[lug] Nagios help?
Matt James
matuse at gmail.com
Mon May 30 09:13:46 MDT 2011
On Tue, May 24, 2011 at 9:29 PM, karl horlen <horlenkarl at yahoo.com> wrote:
> also good points. thanks
I apologize for being bit late getting back to this thread, none the less I
am still working my way though this. I am intrigued by the conversation on
passive monitoring, thanks for all of the feedback. I'm glad to hear it is
and does pretty much exactly as I expected. We have a number of remote
sites with limited bandwidth and some have high security requirements so the
firewall / VPN thing comes into play quite often for us. It sounds to me
like a passive monitor of these sites would be the way to go. In that case,
I've been debating between passive monitoring and distributed monitoring. I
also see that nagios can monitor other nagios instances. So each of these
sites has anywhere from 1 - 12 computers with another possibly 1 - 30
Ethernet devices to keep tabs on (UPS, Wi-Max radios, etc.) In these cases,
or maybe just for the larger deployments, would it not be better to run a
local nagios monitoring server and have it report back to a central nagios
server? Can I do passive checks nagios to nagios?
Although I am struggling to understand why when I do a ./configure in the
add-ons I get a boat load of errors and only about 1/2 of the add-ons
compile, I think I'm struggling more with the logic of it all. Where is the
best place to physically put the monitoring server in any given topology?
How worried should I be about redundant paths for my monitoring server to
reach the outside world and deliver notifications? I get the feeling that
these sorts of things can only be learned by experience. I'm also learning
(due to an event last week) that some problems can masquerade themselvs as
something entirely different if you're not paying attention. I had a
wireless bridge dropping about 50% of the traffic across the link, which
showed up as a few down hosts past the link. It took me a few minutes of
head scratching to figure out why a perfectly accessible host would show as
down. So what that means to me is that I don't have my setup very well
tuned - and again, I think this is where some experience might help.
Any further thoughts / comments would be greatly appreciated. Thanks again
to everyone who already posted.
Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lug.boulder.co.us/pipermail/lug/attachments/20110530/f8d3d2a6/attachment.html>
More information about the LUG
mailing list