You want 3 so that you always have a majority if one leaves.

Understood, thanks for your help.

1 Like

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