How to Check Cluster Health in Elasticsearch Quickly

To check cluster health in Elasticsearch, use the GET /_cluster/health API endpoint. This returns the cluster status such as green, yellow, or red, indicating the overall health of your cluster.

Syntax

The basic syntax to check cluster health is a simple HTTP GET request to the /_cluster/health endpoint. You can add optional parameters to filter or get detailed info.

• GET /_cluster/health: Returns overall cluster health.
• GET /_cluster/health/{index}: Returns health for a specific index.
• Optional query parameters include level (cluster, indices, or shards) and timeout.

http

GET /_cluster/health

# Optional parameters:
# level=cluster|indices|shards
# timeout=30s

Example

This example shows how to check the overall cluster health using curl. The response includes the cluster status, number of nodes, and shard info.

bash

curl -X GET "localhost:9200/_cluster/health?pretty"

Output

{
  "cluster_name" : "my_cluster",
  "status" : "green",
  "timed_out" : false,
  "number_of_nodes" : 3,
  "number_of_data_nodes" : 3,
  "active_primary_shards" : 10,
  "active_shards" : 20,
  "relocating_shards" : 0,
  "initializing_shards" : 0,
  "unassigned_shards" : 0,
  "delayed_unassigned_shards" : 0,
  "number_of_pending_tasks" : 0,
  "number_of_in_flight_fetch" : 0,
  "task_max_waiting_in_queue_millis" : 0,
  "active_shards_percent_as_number" : 100.0
}

Common Pitfalls

Common mistakes when checking cluster health include:

• Not specifying the correct Elasticsearch host or port, causing connection errors.
• Ignoring the status field which shows green, yellow, or red status.
• Assuming yellow means failure; it means some replicas are not allocated but primary shards are fine.
• Not using the ?pretty parameter for readable JSON output during manual checks.

bash

curl -X GET "localhost:9200/_cluster/health"

# Wrong: Using wrong port or host
curl -X GET "localhost:9201/_cluster/health"

# Right: Correct host and port
curl -X GET "localhost:9200/_cluster/health?pretty"

Quick Reference

Parameter	Description	Example
GET /_cluster/health	Get overall cluster health	curl -X GET "localhost:9200/_cluster/health?pretty"
GET /_cluster/health/{index}	Get health for specific index	curl -X GET "localhost:9200/_cluster/health/my_index?pretty"
level	Detail level: cluster, indices, shards	GET /_cluster/health?level=indices
timeout	Wait time for response	GET /_cluster/health?timeout=30s

Key Takeaways

Use the GET /_cluster/health API to quickly check Elasticsearch cluster status.

Cluster status can be green (healthy), yellow (some replicas unassigned), or red (issues).

Add ?pretty to the request URL for easier-to-read JSON output.

Check specific indices by adding the index name to the endpoint.

Ensure you connect to the correct Elasticsearch host and port to avoid errors.