RAM memory monitoring information is not real

[parboleda]

Hello everyone I use Nagios XI to monitor a centos server, the monitoring was configured with "ncpa", in RAM does not show the real value of the used memory, shows that the total of ram is 16 Gb that is correct, the The value of the available memory is also correct, but in the used memory it shows only 1.68 Gb, when in fact it is using almost 16 Gb, and it is not generating alerts to me because it is consuming all the memory, how can I make it show the information Real and send alerts?

Attached Image

[scottwilkerson]

You have 12006 cached which is subtracted out because that will be free'd on demand

Free = Total - Used - Cached

[parboleda]

Good morning, thanks for the answer, it is not very clear to me, if rest those values ​​gives a negative value

Free = 16000 - 15600 - 12006
free = -11606

Is there a way for Nagios to ignore the cache value?

[scottwilkerson]

NCPA uses psutils to calculate the memory, here are the docs
https://psutil.readthedocs.io/en/latest/#memory

Is there a way for Nagios to ignore the cache value?

Not with NCPA, but There are other memory plugins available on the Nagios exchange
http://exchange.nagios.org/

[parboleda]

Good afternoon, thanks for the information, there is some other recommended method different to NCPA to monitor the memory that throws the values ​​that I need and that is not very complex to configure

[ssax]

I think you could probably use this:

Code: Select all

https://support.nagios.com/kb/article.php?id=171

[parboleda]

Thanks for the information

[mbellerue]

Do you have any other questions, or can I close this thread?