Number of master nodes
The cluster has fewer master-eligible nodes than recommended for production resilience. With too small a master-eligible pool, losing the elected master can block cluster management until a new election can form a quorum.
For a complete list of insights, refer to AutoOps insights.
| Field | Value |
|---|---|
| Component | Elasticsearch |
| Severity | High |
| Scope | Cluster |
| Domains | configuration, high-availability, stability |
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.
The cluster has fewer master-eligible nodes than recommended for high availability.
AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.
More dedicated master nodes
Condition: Always shown for this insight.
With es-data-01 master nodes and es-data-01 data nodes, add masters until the cluster has three dedicated master nodes for high availability.
Each cluster has one elected master chosen from the master-eligible nodes. That master owns cluster-wide settings and actions such as creating or deleting indices, adding or removing nodes, and allocating shards.
If the elected master becomes unavailable, the remaining master-eligible nodes need a quorum to elect a replacement. Three master-eligible nodes is the usual production minimum so the cluster can survive a single master failure. Fewer nodes raise the chance that a restart or disconnect leaves the cluster cannot manage itself.