Dedicated master nodes
Your cluster is large enough that dedicated master nodes are recommended. When data nodes also run master duties under load, cluster management can lag and stability suffers.
For a complete list of insights, refer to AutoOps insights.
| Field | Value |
|---|---|
| Component | Elasticsearch |
| Severity | Low |
| Scope | Cluster |
| Domains | performance, configuration, high-availability |
You can customize these settings to adjust when AutoOps detects this event and presents the insight. Refer to AutoOps event settings for details.
The default customization settings are:
| Setting | Type | Default |
|---|---|---|
| Minimum data node count to recommend dedicated masters | Integer | 4 |
The following is an example of what you might see when this insight is triggered. Real insights use live data and links from your deployment or cluster.
Data nodes that also hold the master role: 12
AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.
Add dedicated master nodes
Condition: Always shown for this insight.
Add three dedicated master nodes so the cluster stays quorum-safe if one master restarts or disconnects.
The elected master manages cluster metadata: creating and deleting indices, deciding shard allocation, publishing the cluster state, and running lifecycle chores such as Index Lifecycle Management (ILM) and Snapshot Lifecycle Management (SLM). When that work shares a node with heavy indexing or search, the master can run short of resources.
Symptoms can include slow index creation, delayed shard assignment, slower cluster-state updates, and lagging ILM or SLM steps. Dedicated master-eligible nodes that do not hold data keep cluster management off the data path.