Question regarding memory usage on Solaris servers

[Berto]

On the Solaris servers being monitored, memory usage shows stats for 'free', 'used', 'shared', and 'cache'. Is there a way to set alerts around the 'cache' value that is being collected? In other words, if I have a server showing 46GB of memory is being cached and I would like to get an alert when half that value is reached, is that possible?

--Berto

[mcapra]

Just so we're on the same page, what agent are you using to retrieve the system metrics for this machine? NRPE, SNMP, NCPA, etc.

Can you show me the command definition being leveraged to check the machine's available memory?

[Berto]

The agent being used is NRPE and this server is using the xiwizard_nrpe_service template and looking at the common settings tab I see;

check_command: check_nrpe
$USER1$/check_nrpe -H $HOSTADDRESS$ -t 30 -c $ARG1$ $ARG2$

I attached a screenshot of the tab inside Nagios XI.

[ssax]

The plugin that it calls (custom_check_mem) doesn't support it and I was unable to find one specifically for Solaris, it should be fairly easy to add though, please run these commands on the Solaris server and send me the entire output:

Code: Select all

/usr/local/nagios/libexec/custom_check_mem -w 30 -c 10
free -m
free -m | head -1 | gawk '{print $6}'
free -m | head -2 | tail -1
free -m -w | head -2 | tail -1

Thank you