Kibana instances increasing on Docker about kibana-docker

[jarpy]

Hi,

Thanks for the detailed report. Unfortunately, I haven't been able to replicate the behaviour.

Could you provide your kibana.yml? I'm guessing that the ID of your Kibana instance is changing with each docker-compose up.

Thanks.

[skearns64]

Upon startup, Kibana generates an instance UUID if one doesn't exist, which it stores in the Kibana data directory. If that directory is being cleared upon restart, then it will generate a new UUID.

Monitoring will show one "Kibana instance" per UUID, though note that it will only display Kibana Instance/UUIDs that have sent data in the currently selected time window. So if you are looking at the last 1 hour window (the default), it will show all nodes that had data in that window. If you wait an hour after your last restart of Kibana, you will see just one node displayed, I believe.

[mathewvino]

Thanks jarpy/sk

I dont have any kibana.yml --> based on the docker i am using the default which is present inside the docker kibana image which i don't provide any volume mapping.

Currently based on your suggestion i do provide a volume mapping to the kibana data where it create the UUID so my configuration looks like

• /c/Users/dockeroffice/kibanadata/data:/usr/share/kibana/data

Could you please confirm is that the right approach. I have tested multiple times after I stopped the docker-compose and restarted it seems to be working where it shows only only instance all the time

Below is the complete

version: '2'
services:
  kibana:
    image: docker.elastic.co/kibana/kibana:5.1.1
    ports:
      - 5601:5601
    networks:
      - esnet
    environment:
      SERVER_NAME: kibana.docker
      ELASTICSEARCH_URL: http://192.168.99.100:9200/
    volumes:
      - /c/Users/dockeroffice/kibanadata/data:/usr/share/kibana/data
  elasticsearch1:
    image: docker.elastic.co/elasticsearch/elasticsearch:5.1.1
    container_name: elasticsearch1
    environment:
      - node.name=docker-cluster
      - bootstrap.memory_lock=true
      - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
    ulimits:
      memlock:
        soft: -1
        hard: -1
      nofile:
        soft: 65536
        hard: 65536
    mem_limit: 1g
    cap_add:
      - IPC_LOCK
    volumes:
      - /c/Users/dockeroffice/elknode1/data:/usr/share/elasticsearch/data
    ports:
      - 9200:9200
    networks:
      - esnet
 
networks:
  esnet:
    driver: bridge

[jarpy]

I think that's a great solution. Nice work, @mathewvino, and thank you, @skearns64 for the tip.