I planned to use 4 nodes for index, but when I created index I had only 2 nodes for start and planned to buy other nodes when I need more storage space for index.
So I created an index with 40 shards with "index.routing.allocation.include" to 4 nodes (2 exist, and 2 not exist) and start to index data. When I had 90% storage space on both nodes, I bought another node and add it cluster.
I suppose that automatic cluster rebalance must happen, so some shards will automatically go to new node, but this was not happen and I wonder why?
Why do you think it is too many shards and what resources I am wasting?
How allocation settings depends on resharding mechanics in elasticsearch?
So you think that if I disable allocation options elastic will relocate shards to new nodes?
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.