monit-general
[Top][All Lists]
Advanced

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

Re: Using monit to identify an issue/process causing the issue?


From: EzCom Keith
Subject: Re: Using monit to identify an issue/process causing the issue?
Date: Wed, 5 Jan 2011 10:26:20 -0800

On Wed, Jan 5, 2011 at 9:57 AM, Greg Hilton <address@hidden> wrote:
>
> I have monit up and running fine, I have just installed Monit using the 
> instructions and monitrc here (but using latest version of monit)
>
>
>
> http://kb.mediatemple.net/questions/1638/How+do+I+install+Monit+on+a+%28dv%29+Dedicated-Virtual+Server%3F
>
>
>
> I have created and edited monitrc,  but wanted to know how I can configure it 
> to tell me what process is causing my issues below?
>
>
>
> My issue is that I see frequent kmemsize errors in my Virtuozzo QoS logs, as 
> below. Black zone means that a process was killed off to prevent the issue, 
> so I am trying to track down what process is causing the issue.
>
> I see these kind of messages from monit so far:
>
>
>
> Resource limit matched Service smartphonesoft.com
>
>
>
>                 Date:        Mon, 03 Jan 2011 10:05:11 -0800
>
>                 Action:      alert
>
>                 Host:        tabletpcsoft.net
>
>                 Description: loadavg(1min) of 3.4 matches resource limit 
> [loadavg(1min)>3.0]
>
>
>
>
>
> --
> To unsubscribe:
> http://lists.nongnu.org/mailman/listinfo/monit-general

Hi Greg,

Are there not other system logs that you can cross-reference through
grep ( using the QoS date/timestamps) to determine what process
is doing this? Using Monit as a shotgun approach seems like the wrong
step to take at this point. If you have other system logs that
might help pin down a process, THEN you can monitor that process with
Monit - though realistically all Monit can do for the process
is restart it, alert you, etc. But any way, you say you need to find
what process is causing the memory issue. The system logs can
tell you what was killed off (usually), and/or restarted around the
date/timestamps, but those processes might just be victims of whatever
is really causing the problem.

Hopefully you have posted your problem to other Fora/Groups associated
with your hardware and software, as they might prove infinitely
more helpful. Make sure to give such information as length of time
server has been in service? Has it always done this, or just recently?
Any hardware or software changes coinciding around this problem, or
did it appear out of nowhere?

If you have the luxury of monitoring yourself, you might want to keep
an eye on TOP and see if you have any services or processes that
are slowly spinning out on you... sorry cannot be more helpful! : )

- Keith



reply via email to

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