check_mk for nagios 笔记

继吉

　　all WATO config files:
　　/usr/local/check_mk/etc/conf.d/wato
/usr/local/check_mk/etc/multisite.d/wato

　　

　　Global settings tuning
　　
　　
　　

　　add host/service/contact to hostgroups/servicegroups/contactgroups
　　

　　1. create hostgroups/servicegroups/contactgroups

　　

　　2. add host/service/contact to hostgroups/servicegroups/contactgroups

　　

　　3. for add contact to contactgroups

　　

　　4. apply changes

　　

　　SNMP Checks config
　　

　　1. on Cisco router/switch to set snmp community:
snmp-server community public ro

2. on Check_MK to check snmp manually:
snmpwalk -v 2c -c public cisco_router_ip

3. on Check_mk server WATO

　　

　　
　　Notes: default community is "public", you can config individually in WATO

　　/usr/local/check_mk/bin/cmk -D cisco3745 to check it's config or in WATO
　　

　　Add just ping check for devices
　　

　　1. add hosts in WATO

　　

　　5. check it in WATO

　　

　　

　　Host Tags
　　
1. create host tags in WATO

　　

　　Monitoring of Logfiles with Logwatch
　　

　　1. make sure to install logwatch rpm
　　rpm -ivh check_mk-agent-1.2.4-1.noarch.rpm
rpm -ivh check_mk-agent-logwatch-1.2.4-1.noarch.rpm
　　

　　2. vi /etc/check_mk/logwatch.cfg
/var/log/messages
　　I OK              # lines containing OK are ignored, priority from top to bottom

　　C Critical        # lines containing Critical

　　W Warn            # lines containing Warn

　　W Error           # lines containing Error

　　W Fail            # lines containing Fail

　　

　　/var/log/maillog
　　I OK              # lines containing OK are ignored, priority from top to bottom

　　C Critical        # lines containing Critical

　　W Warn            # lines containing Warn

　　W Error           # lines containing Error

　　W Fail            # lines containing Fail
　　
/usr/local/nagios/var/nagios.log
I OK              # lines containing OK are ignored, priority from top to bottom
　　C Critical        # lines containing Critical

　　W Warn            # lines containing Warn

　　W Error           # lines containing Error

　　W Fail            # lines containing Fail
　　Notes: it's case sensitive for messages, so you should define it carefully

　　

　　3. You can also use WATO
　　/usr/local/check_mk/bin/cmk -I localhost
　　vi /usr/local/check_mk/var/autochecks/localhost.mk to remove unwanted item as you needed

　　/usr/local/check_mk/bin/cmk -R
　　

　　4. you can test it with logger command
　　logger "this is a Error message"

　　

　　Monitoring MySQL
　　

　　/usr/local/check_mk/bin/cmk -d localhost | head

　　PluginsDirectory: /usr/lib/check_mk_agent/plugins
LocalDirectory: /usr/lib/check_mk_agent/local
SpoolDirectory: /etc/check_mk/spool
AgentDirectory: /etc/check_mk
　　

　　1. cp -a /usr/local/check_mk/share/agents/plugins/mk_mysql /usr/lib/check_mk_agent/plugins/
　　chmod 755 /usr/lib/check_mk_agent/plugins/*
　　

　　2. You can also use WATO
　　/usr/local/check_mk/bin/cmk -I localhost
　　vi /usr/local/check_mk/var/autochecks/localhost.mk to remove unwanted item as you needed

　　/usr/local/check_mk/bin/cmk -R
　　

　　

　　Monitoring any linux job
　　

　　1. please refer http://hj192837.blog.运维网.com/655995/1352370 for testing script

2. vi /etc/cron.d/check-cisco
MAILTO=
0 10,17 * * * root mk-job ping-cisco-job /tmp/pingtest.sh

service crond restart

3. after the first run of the job the only thing to do is to run a service inventory on the host and restart your monitoring process.
You can also use WATO
/usr/local/check_mk/bin/cmk -I localhost
vi /usr/local/check_mk/var/autochecks/localhost.mk to remove unwanted item as you needed
/usr/local/check_mk/bin/cmk -R

　　

　　Setup Distributed Monitoring
　　

　　1. on distributed monitored check_mk server
　　cp -a /usr/local/check_mk/bin/unixcat /usr/bin
　　vi /etc/xinetd.d/livestatus
　　service livestatus
{
    type        = UNLISTED
    port        = 6557
    socket_type    = stream
    protocol    = tcp
    wait        = no
# limit to 100 connections per second. Disable 3 secs if above.
    cps             = 100 3
# set the number of maximum allowed parallel instances of unixcat.
# Please make sure that this values is at least as high as
# the number of threads defined with num_client_threads in
# etc/mk-livestatus/nagios.cfg
        instances       = 500
# limit the maximum number of simultaneous connections from
# one source IP address
        per_source      = 250
# Disable TCP delay, makes connection more responsive
    flags           = NODELAY
    user        = nagios
    server        = /usr/bin/unixcat
    server_args     = /usr/local/nagios/var/rw/live
# configure the IP address(es) of your Nagios server here:
#    only_from       = 127.0.0.1 10.0.20.1 10.0.20.2
    disable        = no
}
　　

　　2. service xinetd restart
　　

　　3. on distributed monitoring check_mk server

　　

　　4. now it's ok, only works for Perf-O-Meter column for PNP graph in current version

　　

　　5. using mod_proxy to get PNP graph -- undergoing

　　

　　Setup Legacy Checks (no agent or SNMP on client)

　　

　　1. on Nagios check_mk server
　　vi /etc/hosts
　　192.168.1.11    linuxtestclient
　　

　　vi /usr/local/check_mk/etc/main.mk
　　all_hosts = [ 'localhost','linuxtestclient' ]
　　#Legacy Checks
extra_nagios_conf += r"""
# You can get command definition from nagios command.cfg
define command {
    command_name    check_ssh
    command_line    $USER1$/check_ssh $ARG1$ $HOSTADDRESS$
}
"""

legacy_checks = [
# command_name!argument, service description, pnp enable, hostname
( ( "check_ssh", "SSH Status", True), [ "linuxtestclient" ] ),
]
　　

　　2. /usr/local/check_mk/bin/cmk -R

　　

　　Setup MRPE on client with agent
　　

　　1. on client
　　vi /etc/check_mk/mrpe.cfg
　　#service description, command line for calling the plugin
　　LOAD    /usr/local/nagios/libexec/check_load -w 5.0,4.0,3.0 -c 10.0,6.0,4.0
　　

　　2. on Nagios check_mk server
　　vi /etc/hosts
　　192.168.1.11    linuxtestclient
　　

　　vi /usr/local/check_mk/etc/main.mk
　　all_hosts = [ 'localhost','linuxtestclient' ]
　　

　　3. /usr/local/check_mk/bin/cmk -I --checks=mrpe linuxtestclient
　　/usr/local/check_mk/bin/cmk -R

　　

　　使用check_mk MRPE监控percona xtradb cluster
on pxc node:
rpm -ivh percona-nagios-plugins-1.1.4-1.noarch.rpm
yum -y install xinetd
rpm -ivh check_mk-agent-1.2.4p5-1.noarch.rpm check_mk-agent-logwatch-1.2.4p5-1.noarch.rpm

mkdir /etc/nagios
vi /etc/nagios/mysql.cnf
[client]
user = root
password = s3cret

vi /etc/check_mk/mrpe.cfg
#service description, command line for calling the plugin
Threads_running    /usr/lib64/nagios/plugins/pmp-check-mysql-status -x Threads_running -w 20 -c 40