Deployment Architecture

How to migrate from an indexer clustering to a standalone environment?

Path Finder

Weird question, but does anyone have a graceful procedure to go from clustered environment to stand alone?

Current setup:

1 master node
2 indexers clustered
1 stand alone search head

Company wants to go with:
2 search heads and 1 indexers to prepare for a multisite cluster down the road. Master Node will just sit idle for now. I know there's an "offline" command that we can run to take out the indexer from the cluster and then just need to re-point the search head to the correct peer, but I wanted to see if there's a graceful way to do this or if I'm going to run into any issues using the steps above.


0 Karma

Splunk Employee
Splunk Employee

If you're going to have multi-site Clustering, why do you need to make it non-Indexer Clustering env?

Probably the following steps would work. I strongly recommend to test it in a test env before trying it if possible.

  1. Change CM's RF and SF to 1
  2. Restart CM
  3. Remove excess buckets to make sure only one copy per bucket
  4. Stop CPs(Indexers) and remove Clustering settings
  5. Move apps from etc/slave-apps to etc/apps
  6. Start the Indexers
  7. Stop SH and remove clustering configurations
  8. Start SH and add those indexers as Search Peers from Splunkweb
  9. Stop CM