Just some random curl command that help me with the administration and the configuration of an Elasticsearch cluster.

List of cats:

http://localhost:9200/_cat/

List indices:

http://localhost:9200/_cat/indices?v

Health of the system (should contain the word green)

http://localhost:9200/_cat/health

http://localhost:9200/_all/_settings

The health of the cluster:

http://localhost:9200/_cluster/health

Cluster settings

curl http://localhost:9200/_cluster/settings

Delete index:

curl  -X DELETE http://localhost:9200/index-name

Set temporary (transient) the low disk watermark

curl -X PUT http://localhost:9200/_cluster/settings  -H "Content-Type: application/json"  -d '{ "transient": {"cluster.routing.allocation.disk.watermark.low": "5%" } }'

Set the number of replicas to 1

curl -X PUT http://localhost:9200/experimental/_settings  -H "Content-Type: application/json"  -d '{ "index": {"number_of_replicas" : 1} }'

Tell the index to have no replicas. That is to have exactly one copy of each piece of data. No redundancy. Useful if you only have a single node in your "cluster".

curl -X PUT http://localhost:9200/experimental/_settings  -H "Content-Type: application/json"  -d '{ "index": {"number_of_replicas" : 0} }'