hostgroup member limit

Marc Powell mpowell at ena.com
Tue Apr 15 18:09:22 CEST 2003


Sure, no problem. Note that I didn't say it wasn't without its quirks,
just that expanding the hostgroup limit didn't break it ;)

For those who are curios, we've implemented a distributed monitoring
system with 4 data collectors, the largest of which is polling 692
hosts. Our aggregate checks are approximately 1800 check_ping and 300
miscellaneous check_http variants, check_dns, check_smtp, etc... All
data collectors are reporting back to 2 central hosts via nsca. I'd like
to have only 1 central host but performance issues prevent it so I've
got one host that uses a flat-file back end and another that uses a
postgres back end. We use the database backend because we need to do
some fairly fluid reporting to our major customer going back up to 5
years so we suck that information over to SQL Server 2000 for archiving.
The problem with using the database backend in this scenario is that
service checks come in faster than nagios can process the pipe and
update the database. I have to restart nagios 2 times each day on this
machine (bleh) in order to clear up the 7000+ backlogged processes. I
don't have any such problems with the flat-file backend central host. My
only real limitation there is that the Status Summary views take up to
3-4 minutes to generate so I've just created a cron job that creates
them as static HTML every 5 minutes.

Other caveats --
	Host checks are not performed (check is defined as check_dummy
0) so parenting, etc. is irrelevant.

I guess that's it...

- Data Collectors are PIII 800MHz with 512Megs RAM, Redhat 7.3. Load is
minimal and each Data Collector is running Nagios, Cricket and
Smokeping.

- Central Host #1 (Postgres) is a quad PIII 550 with 2GB Ram, Redhat
7.3. This box is solely dedicated to Nagios/Cricket/Smokeping
aggregation (not used as user front-end though). Load typically sucks
and can be as high as 800 for the 5 minute average. The machine is still
usable though and the high load average is the result of the high number
of nsca/nagios processes hanging out waiting to write to the command
pipe.

- Central Host #2 (flat-file) is a quad PIII 500 with 1GB Ram, Redhat
7.3. Nagios aggregation point and a host of other services. User
front-end as well. Load is minimal and this is actually a heavily used
box for both interactive and non-interactive services.

--
Marc

-----Original Message-----
From: Ethan Galstad [mailto:nagios at nagios.org] 
Sent: Monday, April 14, 2003 8:08 PM
To: nagios-users at lists.sourceforge.net
Subject: RE: [Nagios-users] hostgroup member limit

That's the largest install I've heard of thus far.  I'm surprised 
Nagios isn't barfing all over itself with that number of services. :-
)  Care to add your installation to the user profiles listing?

http://www.nagios.org/userprofiles/


On 14 Apr 2003 at 18:40, Marc Powell wrote:

> 2000+ but I have a particular hostgroup that has 692 members and
another
> has 541 members. I've got about 200 hostgroups with some amount of
> overlap for more logical groupings.
> 
> marc
> 
> -----Original Message-----
> From: Stanley Hopcroft [mailto:Stanley.Hopcroft at IPAustralia.Gov.AU] 
> Sent: Monday, April 14, 2003 6:34 PM
> To: Marc Powell
> Cc: nagios-users at lists.sourceforge.net
> Subject: Re: [Nagios-users] hostgroup member limit
> 
> Dear Sir,
> 
> Please tell me to look in the monster Nagios installations if
necessary,
> 
> otherwise
> 
> On Mon, Apr 14, 2003 at 01:15:22PM -0500, Marc Powell wrote:
> > 
> > I've increased mine to 49152 with no apparent problems but YMMV.
> > 
> 
> how many hosts are you monitoring ?
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> 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
> 



Ethan Galstad,
Nagios Developer
---
Email: nagios at nagios.org
Website: http://www.nagios.org



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
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


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
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