CODEWITHSUNDEEP
elasticsearchelasticsearch-plugin
AbhishekAsh
AbhishekAsh

Reputation: 401

Can i survive without replication in elastic search?

I have gone through several documents but have not found much information on the following :

1) Can i survive with 0 replica shards ? We have tonnes of data and right now dont have enough infra to replicate the total data. We are ok with reindexing if we have lost some data. That at this point is acceptable. We will have decent size index but will have thousands of index.

2) If I go ahead with point 1 , if a node goes down , how do i know which how much data is lost and which all index needs to re-indexed.

3) Other than HQ , any other plugin which helps me with the above things ?

Looking forward to some insights on the above points .

Upvotes: 0

Views: 454

Answers (2)

shubham
shubham

Reputation: 482

Yes you can survive with 0 replica shards.

But if you have a multi-node clusters it is recommended to have at least 1 replica shards for every primary shards. So that if a node gets down you still have the whole index and you can still serves search request Replicas are used to increase search performance and for fail-over.

And if you have a single node cluster then a replica shard is never going to be allocated on the same node where the related primary is (it would pretty much be like putting a backup on the same disk as the original data).

Upvotes: 1

dadoonet
dadoonet

Reputation: 14492

1) be careful with over allocation of shards.

2) elasticsearch will tell you with _cat/indices which indices are RED.

3) xpack has a free version

Upvotes: 1

Related Questions