5 # Monitor Dell server hardware status using Dell OpenManage Server
6 # Administrator, either locally via NRPE, or remotely via SNMP.
10 # Copyright (C) 2009 Trond H. Amundsen
12 # This program is free software: you can redistribute it and/or modify
13 # it under the terms of the GNU General Public License as published by
14 # the Free Software Foundation, either version 3 of the License, or
15 # (at your option) any later version.
17 # This program is distributed in the hope that it will be useful, but
18 # WITHOUT ANY WARRANTY; without even the implied warranty of
19 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
20 # General Public License for more details.
22 # You should have received a copy of the GNU General Public License
23 # along with this program. If not, see <http://www.gnu.org/licenses/>.
26 require 5.006; # Perl v5.6.0 or newer is required
29 use POSIX qw(isatty ceil);
30 use Getopt::Long qw(:config no_ignore_case);
32 # Global (package) variables used throughout the code
33 use vars qw( $NAME $VERSION $AUTHOR $CONTACT $E_OK $E_WARNING $E_CRITICAL
34 $E_UNKNOWN $FW_LOCK $USAGE $HELP $LICENSE
35 $snmp_session $snmp_error $omreport $globalstatus $global
36 $linebreak $omopt_chassis $omopt_system $blade
37 $exit_code $snmp $original_sigwarn
38 %check %opt %perfdata %reverse_exitcode %status2nagios
39 %snmp_status %snmp_probestatus %probestatus2nagios %sysinfo
40 %blacklist %nagios_alert_count %count
41 @perl_warnings @controllers @enclosures
42 @report_storage @report_chassis @report_other
45 #---------------------------------------------------------------------
46 # Initialization and global variables
47 #---------------------------------------------------------------------
49 # Small subrouting to collect any perl warnings during execution
50 sub collect_perl_warning {
51 push @perl_warnings, [@_];
54 # Set the WARN signal to use our collect subroutine above
55 $original_sigwarn = $SIG{__WARN__};
56 $SIG{__WARN__} = \&collect_perl_warning;
58 # Version and similar info
59 $NAME = 'check_openmanage';
60 $VERSION = '3.5.4-beta4';
61 $AUTHOR = 'Trond H. Amundsen';
62 $CONTACT = 't.h.amundsen@usit.uio.no';
70 # Firmware update lock file [FIXME: location on Windows?]
71 $FW_LOCK = '/var/lock/.spsetup'; # default on Linux
74 $USAGE = <<"END_USAGE";
75 Usage: $NAME [OPTION]...
83 -p, --perfdata Output performance data
84 -t, --timeout Plugin timeout in seconds
85 -c, --critical Customise temperature critical limits
86 -w, --warning Customise temperature warning limits
87 -d, --debug Debug output, reports everything
88 -h, --help Display this help text
89 -V, --version Display version info
93 -H, --hostname Hostname or IP of the server (needed for SNMP)
94 -C, --community SNMP community string
95 -P, --protocol SNMP protocol version
96 --port SNMP port number
100 -i, --info Prefix any alerts with the service tag
101 -e, --extinfo Append system info to alerts
102 -s, --state Prefix alerts with alert state
103 --short-state Prefix alerts with alert state (abbreviated)
104 -o, --okinfo Verbosity when check result is OK
105 --htmlinfo HTML output with clickable links
107 CHECK CONTROL AND BLACKLISTING:
109 -a, --all Check everything, even log content
110 -b, --blacklist Blacklist missing and/or failed components
111 --only Only check a certain component or alert type
112 --check Fine-tune which components are checked
114 For more information and advanced options, see the manual page or URL:
115 http://folk.uio.no/trondham/software/check_openmanage.html
118 # Version and license text
119 $LICENSE = <<"END_LICENSE";
