I have 7 servers scattered who knows where each with a single node. I would like to create a single pane of glass of sorts, so I would like to setup a single server to have a copy of each node for each server. So each server has a node that syncs to a node on the "master" server, but if anything were to happen to the master the data is still fine on the individual servers. I would like to keep those node separate from one another as well, so 7 nodes on the master each syncing with the data collecting nodes, but unable to overwrite the data on the collectors. I'm sure it's an odd request but any info would be helpful.
The part that worries me most is the fact the servers might be anywhere. Elasticsearch clusters require a low latency between the nodes, most likely a local network. There have been some discussions across deploying clusters across multiple data centers, but we don't have anything yet.
The network I am working with is local. there are 3 node that are not, 1 uses a VPN so we will exclude that one. the other two are tied into the same layer 2 network via fiber and are less than 30 miles away.
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.