첫 클러스터 설정

Let's get a single-node Elasticsearch cluster running locally — your launchpad for experimenting with indexing and search.

Elasticsearch runs on Java, so you'll need a JRE — recent versions want Java 11 or later. The command below checks your version.

Grab Elasticsearch from the official Elastic site: a .zip for Windows, a .tar.gz for Linux or macOS. Pick the latest stable release.

Extract the archive to create the Elasticsearch directory. On Linux or macOS the tar command does it — just swap in your version number.

Inside the unpacked folder, the key directories are bin/ (executables), config/ (settings), data/ (your index), and logs/.

For a single node, defaults are usually fine. The main config file is config/elasticsearch.yml, where cluster.name and node.name live.

Time to start the node: cd into the directory and run the startup script. Elasticsearch runs in the foreground, so keep this terminal open.

Verify it's up by hitting its default port, 9200, from a new terminal with the curl command below.

A healthy node returns JSON with cluster_name, node_name, the version, and the tagline You Know, for Search — proof your node is live.

To shut down gracefully, go back to the terminal running Elasticsearch and press Ctrl+C. Closing the window may leave it running in the background.

Done! You installed, started, verified, and stopped your first single-node cluster. Next, we'll index and search real documents.