121 Copyright (C) 2009 $AUTHOR
122 License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
123 This is free software: you are free to change and redistribute it.
124 There is NO WARRANTY, to the extent permitted by law.
126 Written by $AUTHOR <$CONTACT>
129 # Options with default values
130 %opt = ( 'blacklist' => [],
134 'timeout' => 30, # default timeout is 30 seconds
144 'okinfo' => 0, # default "ok" output level
145 'linebreak' => undef,
150 'port' => 161, # default SNMP port
152 'community' => 'public', # SMNP v1 or v2c
154 'username' => undef, # SMNP v3
155 'authpassword' => undef, # SMNP v3
156 'authkey' => undef, # SMNP v3
157 'authprotocol' => undef, # SMNP v3
158 'privpassword' => undef, # SMNP v3
159 'privkey' => undef, # SMNP v3
160 'privprotocol' => undef, # SMNP v3
164 GetOptions('b|blacklist=s' => \@{ $opt{blacklist} },
165 'check=s' => \@{ $opt{check} },
166 'c|critical=s' => \@{ $opt{critical} },
167 'w|warning=s' => \@{ $opt{warning} },
168 't|timeout=i' => \$opt{timeout},
169 'd|debug' => \$opt{debug},
170 'h|help' => \$opt{help},
171 'V|version' => \$opt{version},
172 'p|perfdata:s' => \$opt{perfdata},
173 'i|info' => \$opt{info},
174 'e|extinfo' => \$opt{extinfo},
175 'htmlinfo:s' => \$opt{htmlinfo},
176 'postmsg=s' => \$opt{postmsg},
177 's|state' => \$opt{state},
178 'short-state' => \$opt{shortstate},
179 'o|ok-info=i' => \$opt{okinfo},
180 'l|linebreak=s' => \$opt{linebreak},
181 'a|all' => \$opt{all},
182 'only=s' => \$opt{only},
183 'omreport=s' => \$opt{omreport},
184 'port=i' => \$opt{port},
185 'H|hostname=s' => \$opt{hostname},
186 'C|community=s' => \$opt{community},
187 'P|protocol=i' => \$opt{protocol},
188 'U|username=s' => \$opt{username},
189 'authpassword=s' => \$opt{authpassword},
190 'authkey=s' => \$opt{authkey},
191 'authprotocol=s' => \$opt{authprotocol},
192 'privpassword=s' => \$opt{privpassword},
193 'privkey=s' => \$opt{privkey},
194 'privprotocol=s' => \$opt{privprotocol},
195 ) or do { print $USAGE; exit $E_UNKNOWN };
197 # If user requested help
203 # If user requested version info
211 print "PLUGIN TIMEOUT: $NAME timed out after $opt{timeout} seconds\n";
216 # If we're using SNMP
217 $snmp = defined $opt{hostname} ? 1 : 0;
219 # SNMP session variables
220 $snmp_session = undef;
223 # The omreport command
226 # Check flags, override available with the --check option
227 %check = ( 'storage' => 1, # check storage subsystem
228 'memory' => 1, # check memory (dimms)
229 'fans' => 1, # check fan status
230 'power' => 1, # check power supplies
231 'temp' => 1, # check temperature
232 'cpu' => 1, # check processors
233 'voltage' => 1, # check voltage
234 'batteries' => 1, # check battery probes
235 'amperage' => 1, # check power consumption
236 'intrusion' => 1, # check intrusion detection
237 'alertlog' => 0, # check the alert log
238 'esmlog' => 0, # check the ESM log (hardware log)
239 'esmhealth' => 1, # check the ESM log overall health
243 $linebreak = isatty(*STDOUT) ? "\n" : '<br/>';
245 # Line break from option
246 if (defined $opt{linebreak}) {
247 if ($opt{linebreak} eq 'REG') {
250 elsif ($opt{linebreak} eq 'HTML') {
251 $linebreak = '<br/>';
254 $linebreak = $opt{linebreak};
258 # Exit with status=UNKNOWN if there is firmware upgrade in progress
259 if (!$snmp && -f $FW_LOCK) {
260 print "MONITORING DISABLED - Firmware update in progress ($FW_LOCK exists)\n";
264 # List of controllers and enclosures
265 @controllers = (); # controllers
266 @enclosures = (); # enclosures
269 @report_storage = (); # messages with associated nagios level (storage)
270 @report_chassis = (); # messages with associated nagios level (chassis)
271 @report_other = (); # messages with associated nagios level (other)
273 # Counters for everything
276 'pdisk' => 0, # number of physical disks
