[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Nagios monitor script for gdth RAID controller

[Petter Reinholdtsen]
> Can anyone have a look at this draft script, and let me know what
> the exit codes and output must be to have nagios accept it as a
> monitor script?

I managed to find some documentation, and also concluded that the
subshell in the original script made it hard to return the correct
exit code, so I rewrote it in perl.  Please have a look at this
version, and let me know if I am on the right track.

# Author: Petter Reinholdtsen
# Date:   2006-05-15
# Nagios module to report the RAID status from the gdth linux kernel
# module.
# Based on documentation found in
# <URL:http://nagiosplug.sourceforge.net/developer-guidelines.html>

use strict;
use warnings;

use Getopt::Long;

my $verbose;
my $help;
my $RC = 0;

sub usage {
    print "Usage: $0 [-h|--help]\n";
    print "  Reports the gdth status as found in /proc/scsi/gdth/*\n"

my $result = GetOptions ("help" => \$help, "verbose+" => \$verbose);

if ($help) {
    exit 0;

if ( ! -d "/proc/scsi/gdth" ) {
    print "No gdth status file found, /proc/scsi/gdth/ missing";
    exit 3;

chdir "/proc/scsi/gdth";

my $RESULT = "";

for my $controller (<*>) {
    open(INFO, "< $controller") || die "Unable to read from $controller";
    # Have to use 'cat', as redirect into egrep didn't work with the
    # file in /proc/.
    while (<INFO>) {
        if (m/^\s+Number:\s+(\d+)\s+Status:\s+(\S+)$/) {
            my ($num, $status) = ($1,$2);
            if ("ok" ne $status) {
                $RC = 2;
                $RESULT="$RESULT ID $controller volume $num($status)";
#               print "Gdth RAID controller $controller volume $num ($status) NOT OK\n"
            } else {
#               print "Gdth RAID controller $controller volume $num OK\n"
if ( 0 == "$RC" ) {
    print "gdth RAID OK\n";
} else {
    print "gdth RAID $RESULT NOT OK\n";
exit $RC;

Reply to: