nagios-scripts

Collection of various Nagios scripts

check_kafka_lag.sh

Used to check the lag between to Kafka instnaces, one used as a "producer" and the other as a "consumer". Don't know if it's really useful outside my company but anyway, here it is

check_graphite.sh

I know that there are at least two other projects here on GitHub for the same purpose but they are both written in Ruby and I've developed an idiosincrasy towards that language, as a sysadmin (I won't rant here :P). Moreover, both of them are not flexible enough, cause I need dates range support and different calculation modes (they just return the last valid value). So here it is this simple bash script to check if a Graphite series is within two given thresholds. Check the program help for the options available

This script has a dependency on JSON.sh, you can find this great bash script here: https://github.com/dominictarr/JSON.sh

Nagios integration examples

checkcommands.cfg:

define command {
    command_name    check_graphite
    command_line    $USER1$/check_graphite.sh -g 'http://graphite.server/' -w $ARG1$ -c $ARG2$ -m "$ARG3$" -t $ARG4$ -s "$ARG5$" -e "$ARG6$" -a $ARG7$
}

!!! Pay attention to the double quotes !!!

Service definition

define service {
        check_command                  check_graphite!10!30!integral\(path.to.your.metric.\)!My_tag!30 min ago!now!avg
        host_name                      my.host.tld 
        use                            generic-service
        service_description            My description
        action_url                     /pnp/index.php?host=$HOSTNAME$&srv=$SERVICEDESC$
}

Optionally if you are using Puppet, this is the service definition to generate the above example

nagios_service { "graphite check 1":
        tag                 => "nagios-service",
        host_name           => $::hostname,
        service_description => "My description"
        use                 => "generic-service",
        check_command       => "check_graphite!10!30!integral\(path.to.your.metric.\)!My_tag!30 min ago!now!avg",
        action_url          => "/pnp/index.php?host=\$HOSTNAME\$&srv=\$SERVICEDESC\$"
}

!!! Remember to escape the parenthesis !!!

check_ndb

Previously in a separate repo, it is a scripts to monitor MySQL NDB Cluster from Nagios. It connects to the NDB Manager and check the memory used by the cluster for the specified node