Best practice for backup/restore per tenant in a multi-tenant cluster

Elastic Stack Elasticsearch
1

In a multi-tenant cluster, where each document has a tenant id , what is the best practice for backup/restore of data specific to a tenant?

The use case us that we want to backup all the documents for a specific tenant T1at a specific point of time and later want to restore that snapshot of data from for T1 that will overwrite all the current data specific to T1.

We cannot do index snapshot/restore since the same index is shared by multiple tenants. We can do a do a scan/scroll to quickly read the data, but is there a best practice on how to store the data and restore that snapshot that overwrites the current data for a specific tenant?

Thanks for any info regarding this.

2

Didn't test this but maybe you can try it. You can create alias based on T1 so that alias has only data for tenant T1. Then take the snapshot on that alias. Not sure if ES support this but something worth trying. Good luck!

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.