help setting monitors

[rnovoa]

Like it

Up

0

Down

Drop it

::

Hi,

Monitors are modules of type *PROC, meaning there are only two possible values, up (1) and down (0). They are very useful to get a quick view of the state of hosts, services… or any two-state variable you want to monitor.

You can, for example, write a module (monitor) that returns 0 (down) if cpu usage goes above 80%, and 1 (up) otherwise.

But you cannot make a monitor out of a module that gives you the cpu usage percentage at a given time.

Hope that made sense 🙂

I’m trying to figure out how to set monitors.

so I set one up for cpu usage, so I go to manage agents, click the agent for the PC to monitor (a win2k3 server in this case) then go to modules, then under module assignment I select “cpu usage/” then click update and enter an interval, put module type as generic_data (the default only shows 0 or 1) and hit ok.

But it does not show up in the monitor view. Is there something I’m doing wrong? also is there any “add on packs” available? I see theres no way to check ram, network usage, etc.

[red-squirrel]

Like it

Up

0

Down

Drop it

::

But if you can only get 0 or 1, then why is there the “view data” with the graphs? I seen screenshots with cpu graphs and such, just wondering how to go about setting those up for cpu, ram, etc…

[manu]

Like it

Up

0

Down

Drop it

::

It has the option to “view data” in the graphs because, perhaps, you’d like to see in the graph when something that should be up went down during over time.

In which screenshots did you see monitors for cpu or ram usage?

[rnovoa]

Like it

Up

0

Down

Drop it

::

I was wrong about (at least :D) one thing. Monitors do have two possible states, up and down, but they are not restricted to 1 and 0. Any value greater than 0 means up, not just 1.

I am not sure what the question is. Like manu pointed out, even if you have a monitor that returns 1 or 0, being able to view the actual data along with a time-stamp can be quite useful.

You want to display a CPU graph, but are you having trouble building the graph itself or assigning the module? What is the exact problem?

But if you can only get 0 or 1, then why is there the “view data” with the graphs? I seen screenshots with cpu graphs and such, just wondering how to go about setting those up for cpu, ram, etc…

[red-squirrel]

Like it

Up

0

Down

Drop it

::

Basically I want something like this:

http://pandora.sourceforge.net/screenshots/13_20M/pandora_14.jpg

Also would like to do it for Ram, but can’t seem to find a generic Ram one. there is for cisco and such but not for just regular windows/linux box. Is there a module addon pack I can get to add more monitoring options, by any chance?

[manu]

Like it

Up

0

Down

Drop it

::

In that picture, only the “Host Alive” is a monitor, the rest of them ain’t.

Aside from that, the RAM modules are in the pandora_agent.conf of each agent (either linux and windows), once you run the agents the data will be send in the XML

Manuel.