Nagios Strategies

Todd Mcneill todd.mcneill at pmigroup.com
Wed Dec 27 19:42:55 CET 2006


I was thinking of something a little more detailed.  For example, say I
have 3 servers, a web server (A), an application server (B), and a
database server (C).  Each of these servers support 3 applications, each
of which has a component on each of the servers, registered in Nagios as
a service.  Each of these components for an application is
interdependent, so if the database component for application 3 is having
issues, it would be nice to see this visually represented in an
application view.

As another example, we have applications that are interconnected using a
messaging service, such that application A is dependent on application B
for certain types of transactions.  If application B is having
performance issues which may potentially impact application A, it would
be nice to see this visually represented.

A third example would be SAN.  All of my databases are SAN attached.  I
can write checks which can interrogate the SAN.  If, for example, I have
meta LUNs for applications A and C which span common physical spindles
(RAID groups), a rebuild operation on the RAID group could impact
performance on those dependent applications.  To see these dependencies
visually would be fantastic.

Something like this would assist in root cause analysis, instead of just
using the BOFH excuse-of-the-day :)

Todd

-----Original Message-----
From: Brian Loe [mailto:knobdy at gmail.com] 
Sent: Thursday, December 21, 2006 5:31 PM
To: Chris Moody
Cc: Todd Mcneill; Nagios Users mailinglist
Subject: Re: [Nagios-users] Nagios Strategies

I'm not sure on the ease, but it seems like it should be doable now
for any service/application for which you can run a Nagios check
against. The parent being the system (ping?), the next some base OS
subsystem, next the app that depends on it (log file checks or some
other homegrown verification), etc.. Right? Or am I over-simplifying
the issue?

I do this, as do a lot of us I'm sure, in a very simple form now.
Nagios checks that my cacti server is up by checking the host, apache
then the cacti log - and they appear that way on the map as well.

On 12/21/06, Chris Moody <cmoody at qualcomm.com> wrote:
> This would be AWESOME(!!!) if it were easily implementable.
>
> Cheers,
> -Chris
>
> Todd Mcneill wrote:
>
> > It might also be interesting to see if there is a way to visually
> > represent these service dependencies on the Status Map.  I have
people
> > that are interested in viewing the status of the entire multi-tier
> > application stack by application, but this is difficult the way it
is
> > represented now.  I can create Host Groups for each application, and
> > that should translate to a drawing layer on the Status Map, but from
> > what I understand, it won't necessarily show me the status of the
> > application if, for example, an instance of an httpd daemon that the
> > application depends on goes down.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null





More information about the Users mailing list