277 'vdisk' => 0, # number of logical drives (virtual disks)
278 'temp' => 0, # number of temperature probes
279 'volt' => 0, # number of voltage probes
280 'amp' => 0, # number of amperage probes
281 'intr' => 0, # number of intrusion probes
282 'dimm' => 0, # number of memory modules
283 'fan' => 0, # number of fan probes
284 'cpu' => 0, # number of CPUs
285 'bat' => 0, # number of batteries
286 'power' => 0, # number of power supplies
288 'Critical' => 0, # critical entries in ESM log
289 'Non-Critical' => 0, # warning entries in ESM log
290 'Ok' => 0, # ok entries in ESM log
293 'Critical' => 0, # critical entries in alert log
294 'Non-Critical' => 0, # warning entries in alert log
295 'Ok' => 0, # ok entries in alert log
302 # Global health status
303 $global = 1; # default is to check global status
304 $globalstatus = $E_OK; # default global health status is "OK"
306 # Nagios error levels reversed
310 $E_WARNING => 'WARNING',
311 $E_CRITICAL => 'CRITICAL',
312 $E_UNKNOWN => 'UNKNOWN',
315 # OpenManage (omreport) and SNMP error levels
318 'Unknown' => $E_CRITICAL,
319 'Critical' => $E_CRITICAL,
320 'Non-Critical' => $E_WARNING,
322 'Non-Recoverable' => $E_CRITICAL,
323 'Other' => $E_CRITICAL,
334 6 => 'Non-Recoverable',
337 # Probe Status via SNMP
340 1 => 'Other', # probe status is not one of the following:
341 2 => 'Unknown', # probe status is unknown (not known or monitored)
342 3 => 'Ok', # probe is reporting a value within the thresholds
343 4 => 'nonCriticalUpper', # probe has crossed upper noncritical threshold
344 5 => 'criticalUpper', # probe has crossed upper critical threshold
345 6 => 'nonRecoverableUpper', # probe has crossed upper non-recoverable threshold
346 7 => 'nonCriticalLower', # probe has crossed lower noncritical threshold
347 8 => 'criticalLower', # probe has crossed lower critical threshold
348 9 => 'nonRecoverableLower', # probe has crossed lower non-recoverable threshold
349 10 => 'failed', # probe is not functional
352 # Probe status translated to Nagios alarm levels
355 'Other' => $E_CRITICAL,
356 'Unknown' => $E_CRITICAL,
358 'nonCriticalUpper' => $E_WARNING,
359 'criticalUpper' => $E_CRITICAL,
360 'nonRecoverableUpper' => $E_CRITICAL,
361 'nonCriticalLower' => $E_WARNING,
362 'criticalLower' => $E_CRITICAL,
363 'nonRecoverableLower' => $E_CRITICAL,
364 'failed' => $E_CRITICAL,
367 # System information gathered
370 'bios' => 'N/A', # BIOS version
371 'biosdate' => 'N/A', # BIOS release date
372 'serial' => 'N/A', # serial number (service tag)
373 'model' => 'N/A', # system model
374 'osname' => 'N/A', # OS name
375 'osver' => 'N/A', # OS version
376 'om' => 'N/A', # OMSA version
377 'bmc' => 0, # HAS baseboard management controller (BMC)
378 'rac' => 0, # HAS remote access controller (RAC)
379 'rac_name' => 'N/A', # remote access controller (RAC)
380 'bmc_fw' => 'N/A', # BMC firmware
381 'rac_fw' => 'N/A', # RAC firmware
384 # Adjust which checks to perform
385 adjust_checks() if defined $opt{check};
387 # Blacklisted components
388 %blacklist = defined $opt{blacklist} ? %{ get_blacklist() } : ();
390 # If blacklisting is in effect, don't check global health status
391 if (scalar keys %blacklist > 0) {
395 # Take into account new hardware and blades
396 $omopt_chassis = 'chassis'; # default "chassis" option to omreport
397 $omopt_system = 'system'; # default "system" option to omreport
398 $blade = 0; # if this is a blade system
400 # Some initializations and checking before we begin
402 snmp_initialize(); # initialize SNMP
403 snmp_check(); # check that SNMP works
404 snmp_detect_blade(); # detect blade via SNMP
407 # Find the omreport binary
409 # Check help output from omreport, see which options are available.
410 # Also detecting blade via omreport.
