Getting decent availability stats for servers at remote sites

[Anthony Wright]

I have a central site and servers at a number of remote sites all
connected together over the internet. The users of these servers are
also at the remote site, so if their internet connection fails they can
carry on using the server without interruption. The connections are not
under my control, but the servers are.

I run Nagios at the central site, and while normally the connections to
the remote sites are good, if a connection fails, Nagios records the
servers at the site as unavailable. This means that when I come to look
at my weekly/monthly server availability stats what I am looking at is a
combination of server and network connection availability (and it's
mainly network faults that cause outages). This becomes a big problem
when I want to start offering SLAs on these remote servers because
Nagios isn't recording the availability of the server.

I still want to maintain a central monitoring server, but I wondered
about installing a monitoring server/proxy/agent at each remote site so
that it would record the availability of all the servers at the site,
and would periodically feed that back to the central server. Thus if the
internet connection failed, when it was restored the site server would
tell the central server that everything had been fine with the servers
while they were disconnected.

Is there a way to do this with Nagios and/or net-snmp?

thanks,

Anthony Wright.


------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
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