Elasticsearch Single Node with Docker Compose

Balu
Balu
Elasticsearch Single Node with Docker Compose

Preface

This post assumes that you have some basic understanding of Docker/Podman, Docker Compose, and the key components used in the docker ecosystem. Get up to speed, with the Prepare Your Container Environment with Docker or Podman section of Docker docs.

  1. Install Docker or Podman
  2. install Docker Compose

Deploy Elasticsearch Single Node with Docker Compose

Create the docker-compose.yml with the following:

# docker-compose.yml
services:
  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch-wolfi:8.17.4
    container_name: elasticsearch
    environment:
      - cluster.name=oio-es-cluster
      - node.name=es-node
      - discovery.type=single-node
      - bootstrap.memory_lock=true
      - network.host=0.0.0.0
      - xpack.license.self_generated.type=trial
      - xpack.security.enabled=false
      - "ES_JAVA_OPTS=-Xms1024m -Xmx1024m"
    ports: ['9200:9200']
    volumes:
      - 'es_data:/usr/share/elasticsearch/data'
    healthcheck:
      test: curl -s http://localhost:9200 >/dev/null || exit 1
      interval: 30s
      timeout: 10s
      retries: 5
volumes:
  es_data:

Run

docker-compose up -d

Container Status

docker-compose ps -a
docker container ls
docker ps -a

Cluster Health & Stats APIs

Once the container is up, explore the cluster:

APICommand
Node listcurl -s localhost:9200/_cat/nodes?pretty
Cluster healthcurl -s localhost:9200/_cat/health?pretty
Cluster statscurl -s localhost:9200/_cluster/stats?human&pretty
Node statscurl -s localhost:9200/_nodes/stats?pretty
Specific nodecurl -s localhost:9200/_nodes/es-node/stats?pretty
List indicescurl -s localhost:9200/_cat/indices?pretty
All indicescurl -s localhost:9200/_cat/indices?expand_wildcards=all&pretty
Index-level statscurl -s localhost:9200/_nodes/stats/indices?pretty
Pluginscurl -s localhost:9200/_nodes/plugins

CRUD Operations

Create Index

Without mappings: Elasticsearch auto-generates field types on the first document (dynamic mapping).
With mappings: You define field types up front for better control (e.g., text vs keyword). 
curl -X PUT http://localhost:9200/ramayana_characters \
  -H "Content-Type: application/json" -d '
{
  "mappings": {
    "properties": {
      "name": { "type": "text" },
      "description": { "type": "text" }
    }
  }
}'
{"acknowledged":true,"shards_acknowledged":true,"index":"ramayana_characters"}

Insert Document

curl -X POST http://localhost:9200/ramayana_characters/_doc \
  -H "Content-Type: application/json" -d '
{
  "name": "Rama",
  "description": "Hero of the Ramayana, seventh avatar of Vishnu."
}'
{"_index":"ramayana_characters","_id":"_g1qMpYBhtVSA9-yauE0","_version":1,"result":"created", ...}

Copy the _id from the output — you’ll need it for the next queries.

Select All

curl -X GET 'http://localhost:9200/ramayana_characters/_search?pretty'
{ "hits": { "total": { "value": 1 }, "hits": [{ "_id": "_g1qMpYBhtVSA9-yauE0", "_source": { "name": "Rama", "description": "Hero of the Ramayana, seventh avatar of Vishnu." } }] } }

Select by ID

export DOC_ID=_g1qMpYBhtVSA9-yauE0
curl -X GET "http://localhost:9200/ramayana_characters/_doc/${DOC_ID}?pretty"

Update Document

curl -X POST "http://localhost:9200/ramayana_characters/_update/${DOC_ID}?pretty" \
  -H "Content-Type: application/json" -d '
{
  "doc": {
    "name": "Raama"
  }
}'
{"_index":"ramayana_characters","_id":"...","_version":2,"result":"updated"}

Try the GET call again to see the updated value.

Delete Document

curl -X DELETE "http://localhost:9200/ramayana_characters/_doc/${DOC_ID}?pretty"
{"_index":"ramayana_characters","_id":"...","_version":3,"result":"deleted"}

Homework

#TaskGoal
1Deploy & ExploreRun Docker Compose, use curl to hit every health API
2Create & QueryCreate an index, insert 5 documents, run _search and get-by-ID
3Understand MappingsCompare behaviour with and without explicit mappings
4Yellow Cluster?Investigate why health turns yellow after creating an index (hint: replica shards)

Conclusion

With this setup, you can quickly spin up a single-node Elasticsearch stack for development or learning purposes, and start exploring powerful search and analytics capabilities using RESTful APIs.