nagios ncpa listener logs

[iinfi]

I installed ncpa agent on a windows server 2012 R2 box and see the following messages in the ncpa_listener.log.
What could be the issue? The service starts but netstat -ab doesnt show the port number 5693.
IIS was running on that machine on port 80. Could that be an issue?

tried a different port number to start ncpa (15693) still no success. Same message as below.

INFO:root:Looking for config at: C:\Program Files (x86)\Nagios\NCPA\etc\ncpa.cfg
INFO:root:Looking for plugins at: C:\Program Files (x86)\Nagios\NCPA\plugins/
ERROR:root:couldn't find any physical disk
Traceback (most recent call last):
File "C:\Users\nscott\Documents\Git\ncpa\agent\ncpa_windows.py", line 86, in start
File "C:\Users\nscott\Documents\Git\ncpa\agent\listener\server.py", line 11, in <module>
File "C:\Users\nscott\Documents\Git\ncpa\agent\listener\psapi.py", line 176, in <module>
File "C:\Python27\lib\site-packages\psutil\__init__.py", line 1229, in disk_io_counters
RuntimeError: couldn't find any physical disk

[lmiltchev]

Change the port # in the ncpa.cfg to the default one:

Code: Select all

port = 5693

save and exit. Make sure your firewall is not blocking the port. Restart the service and try accessing the API in a browser:

Code: Select all

https://<ip address>:5693/

Click on the API menu and check if you can see any disks listed. You can try:

Code: Select all

https://<ip address>:5693/api/disk/

[iinfi]

Thanks for your prompt response
yes the port was default when i first tried it and firewall is off on my machine.
I see this error on a physical machine with WS2012R2. On a physical box with WS2012 it works. Makes me wonder whether 2012 r2 is supported or not by the NCPA agent

[tmcdonald]

I'm not 100% sure if this would matter (I'm more of a Linux guy), but did you run the installer as Admin? It could be that due to a group policy setting the service cannot detect the disks.

[iinfi]

I ran it as admin and the machine is not joined to the domain.
from a windows machine.. all i need is to collect Windows Event Logs, Check network utilization, CPU and Memory. Is there a way I can do it without Nagios agent on the client machine?
Can I do the same for a Linux machine also without an agent?

[slansing]

Absolutely, I'd recommend taking a look at the WMI Monitoring Wizard:

http://assets.nagios.com/downloads/nagi ... ios-XI.pdf

If you have any questions let us know!

[iinfi]

Thanks i'll take a look. Does a agentless monitoring exist for Linux also?

[iinfi]

i am bumping my post.
Does a agentless monitoring exist for Linux also?

Thanks

[tmcdonald]

That depends on what you mean by "agentless". If you mean exposing the internal metrics of a Linux machine with no prior configuration, then no this is not possible. *Something* needs to be running on the machine in order to query it, however it does not necessarily need to be a Nagios solution. NCPA does work on Linux, but as for a non-Nagios solution I am not aware of one that we would have a plugin for.

And as a side note for the future, we answer older posts before newer ones, so bumping your post, especially if you do it frequently, will possibly delay us getting to you.

[iinfi]

Thanks for your reply. Sorry for bumping the post.

Basically I want to use Nagios XI for monitoring linux servers but not install any agent on it. Is that possible?


Secondly
I am in the evaluation phase of Nagios XI. Coming back to windows. Does it allow me to export all the Windows event logs? Currently I can see the windows events only when the system is up. I want to export logs even when the system is down.
Is that supported?