Loading

No Shards In Data Node

One or more data nodes have no shards allocated. That can be intentional during maintenance or decommissioning when data was drained first.

Note

For a complete list of insights, refer to AutoOps insights.

Field Value
Component Elasticsearch
Severity Low
Scope Node
Domains performance, resource-utilization, shard-allocation

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.

12 data node was detected as not containing any shards, meaning it doesn’t have any data on it. The affected node is: es-data-01 and es-data-02

Note

AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.

Enable shard allocation

Condition: Always shown for this insight.

Reset cluster.routing.allocation.enable so primary and replica shards can allocate to nodes.

				PUT _cluster/settings
					{
  "transient": {
    "cluster.routing.allocation.enable": null
  },
  "persistent": {
    "cluster.routing.allocation.enable": null
  }
}
Note

Requires the manage cluster privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Impact: A node with no shards, meaning no data, could be configured such intentionally or unintentionally. If it is intentional, no further action is required. There can be a few reasons why a data node could have no shards. The two most common reasons are:

  1. A hardware change needs to be made and to prevent data loss, data was moved out of the node before the maintenance.
  2. A node needs to be decommissioned and to prevent data loss, data was moved away from the node before shutting it down. Nodes might also lack shards due to bad configuration, for instance, bad allocation routing or previously valid allocation routing but the underlying node configurations (IP range, name, tags, and so on) changed for some reason.