Installing a secondary Monitoring Clusteredit

First, prepare the monitoring cluster — the cluster you will use to store and analyze Marvel’s data. For this example, we will assume two hosts, es-mon-1 and es-mon-2, are in the monitoring cluster.

The monitoring clusteredit

For each node in the monitoring cluster:

  1. Prevent the nodes in the monitoring cluster from generating their own stats, by adding the following line to config/elasticsearch.yml:

    marvel.agent.enabled: false
  2. Install the plugin. From the Elasticsearch home directory run

    bin/plugin -i elasticsearch/marvel/latest
  3. Restart Elasticsearch

The production clusteredit

  1. Tell each node where to send its stats by adding this line to `config/elasticsearch.yml:

    marvel.agent.exporter.es.hosts: ["es-mon-1:9200","es-mon-2:9200"]

    This line adds two hosts of your monitoring cluster to receive the data, a primary and a backup.

  2. Install the plugin. From the Elasticsearch home directory run

    bin/plugin -i elasticsearch/marvel/latest
  3. Restart Elasticsearch

Once the plugin is installed on all nodes, you can access the Marvel UI by opening this link with any modern browser.

http://any-server-in-monitoring-cluster:9200/_plugin/marvel/

It may take a minute or two for data to appear.

You may want to temporarily disable shard allocation before you restart your nodes to avoid unnecessary shard reallocation during the install process.