411 check_omreport_options();
415 #---------------------------------------------------------------------
417 #---------------------------------------------------------------------
420 # Store a message in one of the message arrays
423 my ($type, $msg, $exval, $id) = @_;
424 defined $id or $id = q{};
428 'storage' => \@report_storage,
429 'chassis' => \@report_chassis,
430 'other' => \@report_other,
433 return push @{ $type2array{$type} }, [ $msg, $exval, $id ];
438 # Run command, put resulting output lines in an array and return a
439 # pointer to that array
444 open my $CMD, '-|', $command
445 or do { report('other', "Couldn't run command '$command': $!", $E_UNKNOWN)
449 or do { report('other', "Couldn't close filehandle for command '$command': $!", $E_UNKNOWN)
450 and return \@lines };
455 # Run command, put resulting output in a string variable and return it
460 open my $CMD, '-|', $command
461 or do { report('other', "Couldn't run command '$command': $!", $E_UNKNOWN) and return };
462 my $rawtext = do { local $/ = undef; <$CMD> }; # slurping
465 # NOTE: We don't check the return value of close() since omreport
466 # does something weird sometimes.
474 sub snmp_initialize {
475 # Legal SNMP v3 protocols
476 my $snmp_v3_privprotocol = qr{\A des|aes|aes128|3des|3desde \z}xms;
477 my $snmp_v3_authprotocol = qr{\A md5|sha \z}xms;
479 # Parameters to Net::SNMP->session()
482 '-port' => $opt{port},
483 '-hostname' => $opt{hostname},
484 '-version' => $opt{protocol},
487 # Parameters for SNMP v3
488 if ($opt{protocol} == 3) {
490 # Username is mandatory
491 if (defined $opt{username}) {
492 $param{'-username'} = $opt{username};
495 print "SNMP ERROR: With SNMPv3 the username must be specified\n";
499 # Authpassword is optional
500 if (defined $opt{authpassword}) {
501 $param{'-authpassword'} = $opt{authpassword};
504 # Authkey is optional
505 if (defined $opt{authkey}) {
506 $param{'-authkey'} = $opt{authkey};
509 # Privpassword is optional
510 if (defined $opt{privpassword}) {
511 $param{'-privpassword'} = $opt{privpassword};
514 # Privkey is optional
515 if (defined $opt{privkey}) {
516 $param{'-privkey'} = $opt{privkey};
519 # Privprotocol is optional
520 if (defined $opt{privprotocol}) {
521 if ($opt{privprotocol} =~ m/$snmp_v3_privprotocol/xms) {
522 $param{'-privprotocol'} = $opt{privprotocol};
525 print "SNMP ERROR: Unknown privprotocol '$opt{privprotocol}', "
526 . "must be one of [des|aes|aes128|3des|3desde]\n";
531 # Authprotocol is optional
532 if (defined $opt{authprotocol}) {
533 if ($opt{authprotocol} =~ m/$snmp_v3_authprotocol/xms) {
534 $param{'-authprotocol'} = $opt{authprotocol};
537 print "SNMP ERROR: Unknown authprotocol '$opt{authprotocol}', "
538 . "must be one of [md5|sha]\n";
543 # Parameters for SNMP v2c or v1
544 elsif ($opt{protocol} == 2 or $opt{protocol} == 1) {
545 $param{'-community'} = $opt{community};
548 print "SNMP ERROR: Unknown SNMP version '$opt{protocol}'\n";
552 # Try to initialize the SNMP session
553 if ( eval { require Net::SNMP; 1 } ) {
554 ($snmp_session, $snmp_error) = Net::SNMP->session( %param );
555 if (!defined $snmp_session) {
556 printf "SNMP: %s\n", $snmp_error;
561 print "You need perl module Net::SNMP to run $NAME in SNMP mode\n";
568 # Checking if SNMP works by probing for "chassisModelName", which all
569 # servers should have
572 my $chassisModelName = '1.3.6.1.4.1.674.10892.1.300.10.1.9.1';
573 my $result = $snmp_session->get_request(-varbindlist => [$chassisModelName]);
575 # Typically if remote host isn't responding
576 if (!defined $result) {
577 printf "SNMP CRITICAL: %s\n", $snmp_session->error;
