pkg-monitoring-plugins/contrib/check_snmp_disk_monitor.pl

186 lines
6.2 KiB
Perl

#!/usr/local/bin/perl
# author: Al Tobey <albert.tobey@priority-health.com>
# what: monitor diskspace using the host-resources mib
# license: GPL - http://www.fsf.org/licenses/gpl.txt
#
# Todo:
use strict;
require 5.6.0;
use lib qw( /opt/nagios/libexec );
use utils qw(%ERRORS $TIMEOUT &print_revision &support &usage);
use SNMP 5.0;
use Getopt::Long;
use vars qw( $exit $message $opt_version $opt_timeout $opt_help $opt_command $opt_host $opt_community $opt_verbose $opt_warning $opt_critical $opt_port $opt_mountpoint $opt_stats $snmp_session $PROGNAME $TIMEOUT %mounts );
$PROGNAME = "snmp_disk_monitor.pl";
$opt_verbose = undef;
$opt_host = undef;
$opt_community = 'public';
$opt_command = undef;
$opt_warning = 99;
$opt_critical = 100;
$opt_port = 161;
$opt_stats = undef;
$message = undef;
$exit = 'OK';
%mounts = ();
sub process_options {
my( $opt_crit, $opt_warn ) = ();
Getopt::Long::Configure( 'bundling' );
GetOptions(
'V' => \$opt_version, 'version' => \$opt_version,
'v' => \$opt_verbose, 'verbose' => \$opt_verbose,
'h' => \$opt_help, 'help' => \$opt_help,
's' => \$opt_stats, 'statistics' => \$opt_stats,
'H:s' => \$opt_host, 'hostname:s' => \$opt_host,
'p:i' => \$opt_port, 'port:i' => \$opt_port,
'C:s' => \$opt_community, 'community:s' => \$opt_community,
'c:i' => \$opt_crit, 'critical:i' => \$opt_crit,
'w:i' => \$opt_warn, 'warning:i' => \$opt_warn,
't:i' => \$TIMEOUT, 'timeout:i' => \$TIMEOUT,
'm:s' => \$opt_mountpoint, 'mountpoint:s'=> \$opt_mountpoint
);
if ( defined($opt_version) ) { local_print_revision(); }
if ( defined($opt_verbose) ) { $SNMP::debugging = 1; }
if ( !defined($opt_host) || defined($opt_help) || !defined($opt_mountpoint) ) {
print_help();
exit $ERRORS{UNKNOWN};
}
$opt_mountpoint = [ split(/,/, $opt_mountpoint) ];
}
sub local_print_revision {
print_revision( $PROGNAME, '$Revision: 82 $ ' )
}
sub print_usage {
print "Usage: $PROGNAME -H <host> -C <snmp_community> [-s] [-w <low>,<high>] [-c <low>,<high>] [-t <timeout>] -m <mountpoint>\n";
}
sub print_help {
local_print_revision();
print "Copyright (c) 2002 Al Tobey <albert.tobey\@priority-health.com>\n\n",
"SNMP Disk Monitor plugin for Nagios\n\n";
print_usage();
print <<EOT;
-v, --verbose
print extra debugging information
-h, --help
print this help message
-H, --hostname=HOST
name or IP address of host to check
-C, --community=COMMUNITY NAME
community name for the host's SNMP agent
-m, --mountpoint=MOUNTPOINT
a mountpoint, or a comma delimited list of mountpoints
-w, --warning=INTEGER
percent of disk used to generate WARNING state (Default: 99)
-c, --critical=INTEGER
percent of disk used to generate CRITICAL state (Default: 100)
-s, --statistics
output statistics in Nagios format
EOT
}
sub verbose (@) {
return if ( !defined($opt_verbose) );
print @_;
}
sub check_for_errors {
if ( $snmp_session->{ErrorNum} ) {
print "UNKNOWN - error retrieving SNMP data: $snmp_session->{ErrorStr}\n";
exit $ERRORS{UNKNOWN};
}
}
# =========================================================================== #
# =====> MAIN
# =========================================================================== #
process_options();
alarm( $TIMEOUT ); # make sure we don't hang Nagios
$snmp_session = new SNMP::Session(
DestHost => $opt_host,
Community => $opt_community,
RemotePort => $opt_port,
Version => '2c'
);
# retrieve the data from the remote host
my( $mps, $alloc, $size, $used ) = $snmp_session->bulkwalk( 0, 4, [['hrStorageDescr'],['hrStorageAllocationUnits'],['hrStorageSize'],['hrStorageUsed']] );
check_for_errors();
alarm( 0 ); # all done with the network connection
# move all the data into a nice, convenient hash for processing
foreach my $mp ( @$mps ) { $mounts{$mp->iid}->{mountpoint} = $mp->val; }
foreach my $a ( @$alloc ) { $mounts{$a->iid}->{alloc_units} = $a->val; }
foreach my $si ( @$size ) {
if ( exists($mounts{$si->iid}->{alloc_units}) ) {
$mounts{$si->iid}->{size} = $si->val * $mounts{$si->iid}->{alloc_units};
}
else {
$mounts{$si->iid}->{size} = $si->val;
}
}
foreach my $us ( @$used ) {
if ( exists($mounts{$us->iid}->{alloc_units}) ) {
$mounts{$us->iid}->{used} = $us->val * $mounts{$us->iid}->{alloc_units};
}
else {
$mounts{$us->iid}->{used} = $us->val;
}
}
# now find the mountpoint or mountpoints that were actually requested and push onto an array for output
my @matches = ();
foreach my $mp ( @$opt_mountpoint ) {
my $found = scalar(@matches); # count all matches
foreach my $key ( keys(%mounts) ) {
if ( $mounts{$key}->{mountpoint} eq $mp ) {
# find the percentate - eval to avoid divide by zero errors
eval { $mounts{$key}->{percent_used} = $mounts{$key}->{used} / $mounts{$key}->{size} };
$mounts{$key}->{percent_used} =~ s/^0\.([0-9]{1,2})([0-9]?).*/\1/; # truncate
if ( $2 >= 5 ) { $mounts{$key}->{percent_used}++ }; # round the number number
verbose "mountpoint $mp has ", $mounts{$key}->{percent_used}, "% used, ",
$mounts{$key}->{size}, " bytes and ",$mounts{$key}->{used}, " used\n";
push( @matches, $mounts{$key} );
}
}
if ( scalar(@matches) == $found ) {
print "UNKNOWN - could not locate mountpoint $mp on host\n";
exit $ERRORS{UNKNOWN};
}
}
# now run through and check the thresholds
foreach my $mp ( @matches ) {
if ( $mp->{percent_used} >= $opt_warning ) {
$exit = 'WARNING';
if ( $mp->{percent_used} >= $opt_critical ) { $exit = 'CRITICAL'; }
}
$message .= $mp->{percent_used}.'% used on '.$mp->{mountpoint}.', ';
}
$message =~ s/,\s*$//;
# append statistics if requested
if ( defined($opt_stats) ) {
my @tmp = ();
foreach my $mp ( @matches ) {
push( @tmp, join(',',$mp->{mountpoint},$mp->{size},$mp->{used}) );
}
$message .= '|'.join( ':', @tmp );
}
print "Disk $exit - $message\n";
exit $ERRORS{$exit};