Elasticsearch How to Spin Up an OpenSearch Cluster Using Docker

Average Read Time

4 Mins

Elasticsearch How to Spin Up an OpenSearch Cluster Using Docker

Opster Expert Team - Gustavo

Nov-2021

Average Read Time

4 Mins

Opster Team

October 2021

Average Read Time

4 Mins


In addition to reading this guide, we recommend you run the Elasticsearch Health Check-Up. It will detect issues and improve your Elasticsearch performance by analyzing your shard sizes, threadpools, memory, snapshots, disk watermarks and more.

The Elasticsearch Check-Up is free and requires no installation.

Overview

OpenSearch was created as a community-driven, open source fork of Elasticsearch and Kibana. OpenSearch includes a search engine daemon and dashboards for visualization and user interface.

Opster is committed to the continuous improvement of open source technologies and sees the need for new tools for running OpenSearch. If you’d like to install OpenSearch clusters in AWS, you can do so using Opster’s new Terraform module.

In this article, we will explain all the steps needed to spin up an Opensearch cluster using docker.

Single node setup

The easiest way to start testing Opensearch is running the available docker image. 

Docker environment settings

Windows and Mac: In preferences set RAM to at least 4GB. Remember to press “Apply & Restart”.

Linux: Set vm.max_map_count to at least 262144.

Installation steps

To begin installation, run this command:

docker run -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" 
opensearchproject/opensearch:1.0.0

This will download the Opensearch image from docker and run it.

To test it, open a new terminal and run: 

curl -XGET https://localhost:9200 -u 'admin:admin' --insecure

The response should look like this:

{
  "name" : "abd86d8caffc",
  "cluster_name" : "docker-cluster",
  "cluster_uuid" : "N9h01U9ET6mFtDN6930R2Q",
  "version" : {
    "distribution" : "opensearch",
    "number" : "1.0.0",
    "build_type" : "tar",
    "build_hash" : "34550c5b17124ddc59458ef774f6b43a086522e3",
    "build_date" : "2021-07-02T23:22:21.383695Z",
    "build_snapshot" : false,
    "lucene_version" : "8.8.2",
    "minimum_wire_compatibility_version" : "6.8.0",
    "minimum_index_compatibility_version" : "6.0.0-beta1"
  },
  "tagline" : "The OpenSearch Project: https://opensearch.org/"
}

Opensearch incorporates SSL by default using self-signed certificates. This is one of the differences with Elasticsearch that doesn’t include SSL by default. In Opensearch TLS is optional for the REST layer and mandatory for the transport layer.

Note we use the –insecure flag to skip the local certificate validation, data is still being encrypted.

This flag should not be used in production. For this case, you might prefer to generate your own certificates with a CA (Certificate Authority). If you don’t have access to a CA and want to generate your own self-signed certificates for non-demo purposes, you can follow this guide.

Stopping the container

Find the docker container ID:

docker ps

To stop the container, run:

docker stop 

Multi-node setup

To deploy a more realistic scenario with multiple nodes, we will use docker-compose. This compose file will install Opensearch and Opensearch Dashboards.

Installation steps

  • Download the docker-compose.yml file from this link.
  • Run docker-compose up from the terminal in the docker-compose.yml folder (this may take some time).
  • Open the http://localhost:5601 for the dashboards.
  • Default login credentials are:
    • User: admin
    • Password: admin

Make sure your 5601 and 9200 ports are free (i.e not being used by Elasticsearch).

From DevTools using GET _cat/nodes we can confirm we just spun up a 2-node cluster with the following roles

  • data
  • ingestion
  • master
  • remote_cluster

Stopping the container

To stop the cluster, run:

docker-compose down

To stop the cluster and delete data volumes:

docker-compose down -v

Custom configuration

In both scenarios (docker and docker-compose) you can override the opensearch.yml configurations.

Docker

To override the configurations, you have to use docker -v flag to pass your custom opensearch.yml file: 

-v /path/to/custom-opensearch.yml:/usr/share/opensearch/config/opensearch.yml 

Full command would then be:

docker run -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" -v /path/to/custom-opensearch.yml:/usr/share/opensearch/config/opensearch.yml opensearchproject/opensearch:1.0.0
Docker-compose

For docker-compose you have to add a relative path to your custom opensearch.yml file in the services block. Download the docker-compose.yml from the multi-node installation section above and create a custom one from there.

services:
 opensearch-node1:
   volumes:
     - opensearch-data1:/usr/share/opensearch/data
     - ./custom-opensearch.yml:/usr/share/opensearch/config/opensearch.yml
 opensearch-node2:
   volumes:
     - opensearch-data2:/usr/share/opensearch/data
     - ./custom-opensearch.yml:/usr/share/opensearch/config/opensearch.yml
 opensearch-dashboards
   volumes:
     - ./custom-opensearch_dashboards.yml:/usr/share/opensearch-dashboards/config/opensearch_dashboards.yml

Performance Analyzer

You can also choose wether to enable the performance analyzer for Opensearch. This feature is descended from OpenDistro and runs outside the cluster, ensuring you the ability to solve issues if the cluster is in trouble.

Installation

Enable the Performance Analyzer plugin:

curl -XPOST https://localhost:9200/_plugins/_performanceanalyzer/cluster/config -H 
'Content-Type: application/json' -d '{"enabled": true}' -u 'admin:admin' -k

Enable the Root Cause Analyzer (RCA) framework:

curl -XPOST https://localhost:9200/_plugins/_performanceanalyzer/rca/cluster/config -H 
Content-Type: application/json' -d '{"enabled": true}' -u 'admin:admin' -k

Usage with perftop

To monitor your cluster visually you can use Perftop. Perftop is a lightweight linux tool that generates command line visuals. Perftop for OpenSearch includes the commands to interact with Performance Analyzer and some preset dashboards. Perftop is compatible with Linux and MacOS.

Installation

Download the binary file for your OpenSearch version (1.0 at the moment): https://github.com/opensearch-project/perftop/releases/tag/v1.0.0.0-rc1

Run this command:

./opensearch-perf-top-macos --dashboard ClusterOverview --endpoint localhost:9600

*NPM Version to be available

Dashboards

You can replace ClusterOverview with any of the preset dashboards perftop has available: 

  • ClusterNetworkMemoryAnalysis
  • ClusterThreadAnalysis
  • NodeAnalysis

Or create your own using a JSON file. Documentation for custom perftop dashboards can be found here.



Run the Check-Up to get a customized report like this:

Analyze your cluster