581 # If OpenManage isn't installed or is not working
582 if ($result->{$chassisModelName} =~ m{\A noSuch (Instance|Object) \z}xms) {
583 print "ERROR: (SNMP) OpenManage is not installed or is not working correctly\n";
590 # Detecting blade via SNMP
592 sub snmp_detect_blade {
593 my $DellBaseBoardType = '1.3.6.1.4.1.674.10892.1.300.80.1.7.1.1';
594 my $result = $snmp_session->get_request(-varbindlist => [$DellBaseBoardType]);
596 # Identify blade. Older models (4th and 5th gen models) and/or old
597 # OMSA (4.x) don't have this OID. If we get "noSuchInstance" or
598 # similar, we assume that this isn't a blade
599 if ($result->{$DellBaseBoardType} eq '3') {
606 # Locate the omreport binary
609 # If user has specified path to omreport
610 if (defined $opt{omreport} and -x $opt{omreport}) {
611 $omreport = qq{"$opt{omreport}"};
615 # Possible full paths for omreport
618 '/usr/bin/omreport', # default on Linux
619 '/opt/dell/srvadmin/bin/omreport', # default on Linux with OMSA 6.2.0
620 '/opt/dell/srvadmin/oma/bin/omreport.sh', # alternate on Linux
621 '/opt/dell/srvadmin/oma/bin/omreport', # alternate on Linux
622 'C:\Program Files (x86)\Dell\SysMgt\oma\bin\omreport.exe', # default on Windows x64
623 'C:\Program Files\Dell\SysMgt\oma\bin\omreport.exe', # default on Windows x32
624 'c:\progra~1\dell\sysmgt\oma\bin\omreport.exe', # 8bit legacy default on Windows x32
625 'c:\progra~2\dell\sysmgt\oma\bin\omreport.exe', # 8bit legacy default on Windows x64
628 # Find the one to use
630 foreach my $bin (@omreport_paths) {
632 $omreport = qq{"$bin"};
637 # Exit with status=UNKNOWN if OM is not installed, or we don't
638 # have permission to execute the binary
639 if (!defined $omreport) {
640 print "ERROR: Dell OpenManage Server Administrator (OMSA) is not installed\n";
647 # Checks output from 'omreport -?' and searches for arguments to
648 # omreport, to accommodate deprecated options "chassis" and "system"
649 # (on newer hardware), as well as blade servers.
651 sub check_omreport_options {
652 foreach (@{ run_command("$omreport -? 2>&1") }) {
653 if (m/\A servermodule /xms) {
654 # If "servermodule" argument to omreport exists, use it
655 # instead of argument "system"
656 $omopt_system = 'servermodule';
658 elsif (m/\A mainsystem /xms) {
659 # If "mainsystem" argument to omreport exists, use it
660 # instead of argument "chassis"
661 $omopt_chassis = 'mainsystem';
663 elsif (m/\A modularenclosure /xms) {
664 # If "modularenclusure" argument to omreport exists, assume
665 # that this is a blade
673 # Read the blacklist option and return a hash containing the
674 # blacklisted components
680 if (scalar @{ $opt{blacklist} } >= 0) {
681 foreach my $black (@{ $opt{blacklist} }) {
684 open my $BL, '<', $black
685 or do { report('other', "Couldn't open blacklist file $black: $!", $E_UNKNOWN)
698 return {} if $#bl < 0;
700 # Parse blacklist string, put in hash
701 foreach my $black (@bl) {
702 my @comps = split m{/}xms, $black;
703 foreach my $c (@comps) {
704 next if $c !~ m/=/xms;
705 my ($key, $val) = split /=/xms, $c;
706 my @vals = split /,/xms, $val;
707 $blacklist{$key} = \@vals;
715 # Read the check option and adjust the hash %check, which is a rough
716 # list of components to be checked
721 # Adjust checking based on the '--all' option
724 if (defined $opt{only} and $opt{only} !~ m{\A critical|warning \z}xms) {
725 print qq{ERROR: Wrong simultaneous usage of the "--all" and "--only" options\n};
728 if (scalar @{ $opt{check} } > 0) {
729 print qq{ERROR: Wrong simultaneous usage of the "--all" and "--check" options\n};
733 # set the check hash to check everything
