First post: looking to find out whether nagios can be adapted to work on a large scale

[Jonathan]

Howdy!

I'm a new list member, as we're just getting started with Nagios.  It's a
pleasure to make everyone's aquantance.

We're currently setting up Nagios to do distributed monitoring on our server
farm.  However, we recently were handed some requirements to build a
distributed monitoring solution for end-users.  I've been pounding the pixel
pavement looking for some kind of monitoring package that's robust enough to
provide end user access to tens-of-thousands of end user customers, and I'm
wondering if Nagios is flexible enough that it can be adapted.

We have a basic website (php auth, design, xhtml is done) framework, and are
looking to put together a core offering.  It's free to end-users, and what
it does is simple website monitoring.  We'd like to be able to write a php
or perl class to call something in Nagios to handle 

We're looking to set up a *free* version of something like this:
http://www.watchmouse.com/en/index.php
http://www.nmsalert.com/
http://netwhistle.com/overview/default.aspx

Picture of our project's homepage (draft):
http://www.firebright.com/projects/freeuptime1.jpg

A rough idea of what we want to do is:
-- build the framework for an enterprise class monitoring system. We already
have distributed DNS servers throughout the United States and Canada, so the
network to run this stuff is already there.

Obviously, this would require a database backed (mysql or postgres, either
is fine with us) website with multi-level authentication capabilities. We'd
need to be able to define account levels and capabilities based on those
account levels, specifically the number and type of monitoring services that
can be invoked. We figure something keyed off a (example here in mysql
format) database like this:

username VARCHAR(100) NOT NULL, 
password VARCHAR(100) NOT NULL, 
auth_level INT(11) NOT NULL, 

That auth level then has another table or tables which defines service
parameter(s) -- simple.

Then we need a monitoring engine to be written. Obviously, if we wanted to
hack this system together in a day, we could use a simple monitoring script
and adapt it to our use (something like psi would be easy to use).  This is
maybe where Nagios could come in, but we're not sure.  Is this something
that can be worked together?

The engine should be able to set up monitors for:

connect
http
httpd
pop3
smtp
ftp
imap
telnet
dnsa
dnsns
match
ping

I mean, eventually we'd like to be able to offer something like:
http://www.watchmouse.com/en/checkit.php

And you should be able to set the monitoring period to as low as 70080
checks/month (5 minutes) with a default for accounts of 1 hour.
Anyways, I hope that's enough detail for everyone.  That's what we're trying
to put together.  Yes, I know it's shameless self-promotion, but we figure
it's better that than 10,000 banner ads all over the internet -- at least we
can give something back.
I'd love some tips on whether this is something I should consider hacking
with Nagios, or whether we're going to have the build the functionality from
scratch.
Thanks a lot.  I sincerely appreciate it!
Best,
Jonathan
----
Email: j at FireBright.com
For Support: http://helpdesk.FireBright.com/
Toll Free: (866) FIREBRIGHT x901
Fax:  (916) 359 0151

FireBright, Inc.
4660 Natomas Blvd 
Suite 120-112
Sacramento, CA 95835

Please visit our site:
http://www.FireBright.com/






-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
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