Question

AWS Elasticsearch Backups

  • 18 August 2023
  • 2 replies
  • 129 views

Badge +6

There is a requirement in our AWS environment to perform Backups of the Elasticsearch nodes.  From what I have read, CV does not actually support that and the recommendations are to use the Elasticsearch imbedded snapshots and from there we CV could backup up from the S3 buckets.

“Elasticsearch has consistency requirements that span the whole cluster. You must use the built-in snapshot functionality for cluster backups.”

Is this still true?  Are there other ways to perform backups using CommVault?

 

Thanks

Chuck 


2 replies

Userlevel 7
Badge +23

Hey @ChuckC,

Yes there is no discrete integration for ElasticSearch. In terms of best practice, we’d have to rely on the vendors best practices to protect it. It does make sense what they say, especially if its designed as a distributed database where consistency between nodes is key. The same way that allows VSS to capture a consistent file system backup by temporarily suspending I/O, they have most likely built into their built-in snapshot functionality to guarantee consistency, which cannot be guaranteed with other methods. Like you said, we can protect data in an S3 bucket, so that could be a good option to safeguard the backup as a secondary copy.

 

It would be good to test the recovery scenario if using this method to ensure it will actually work given the lack of discrete integration.

Badge +6

Thanks Damian.  My first step is to find out what APPS owner really want to be protected.  I do appreciate you taking the time out to research and respond to my question.

 

Chuck

Reply