Deployment Architecture

Distributed KV store with 2 search head?

janlar
Explorer

Hello.

I'm running on RHEL 7 with 6.6.3 and an Indexer cluster (3 peers), and have 2 Search Heads not in a SHC but connected induvidually to the index cluster.
I try to use KV store with a custom built app where I store data in a Collection (Collections.conf and transforms.conf) that I wish to share with my other search head, but I'm failing and don't know how to proceed. My app is installed on both SH, it works locally on both, but my KV store instance seems locally even though I have set 'replicate = true' in my Collection stanza and 'replication_host = FQDN' in server.conf.

Reading about KV store I can't really find out if it should work unless you have a SHC, but on the other hand I can't find information it wouldn't.
is it possible to share a KV store whitout a SHC?
How do I configure the KV store to be stored at the index cluster and not locally?

The part of the 'replication_host' is a bit un clear what to type there on each host, but the description in server.conf lets me Believe it should be the local hosts FQDN?

Regards,
Jan

0 Karma
1 Solution

outcoldman
Communicator

Hi Jan,

As far as I know, It is not possible (or better to say is not supported) what you want to implement.

replicate = true in collections.conf actually only makes this collections to be available directly on indexers, when you perform search from this search head, it does not affect KVStore replication between Search Heads.

Any reasons why you don't want to enable Search Head Cluster? Seems like you are looking on the replication between Search Heads, and SHC can provide you that.

Also one note, if you will decide to go with SHC - you want to setup minimum 3 nodes for reliability and consistency.

View solution in original post

0 Karma

janlar
Explorer

Thanx.

I guess i'll have to surrender to the SHC anyway. The reason for not going that path straight away is internal politics based on security and delegation by duty. Simply I wanted to have separate instances for some functions. No big deal but good you confirmed what I believed that it' not possible.

0 Karma

outcoldman
Communicator

Hi Jan,

As far as I know, It is not possible (or better to say is not supported) what you want to implement.

replicate = true in collections.conf actually only makes this collections to be available directly on indexers, when you perform search from this search head, it does not affect KVStore replication between Search Heads.

Any reasons why you don't want to enable Search Head Cluster? Seems like you are looking on the replication between Search Heads, and SHC can provide you that.

Also one note, if you will decide to go with SHC - you want to setup minimum 3 nodes for reliability and consistency.

0 Karma
Get Updates on the Splunk Community!

.conf24 | Registration Open!

Hello, hello! I come bearing good news: Registration for .conf24 is now open!   conf is Splunk’s rad annual ...

Splunk is officially part of Cisco

Revolutionizing how our customers build resilience across their entire digital footprint.   Splunk ...

Splunk APM & RUM | Planned Maintenance March 26 - March 28, 2024

There will be planned maintenance for Splunk APM and RUM between March 26, 2024 and March 28, 2024 as ...