I have configured a tribe node. It can talk to three data nodes.
I see a constant TCP flow between a node and the tribe node.
I wonder what is the tcp flow doing ? What sort of information it contains ? I can't find any documentation on this.
Is it possible to make some tuning on this ? Maybe change the refresh time to have less traffic between nodes ?
Since tribe nodes are designed to work as clients across multiple clusters, they need to retrieve the cluster state from all connected clusters to merge them into a global cluster state. This is most likely causing the traffic you are noticing, even if you are not perfocming any other operation.
Refresh time has nothing to do with it since this only controls how often operations are made available for search.
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.