Hi,
Assume I have created index with 3 shards and 1 replica on 3 nodes.i.e
each node having 1 primary 1 replica shard with full of data.Now If I
make down any one of the node,each one of Remaining nodes have 3
shards(primary & replica).My doubt is....Previously we allocate only 2
shards on a node with full of data but what is the meaning of
allocating 3rd shard on the same node.Isnt the disk size full?
Assume I have created index with 3 shards and 1 replica on 3
nodes.i.e each node having 1 primary 1 replica shard with full of
data.Now If I make down any one of the node,each one of Remaining
nodes have 3 shards(primary & replica).My doubt is....Previously we
allocate only 2 shards on a node with full of data but what is the
meaning of allocating 3rd shard on the same node.Isnt the disk size
full?
You would need to ensure each node's capacity is large enough to
handle such a scenario.
Not sure if you understand the index/shard concept (read about it).
A shard means: your index is splitted into several parts.
Disc space isn't normally the problem when only one index is involved.
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.