Check Command: check_nrpe

Configuration

Firewall Ports

If you have a firewall between your NEMS Server and your host, you must open incoming TCP connections on ports 5666 and 12489 on your host side.

Since check_nrpe can pose a security risk, please do not open these ports to the world. Rather, ensure your NEMS Server is the only outside IP allowed to connect to these ports.

Host-Side Service

The NRPE service must be installed before check_nrpe can be used to monitor the host.

See:

• NRPE For Linux

• NRPE For Windows Clients

Important Note for Users of NEMS 1.5 and Under

In NEMS Linux 1.5 and lower, an older version of NRPE was used. This can be upgraded by running sudo nems-upgrade however you must also make a few minor changes in NEMS NConf as follows:

• Edit check_nrpe in checkcommands, changing the command line to: $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$ (essentially, just removing the quotation marks).

• Edit / Disk Space and /var Disk Space in Advanced Services, changing the ARGS to: check_disk -a ‘-w 80 -c 90 -p / -u GB’ and check_disk -a ‘-w 80 -c 90 -p /var -u GB’ respectively.

Sample Args for check_nrpe

In order to use these commands, NRPE must be installed on the client system using the NEMS Linux installation procedure found here: NRPE For Linux

I suggest you always put NRPE in the service titles you create in order to prevent accidentally assigning a local service to a host who uses NRPE. So instead of calling the advanced service “Check Disk Space /” I’d call it “Check Disk Space / via NRPE”.

Check CPU Temperature of Remote System

• Description: Detect temperature of remote system CPU.

• Client Requirements: Must have lm-sensors installed and working.

• $ARG1$ Syntax: check_cpu_temp -a “WARN CRIT”

$ARG1$ Examples:

Warn if CPU is hotter than 40°C and Critical if over 50°C:

check_cpu_temp -a “40 50”

Warn if CPU is hotter than 35°C and Critical if over 47°C:

check_cpu_temp -a “35 47”

Check Disk Usage of Remote System

• Description: Determine disk usage by percentage.

• Client Requirements: None.

• $ARG1$ Syntax: check_disk -a “-w WARN -c CRIT -p PATH”

$ARG1$ Examples:

Warn if the disk at / is 80% full and Critical if over 90% full and report in gigabytes:

check_disk -a ‘-w 80 -c 90 -p / -u GB’

Warn if the disk mounted on /var is 80% full and Critical if over 90% full and report in gigabytes:

check_disk -a ‘-w 80 -c 90 -p /var -u GB’

Warn if the disk mounted on /mnt/backup is 50% full and Critical if over 70% full and report in gigabytes:

check_disk -a ‘-w 50 -c 70 -p /mnt/backup -u GB’