RAID monitoring

[Karl Katzke]

We're using Nagios here. I've worked with Nagios since it was called NetSaint. 

What I've found is that if you have lots of similar nodes (ex: We're all Dell here. In my last job, where I also used Nagios, our kit was all Tyan and HP), it's REALLY easy to write scripts that monitor anything you'd ever want to monitor... but if you've got 100 nodes with all different services running on them, it's kind of a pain. What I've taken to doing is using Nagios' template and inheritance features extensively inside my host and service files. Then I simply extend/extrude a host, and I can have ten new nodes up in less than five minutes. 

It's just the initial configuration that's tough. 

We use Cacti along with Nagios for all of our pretty-graphs-for-managers needs.

-Karl Katzke

You're right in that Nagios' graphs and displays are kind of hard to work with. 

>>> "J. Epperson" <Dell at epperson.homelinux.net> 11/03/07 1:10 PM >>>
On Sat, November 3, 2007 12:20, Marcus Bointon wrote:
> On 2 Nov 2007, at 21:22, Mike Hanby wrote:
>
>> I use Nagios.
>
>
> I don't know about anyone else, but I find nagios incredibly difficult
> and unpleasant to use. Trivial tasks like adding a node seem massively
> overcomplicated, and its policy of "do nothing out of the box" is
> really unhelpful. I once spent 4 days trying to get it to monitor 4
> nodes and send email on errors and eventually had to give up and write
> my own in about 10 mins that monitored the same stuff and sent email
> and SMS. Even when I've used working nagios setups, the displays are
> not terribly helpful, well designed or informative. Though it doesn't
> solve quite the same problem, I've found munin really good at 'just
> working' and it's really trivial to write new plugins for. I keep
> hearing about people using nagios on big installations - I figure they
> must be complete geniuses or just have lots of spare time...
>
Agreed on Nagios.  I figure the big installations have a full-time Nagios
wizard who does r&d and rollout all day.

Hobbit, the open source successor to the original Big Brother, on
steroids, is my current choice.  Easy to set up, easy to write custom ext
scripts for because it's so well documented, and uses RRD so the data is
extremely extensible.  You can get graphs of how often the CEO logs in if
you wanna.  And the mailing list is very active, great peer support.  Some
folks have farms in the thousands being monitored.

regards,
j.

_______________________________________________
Linux-PowerEdge mailing list
Linux-PowerEdge at dell.com
http://lists.us.dell.com/mailman/listinfo/linux-poweredge
Please read the FAQ at http://lists.us.dell.com/faq