I have the follow cenario: 1 node master, 2 data nodes and 1 client node (kibana+logstash) and a few questions.
Regarding the elasticsearch.yml configuration, should I need to configure an initial list of nodes that will be contacted to discover and form the cluster?
you should make a cluster including all master, data and client nodes that is helpful for load balancing the data also. for this you have to mention all nodes hostname in discovery parameter.
As you told logstash is configured on client node you can mention the IP of elasticsearch as per your configuration ex: localhost if you configured elasticsearch on the same machine or cluster IP from which you are accessing the Elasticsearch cluster. also you can mention the data node ip
process wil be as whenever any logs are coming it will go to client node and due to elasticsearch all data will be reflected in the cluster and request will be followed.
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.