[Top][All Lists]

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

Re: CPU syntax

From: Martin Pala
Subject: Re: CPU syntax
Date: Mon, 4 Aug 2014 14:33:47 +0200


there are two kinds of CPU checks:

1.) system wide (with user/system/wait granularity) ... in the context of "check system ..."
2.) per-process (the Spamassassin example) ... in the context of "check process ..."

On system-wide level there is no "total cpu" load ... in practice it's usually sufficient to use some reasonable value for usr/sys/wait and/or combine it with loadavg test, which is good indicator for real system load.


On 31 Jul 2014, at 18:55, Francisco Reyes <address@hidden> wrote:

I had this in my monitrc
    if cpu usage (user) > 85% then alert
    if cpu usage (system) > 65% then alert
    if cpu usage (wait) > 30% then alert

CPU got to 100% and when I checked MMonit it was 65% user and 35% system.
I went to the monit site to look for a syntax for cpu without specific type and found
Spamassassin daemon (spam scan daemon)
if cpu usage > 99% for 5 cycles then alert

However, when I try
if cpu usage > 85 for 5 cycles then alert
if cpu usage > 95 then alert

Both of those lines give errors.

monit --version
This is Monit version 5.8

There is no way to check total CPU instead of user or system separately?
To unsubscribe:

reply via email to

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