So I was trying to learn elasticsearch for work. So I was looking for some tutorials and I came across this official tutorial. I was going to setup elasticsearch and kibana for it and I thought it’s better to do this in Docker, since it’s easy to use. The issue with Elasticsearch and Kibana is, the two should be linked and that has to be defined by Docker. I had some problems with it and I even posted it on StackOverflow. As per the accepted answer by Andreas Jägle, there are two solutions.

Option 1. Link the two Containers directly

Step 1.1 Create the network

To create the medium for communication.

docker network create mynetwork --driver=bridge

Step 1.2 Run Elasticsearch Container

I’m running version Elasticsearch 2.4

docker run -d -p 9200:9200 -p 9300:9300 --name elasticsearch_2_4 --network mynetwork elasticsearch:2.4

### Step 1.3 Run Kibana Container I’m running version Kibana 4.6

docker run -d --network mynetwork -e ELASTICSEARCH_URL=http://elasticsearch_2_4:9200 -p 5601:5601 kibana:4.6

Note that I’m setting http://elasticsearch_2_4:9200 as the Elasticsearch URL and not localhost. This is where I went wrong.

Now run browser of http://localhost:5601 and it should work where you should see Kibana.

Option 2: Use docker-compose

There’s a tool called docker-compose created just for linking containers. This is a much easier way of doing this.

Step 2.1 Create docker-compose.yml

A file with above name should be created. It would look like this:

version: "2"

services:

  elasticsearch:
    image: elasticsearch:2.4
    ports:
      - "9200:9200"
    volumes:
      - ./esdata/:/usr/share/elasticsearch/data/

  kibana:
    image: kibana:4.6
    ports:
      - "5601:5601"
    environment:
      - ELASTICSEARCH_URL=http://elasticsearch:9200

Note that it’s the same things as above. But notice the differences,

1. No need to define the network since docker-compose handles it.
2. We’re passing a folder so that the saved folder is persisted in the /esdata/ folder in the host machine, so that when we restart docker, the saved data is persisted in the file system.

Step 2.2 Run docker-compose

Just call,

docker-compose up

That’s it! Now run browser of http://localhost:5601 and it should work where you should see Kibana.

Bonus: Use Kibana with Sense

The previous Kibana is the vanilla one. There’s a cool tool called Sense which is an plugin for Kibana. So, to use the pre-built Kibana with sense, I found an alternate Docker image having Kibana with sense

As a result the docker-compose.yml looks like this:

version: "2"

services:

  elasticsearch:
    image: elasticsearch:2.4
    ports:
      - "9200:9200"
    volumes:
      - ./esdata/:/usr/share/elasticsearch/data/

  kibana:
    image: seeruk/docker-kibana-sense:4.5
    ports:
      - "5601:5601"
    environment:
      - ELASTICSEARCH_URL=http://elasticsearch:9200

Cheers