Can you answer a question regarding backing up an indexer cluster?

[christay]

Hi,

I am quite new to the Splunk setup, so as such, pardon my question here.

Given that, if i have a 2 x indexer setup in a Cluster, is it possible for me to shutdown 1 of the indexers ( Save cost as it's running on AWS) and still get the latest data replicated into the shutdown indexer from the active indexer ?

So, in the event that the primary active indexer is down, I can bring up the shutdown indexer to take over the role?

Is the above scenario possible ?

Or do I definitely have to have both the indexer instance online to get the latest data ingested ?

Thanks

[chrisyounger]

Yes indexer clustering requires all indexers only.

If you are on AWS and looking to save money, consider using smart store: https://www.splunk.com/blog/2018/10/11/splunk-smartstore-cut-the-cord-by-decoupling-compute-and-stor... . This will store the data in S3 buckets which different indexers can basically share.

[richgalloway · Accepted solution]

An indexer must be running to receive replicated buckets. While an indexer is down, the cluster master (CM) will keep track of the buckets that need to be replicated to that indexer. Once the indexer comes back up, it will receive replicated buckets in the background while processing new events and search requests.

Indexer clustering is not a cold-backup solution.