check_memcached.pl

Project Notes

This plugin checks Memcached and measures response which can be used for threshold checks. It also retrieves various statistics data available with memcache 'stats' command (what you get when you telnet to the port and do 'stats') and allows to set thresholds either on their direct values or on the rate of change of those variables. Plugin calculates and allows to set threshold on useful statistics like Hitrate (based on rate of change unlike all other plugins that do it based on totals for all time) and Memory Utilization. All data can become performance data for graphing and pnp4nagios template is available.

For help on what parameters this plugin accepts you can just do ./check_memcached.pl --help

Extensive documentation is also available in the header of plugin perl code that you should read.

Sample command and service definitions are below:

define command {
command_name check_memcached
command_line $USER1$/check_memcached.pl -H $HOSTADDRESS$ -p $ARG1$ -T $ARG2$ -R $ARG3$ -U $ARG4$ -a curr_connections,evictions -w ~,~ -c ~,~ -f -A 'utilization,hitrate,response_time,curr_connections,evictions,cmd_set,bytes_written,curr_items,uptime,rusage_system,get_hits,total_connections,get_misses,bytes,time,connection_structures,total_items,limit_maxbytes,rusage_user,cmd_get,bytes_read,threads,rusage_user_ms,rusage_system_ms,cas_hits,conn_yields,incr_misses,decr_misses,delete_misses,incr_hits,decr_hits,delete_hits,cas_badval,cas_misses,cmd_flush,listen_disabled_num,accepting_conns,pointer_size,pid' -P "$SERVICEPERFDATA$"
}

# Arguments and thresholds are:
# ARG1 : Port
# ARG2 : Hitrate Threshold. Below it is <60% for warning, <30% for critical
# ARG3 : Response Time Threshold. Below it is >0.1s for WARNING, >0.2s for critical
# ARG4 : Utilization/Size Threshold. Below it is >95% for warning, >98% for critical

define service {
use prod-service
service_description Memcached: Port 11212
check_command check_memcached!11212!'>0.1,>0.2'!'<60,<30'!'>95,>98'
hostgroups memcached
}