I set up an elasticsearch cluster which has only three nodes, should I set one as master node and other two as data nodes? And do I have to add more nodes for coordinate and ingest?
PS. I will use this cluster in a production environment and data amount is small
For small environments it is fairly common to have three master eligible nodes and then add some non-master eligible nodes.
Best case is to have three master eligible only nodes but that is harder to justify with a small cluster. Second best is to have three master eligible nodes like I said.
Having just one master eligible node or just two isn't good. Go for three.
For small deployments it is generally fine to have 3 nodes with the default configuration (all roles). I’d load gets high or the cluster grows you can introduce dedicated node types.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.