Re: [Nagios-users] Nagios Scalability

[Marc P. <ma...@en...>]

> -----Original Message-----
> From: nag...@li... [mailto:nagios-users-
> bo...@li...] On Behalf Of Mr D
> Sent: Wednesday, August 29, 2007 1:19 PM
> To: nag...@li...
> Subject: [Nagios-users] Nagios Scalability
>=20
> Greets


> First question: is Nagios 3 and up, adequate for a task of this size.
> >From my own experience with Nagios the answer would be yes, but
> potentially 100,000 or more units is an awful lot of units.

I'm not a high-count site like that (~4,000 services) but just my $0.02.
--

Nagios-3 just entered beta status in the last 30 days. Chances are high
that not many sites, let alone sites with high host and service counts,
have put it through the grinder yet. It's also going to depend greatly
on what you're monitoring and how often. 1 ping per device per year is a
heck of a lot easier than running a 30 second plugin per device every
minute, passive checks are better than active checks, etc.

That being said, there have been many performance improvements for large
sites in v3 over v2. Many of them are documented at
http://nagios.sourceforge.net/docs/3_0/largeinstalltweaks.html. The
three biggest in my mind are the parallelization of host checks, the
ability to batch submit passive results from a file and the fast startup
options.=20

Ethan maintains a user profile section on the site at
http://www.nagios.org/userprofiles. The top site, presumably running v2
and added recently purports to be monitoring 1,000,000 services with 20
nagios machines (~50,000/machine). #2 is at 90,000 services with 10
servers. You can browse the rest to see what there is...


--
Marc