How to set Elasticsearch node names

Meaningful node names make Elasticsearch clusters easier to operate by turning shard allocation, logs, and monitoring dashboards into readable, human-friendly outputs instead of a blur of hostnames and IPs.

Elasticsearch reads the node.name setting from /etc/elasticsearch/elasticsearch.yml during startup and advertises it to the cluster. When node.name is unset, the node name defaults to the system hostname and is shown in APIs such as /_cat/nodes and in log messages.

Node names must be unique within a cluster, and renaming requires a service restart to take effect. Changing node.name updates the label used for display and monitoring, but does not change the persistent node ID stored under the node’s data path, so a rename can still confuse inventory tooling if naming conventions are inconsistent.

1. Open the Elasticsearch configuration file.

   $ sudo nano /etc/elasticsearch/elasticsearch.yml

2. Set a unique node.name value in the configuration.

   node.name: node-01

   Keep the name unique within the cluster, and prefer stable names that match monitoring labels (for example tier + sequence).

3. Restart the Elasticsearch service to apply the new name.

   $ sudo systemctl restart elasticsearch

   Restarting a single-node cluster stops search and indexing until Elasticsearch is running again, and restarting multiple nodes at once can make shards temporarily unavailable.

   Related: How to manage the Elasticsearch service with systemctl in Linux

4. Verify the node name in the cluster node list.

   $ curl --silent --show-error --cacert /etc/elasticsearch/certs/http_ca.crt --user elastic "https://localhost:9200/_cat/nodes?v&h=name,ip,node.role,master"
   Enter host password for user 'elastic':
   name    ip         node.role   master
   node-01 192.0.2.40 cdfhilmrstw *

   Clusters without TLS or authentication can use http://localhost:9200 and omit --cacert and --user.