734 map { $_ = 1 } values %check;
739 # Adjust checking based on the '--only' option
740 if (defined $opt{only} and $opt{only} !~ m{\A critical|warning \z}xms) {
742 if (scalar @{ $opt{check} } > 0) {
743 print qq{ERROR: Wrong simultaneous usage of the "--only" and "--check" options\n};
746 if (! exists $check{$opt{only}} && $opt{only} ne 'chassis') {
747 print qq{ERROR: "$opt{only}" is not a known keyword for the "--only" option\n};
751 # reset the check hash
752 map { $_ = 0 } values %check;
754 # adjust the check hash
755 if ($opt{only} eq 'chassis') {
756 map { $check{$_} = 1 } qw(memory fans power temp cpu voltage
757 batteries amperage intrusion esmhealth);
760 $check{$opt{only}} = 1;
766 # Adjust checking based on the '--check' option
767 if (scalar @{ $opt{check} } >= 0) {
768 foreach my $check (@{ $opt{check} }) {
771 open my $CL, '<', $check
772 or do { report('other', "Couldn't open check file $check: $!", $E_UNKNOWN) and return };
785 # Parse checklist string, put in hash
786 foreach my $check (@cl) {
787 my @checks = split /,/xms, $check;
788 foreach my $c (@checks) {
789 next if $c !~ m/=/xms;
790 my ($key, $val) = split /=/xms, $c;
795 # Check if we should check global health status
797 foreach (keys %check) {
798 next CHECK_KEY if $_ eq 'esmlog'; # not part of global status
799 next CHECK_KEY if $_ eq 'alertlog'; # not part of global status
801 if ($check{$_} == 0) { # found something with checking turned off
811 # Runs omreport and returns an array of anonymous hashes containing
813 # Takes one argument: string containing parameters to omreport
820 # Errors that are OK. Some low-end poweredge (and blades) models
821 # don't have RAID controllers, intrusion detection sensor, or
822 # redundant/instrumented power supplies etc.
825 Intrusion\sinformation\sis\snot\sfound\sfor\sthis\ssystem # No intrusion probe
826 | No\sinstrumented\spower\ssupplies\sfound\son\sthis\ssystem # No instrumented PS (blades/low-end)
827 | No\scontrollers\sfound # No RAID controller
828 | No\sbattery\sprobes\sfound\son\sthis\ssystem # No battery probes
829 | Invalid\scommand:\spwrmonitoring # Older OMSAs lack this command(?)
830 # | Current\sprobes\snot\sfound # No power monitoring capability
833 # Errors that are OK on blade servers
836 No\sfan\sprobes\sfound\son\sthis\ssystem # No fan probes
839 # Run omreport and fetch output
840 my $rawtext = slurp_command("$omreport $command -fmt ssv 2>&1");
841 return [] if !defined $rawtext;
843 # Workaround for Openmanage BUG introduced in OMSA 5.5.0
844 $rawtext =~ s{\n;}{;}gxms if $command eq 'storage controller';
846 # Openmanage sometimes puts a linebreak between "Error" and the
848 $rawtext =~ s{^Error\s*\n}{Error: }xms;
850 # Parse output, store in array
851 for ((split m{\n}xms, $rawtext)) {
853 next if m{$ok_errors}xms;
854 next if ($blade and m{$ok_blade_errors}xms);
855 report('other', "Problem running 'omreport $command': $_", $E_UNKNOWN);
858 next if !m/(.*?;){2}/xms; # ignore lines with less than 3 fields
859 my @vals = split /;/xms;
860 if ($vals[0] =~ m/\A (Index|ID|Severity|Processor|Current\sSpeed) \z/xms) {
865 push @output, { map { $_ => $vals[$i++] } @keys };
870 # Finally, return the collected information
876 # Checks if a component is blacklisted. Returns 1 if the component is
877 # blacklisted, 0 otherwise. Takes two arguments:
878 # arg1: component name
879 # arg2: component id or index
882 my $name = shift; # component name
883 my $id = shift; # component id
884 my $ret = 0; # return value
886 if (defined $blacklist{$name}) {
887 foreach my $comp (@{ $blacklist{$name} }) {
888 if (defined $id and ($comp eq $id or uc($comp) eq 'ALL')) {
