Elasticsearch Nagios Checks

Overview

This is a simple Nagios check script to monitor all nodes of your Elasticsearch Cluster.

Authors

Main Author

Fedele Mantuano (Twitter: @fedelemantuano)

Installation

In your Nagios plugins directory run

git clone https://github.com/fedelemantuano/nagios-plugin-elasticsearch.git

Usage

Install in Nagios (Example Heap Used Percent)

Edit your commands.cfg and add the following

define command {
    command_name    check_heap_used_percent
    command_line    /opt/nagios-plugin-elasticsearch/check_elasticsearch.py -D -n $ARG1$ -c $ARG2$ node --heap-used-percent
}

(add -D to the command if you want to add perfdata to the output)

(add -G to the command if you want only output)

(add -D -G to the command if you want only output and graph)

Then you can reference it like the following.

Check Heap Used Percent

define service {
    use                     generic-service
    host_name               node_cluster
    service_description     ELASTICSEARCH - Node Heap used percent
    check_command           check_heap_used_percent!node_name!client_node
}

Help

check_elasticsearch.py

check_elasticsearch.py [-h] [-n NODE_NAME] [-c CLIENT_NODE] [-D] [-G]
                              [-v]
                              {cluster,node,indices} ...

Elasticsearch Nagios checks

positional arguments:
  {cluster,node,indices}
                        All Elasticsearch checks groups
    cluster             All Cluster checks
    node                All Node checks
    indices             All indices checks

optional arguments:
  -h, --help            show this help message and exit
  -n NODE_NAME, --node-name NODE_NAME
                        Node name in the Cluster
  -c CLIENT_NODE, --client-node CLIENT_NODE
                        Client node name (FQDN) for HTTP communication
  -D, --perf-data       Enable Nagios performance data (Valid for all checks
                        groups)
  -G, --only-graph      Enable Nagios to print only message
  -v, --version         show program's version number and exit

check_elasticsearch.py cluster

check_elasticsearch.py cluster [-h] [--cluster-health]

optional arguments:
  -h, --help        show this help message and exit
  --cluster-health  Check the Cluster health (green, yellow, red)

check_elasticsearch.py node

check_elasticsearch.py node [-h] [--heap-used-percent]                                                     
                                   [--documents-count]
                                   [--ratio-search-query-time]

optional arguments:
  -h, --help            show this help message and exit
  --heap-used-percent   Check the Heap used percent
  --documents-count     Documents on node
  --ratio-search-query-time
                        Ratio search query_time_in_millis/query_total

check_elasticsearch.py indices

check_elasticsearch.py indices [-h] [--index INDEX] [--prefix PREFIX]
                                      [--doc-type DOC_TYPE] [--last-entry]

optional arguments:
  -h, --help           show this help message and exit
  --index INDEX        Index name
  --prefix PREFIX      Include only indices beginning with prefix
  --doc-type DOC_TYPE  Include only documents with doc-type
  --last-entry         Check last entry in the index. Only for timestamp in
                       UTC