Deployment Architecture

Why in a 4 site Search Head Cluster captain is getting elected automatically?

schose
Builder

Hi Forum,

I'm integrating a streched SHC running in 4 sites. Each site have 1 SHC Member. When booting the SHC up the captain is elected automatically.

from the docs i learned:
http://docs.splunk.com/Documentation/Splunk/6.6.7/DistSearch/DeploymultisiteSHC

"If the site with a majority of members fails, the remaining members on the minority site cannot elect a new captain. Captain election requires the vote of a majority of members, but only a minority of members are running. The cluster does not function. See "Consequences of a non-functioning cluster."

"If the network between sites fails, the members on each site will attempt to elect a captain. However, only a site that holds a majority of the total members will succeed. That site can continue to function as the cluster indefinitely."

So as I'm having 4 shc members in 4 sites I never have any kind of majority in any site. So I would guess captain election should never work correctly. Any hint why it's working correctly?

Best regards,

Andreas

0 Karma
1 Solution

Steve_G_
Splunk Employee
Splunk Employee

The doc is describing a failure in the the common case of a two-site cluster. The key point is that you must have a majority of members, across all sites, up-and-running and in communication, for the cluster to function. This is also discussed at length elsewhere in the docs. (See, for example, http://docs.splunk.com/Documentation/Splunk/7.1.2/DistSearch/SHCarchitecture .)

As long as you have a majority (>50%) of the entire set of cluster members running and communicating, the cluster can function. It doesn't matter if they're on one site or a hundred. So, for the deployment you describe, you're fine, as long as you don't suffer the loss of two sites, which would cause you to lose the majority.

View solution in original post

Steve_G_
Splunk Employee
Splunk Employee

The doc is describing a failure in the the common case of a two-site cluster. The key point is that you must have a majority of members, across all sites, up-and-running and in communication, for the cluster to function. This is also discussed at length elsewhere in the docs. (See, for example, http://docs.splunk.com/Documentation/Splunk/7.1.2/DistSearch/SHCarchitecture .)

As long as you have a majority (>50%) of the entire set of cluster members running and communicating, the cluster can function. It doesn't matter if they're on one site or a hundred. So, for the deployment you describe, you're fine, as long as you don't suffer the loss of two sites, which would cause you to lose the majority.

View solution in original post

Did you miss .conf21 Virtual?

Good news! The event's keynotes and many of its breakout sessions are now available online, and still totally FREE!