Tribe node setup

Hi All,

I have started working on tribe node setup for elasticsearch i have two clusters in my network i have joined those two using trbe node concept with followign configurtions.

tribe.elasticsearch.cluster.name: elasticsearch
tribe.RND.cluster.name: RND

tribe.elasticsearch.discovery.zen.ping.multicast.enabled: false
tribe.RND.discovery.zen.ping.multicast.enabled: false

tribe.elasticsearch.discovery.zen.ping.unicast.hosts: ["10.0.1.218"]
tribe.RND.discovery.zen.ping.unicast.hosts: ["10.0.1.138","10.0.1.117"]

So I can able to perform search operations and health operation of cluster through this tribe node.
i read master level operations such as indexing will not support using this tribe node .please suggest me do we have any work around to make indexing possible using tribe client so that all other cluster situated will have the same data.
can we maintain same indices on both cluster for data indexing purpose?

Thanks
phani

From my understanding, tribe node allows us to perform federated search where indices are in different clusters and on different networks. If you need to index data, it's better to do so directly with the master(s) or the cluster where you want the data to be indexed, not through the tribe node.

Tribe nodes merge the cluster state (lists of nodes and indices) from multiple clusters. Given this mixing process a condition is that the clusters cannot have indexes with the same name otherwise there is the potential for conflicts and ambiguities.

You need to do that outside tribe nodes. Scaling Elasticsearch Across Data Centers With Kafka | Elastic Blog may help.

Think of tribe nodes as read only.