[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Odd monit configuration help...

From: Martin Pala
Subject: Re: Odd monit configuration help...
Date: Fri, 05 Nov 2004 21:32:35 +0100
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20041007 Debian/1.7.3-5

I think you can use "manual" monitoring mode and modify the rc scripts to enable monitoring such as (based on your samples):

1.) monitrc:
set logfile = syslog
set alert address@hidden
set httpd port 2812 and use address localhost
        allow localhost
        allow monit:monit
include /etc/monit/monit.d/*
>>>NOTE: it is the same as yours<<<

2.) and then monit.d contains files like this - /etc/monit/monit.d/ssh:
check ssh.sshd with pidfile /var/run/
        start program "/etc/init.d/ssh start"
        stop program "/etc/init.d/ssh stop"
        if failed port 22 protocol ssh then restart
        if 5 restarts within 5 cycles then timeout
        mode manual

check file ssh.etc.ssh.sshd_config with path /etc/ssh/sshd_config
        if timestamp changed then exec "/etc/init.d/ssh reload"
        mode manual

check file ssh.usr.sbin.sshd with path /usr/sbin/sshd
        if timestamp changed then exec "/etc/init.d/ssh restart"
        mode manual
>>>NOTE: mode changed to manual => monit is passive on sshd<<<

3.) add following line to /etc/init.d/sshd (symbolic example):
case "$1" in
        /usr/bin/monit monitor sshd
        /usr/bin/monit unmonitor sshd
        /usr/bin/pkill sshd
>>>NOTE: monitoring enabled => monit will be active on sshd<<<


- monit is started in daemon mode (as usual)
- shared configuration defines the services which are not common for all
  servers in "manual" monitoring mode
- rc scripts are modified to enable/disable monitoring when the service
  is started/stopped

Mark Ferlatte wrote:
Hi all,

I'm trying to configure monit to do something that perhaps it's not capable of
doing, but figured I'd ask here before giving up.

I run a reasonably sized cluster of machines (~500), and use monit as a service
monitor and restarter.  This works well.  The problem that I have is that not
all machines in the cluster run the same set of services, and the question of
"which machine runs which service" changes with some frequency.

What I think I want is a way to configure monit so that it knows how to monitor
every service we run, but only actually have it care about the services that
are running when monit starts up (it starts last in our boot process).  This
would let me only have one global monit configuration for the entire cluster,
and let the computer figure out the local set of things to worry about.

Is there are way to do this already that I'm missing?



To unsubscribe:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]