Old-style Health Checks


As of rev. 0.98 a new health check model was introduced in order to simplify the handling of plugins. Most of the "old" health checks have already been ported to the new model and new-style health checks should always be prefererred over their old implementation. Please refer to the testers command, to the list of new-style checks, and to the agent configuration section.

Table of Contents

cpuload - check the CPU load as reported by the 'uptime' command
statusfile - read status information from a file and report it to the Status Collector
bbscript - use BB style monitor script
http - check http response
tcp - check if the host does respond to tcp connection request
ping - does a ping
rpc - does an 'rpc ping
procs - checks for running processes or services (Win32)
diskfree - checks file systems for free space
diskload - check the average disk load (4 minutes) as reported by 'sar'
load - check for CPU idle time, I/O-wait, freeswap as reported by sar (4 minutes period)
dumpdates - check for last backup if using dump/ufsdump
syslog - check system log files
eventlog - check event log on Win32 systems
bbdisplay - heck the log file being generated by bstrapd (var/snmp_traplog)
snmp - remotely monitor hosts running SNMP agents
OV - monitor HP Openview trapd.log
metastat - monitor Solstice Disksuite metadevices
ntp - check if the machine is running an ntp server
mrtg - receive and visualize mrtg data
atmport or etherport - check operating status of specific ports
software - get firmware release via SNMP and check against an expected version
realhttp - execute an HTTP(S) request for a specified url
memory - report memory usage on NT and Linux systems
oracle - tests an oracle database server
tripwire - check system consistency via tripwire
ldap - query an ldap server and attempt to retriew a register
command - execute an arbitrary command
ups - query an uninterruptible power supply
qmqueue - monitor qmail's mail queue