Does Elasticsearch automatically ensures that all the replica shards are distributed in other data nodes irrespective of the number of shards and no of data nodes ?
Say I have 3 ES nodes and For an Index A , I have 10 shards,
Assuming the primary shards is Distributed uniformly in 3 ES Boxes
My question is Does Elasticsearch distributes the replica shards of the 10 primary shards in such a way that If any of the box out of the 3 ES Nodes went down , the data loss won't happens ?
So If I have X number of Data Nodes and Y no of Shards with Z No of Replication Factor ,
ES guarantees Failure of Any Boxes in X won't causes any data loss ?
To be clear ,Say If I have 3 Nodes and 4 shard for my index , Then then the primary and replica shard will
be distributed as below [p - Primary Shard and r-Replica Shard ]
Node 1 - [p1, r2 , r4 ] Node 2 =[p2 , r3] Node 3 - [p3 , p4 , r1 ] ,
So that failure of any nodes won't cause chaos !
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.