Cluster Concurrent Rebalance Low
Cluster concurrent rebalance is below the recommended value, so shard moves proceed slowly when disks are uneven. Nodes nearing full disk might stay hot longer, risking yellow or red allocation and write blocks at flood stage.
For a complete list of insights, refer to AutoOps insights.
| Field | Value |
|---|---|
| Component | Elasticsearch |
| Severity | Medium |
| Scope | Cluster |
| Domains | performance, configuration, shard-allocation |
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 cluster concurrent rebalance | Integer | 2 |
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 current concurrent rebalance is set to: 12.
The recommended setting is: 12.
AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.
Increase concurrent rebalance limit
Condition: Always shown for this insight.
cluster.routing.allocation.cluster_concurrent_rebalance controls how many shards rebalance at once. Raise it with the action below to finish rebalancing sooner.
PUT _cluster/settings
{
"persistent": {
"cluster.routing.allocation.cluster_concurrent_rebalance": 12
}
}
Requires the manage cluster privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.
Impact: The cluster might be cannot rebalance shards away from a node with a disk that is close to full. Some nodes might be cannot allocate shards which could result in the cluster turning yellow or red and not being able to write new data to certain indices. This setting determines the maximum number of shards the cluster can move at any one time to rebalance the distribution of shards across the nodes, obeying any existing shard allocation rules such as allocation filtering and forced awareness. If the setting is set too low, the cluster will be cannot move shards from a full disk to another node with more disk space. This means that the disk might continue to fill up and reach the disk flood stage threshold, at which point it will no longer be possible to write to the index.