897 # Converts the NexusID from SNMP to our version
900 $nexus =~ s{\A \\}{}xms;
901 $nexus =~ s{\\}{:}gxms;
905 # Sets custom temperature thresholds based on user supplied options
906 sub custom_temperature_thresholds {
907 my $type = shift; # type of threshold, either w (warning) or c (critical)
908 my %thres = (); # will contain the thresholds
909 my @limits = (); # holds the input
911 my @opt = $type eq 'w' ? @{ $opt{warning} } : @{ $opt{critical} };
913 if (scalar @opt >= 0) {
914 foreach my $t (@opt) {
918 or do { report('other', "Couldn't open temperature threshold file $t: $!",
919 $E_UNKNOWN) and return {} };
930 # Parse checklist string, put in hash
931 foreach my $th (@limits) {
932 my @tmp = split m{,}xms, $th;
933 foreach my $t (@tmp) {
934 next if $t !~ m{=}xms;
935 my ($key, $val) = split m{=}xms, $t;
936 if ($val =~ m{/}xms) {
937 my ($max, $min) = split m{/}xms, $val;
938 $thres{$key}{max} = $max;
939 $thres{$key}{min} = $min;
942 $thres{$key}{max} = $val;
951 # Gets the output from SNMP result according to the OIDs checked
952 sub get_snmp_output {
953 my ($result,$oidref) = @_;
956 foreach my $oid (keys %{ $result }) {
957 my @dummy = split /\./xms, $oid;
960 my $foo = join q{.}, @dummy;
961 if (exists $oidref->{$foo}) {
962 $output[$id]{$oidref->{$foo}} = $result->{$oid};
969 # Map the controller or other item in-place
971 my ($key, $val, $list) = @_;
973 foreach my $lst (@{ $list }) {
974 if (!exists $lst->{$key}) {
981 # Return the URL for official Dell documentation for a specific
983 sub documentation_url {
986 # create model short form, e.g. "r710"
987 $model =~ s{\A PowerEdge \s (.+?) \z}{lc($1)}exms;
989 # special case for blades (e.g. M600, M710), they have common
991 $model =~ s{\A m\d+ \z}{m}xms;
993 return 'http://support.dell.com/support/edocs/systems/pe' . $model . '/';
996 # Return the URL for warranty information for a server with a given
997 # serial number (servicetag)
1001 # Dell support sites for different parts of the world
1004 'emea' => 'http://support.euro.dell.com/support/topics/topic.aspx/emea/shared/support/my_systems_info/',
1005 'ap' => 'http://supportapj.dell.com/support/topics/topic.aspx/ap/shared/support/my_systems_info/en/details?',
1006 'glob' => 'http://support.dell.com/support/topics/global.aspx/support/my_systems_info/details?',
1009 # warranty URLs for different country codes
1013 'at' => $supportsite{emea} . 'de/details?c=at&l=de&ServiceTag=', # Austria
1014 'be' => $supportsite{emea} . 'nl/details?c=be&l=nl&ServiceTag=', # Belgium
1015 'cz' => $supportsite{emea} . 'cs/details?c=cz&l=cs&ServiceTag=', # Czech Republic
1016 'de' => $supportsite{emea} . 'de/details?c=de&l=de&ServiceTag=', # Germany
1017 'dk' => $supportsite{emea} . 'da/details?c=dk&l=da&ServiceTag=', # Denmark
1018 'es' => $supportsite{emea} . 'es/details?c=es&l=es&ServiceTag=', # Spain
1019 'fi' => $supportsite{emea} . 'fi/details?c=fi&l=fi&ServiceTag=', # Finland
1020 'fr' => $supportsite{emea} . 'fr/details?c=fr&l=fr&ServiceTag=', # France
1021 'gr' => $supportsite{emea} . 'en/details?c=gr&l=el&ServiceTag=', # Greece
1022 'it' => $supportsite{emea} . 'it/details?c=it&l=it&ServiceTag=', # Italy
1023 'il' => $supportsite{emea} . 'en/details?c=il&l=en&ServiceTag=', # Israel
1024 'me' => $supportsite{emea} . 'en/details?c=me&l=en&ServiceTag=', # Middle East
1025 'no' => $supportsite{emea} . 'no/details?c=no&l=no&ServiceTag=', # Norway
1026 'nl' => $supportsite{emea} . 'nl/details?c=nl&l=nl&ServiceTag=', # The Netherlands
1027 'pl' => $supportsite{emea} . 'pl/details?c=pl&l=pl&ServiceTag=', # Poland
1028 'pt' => $supportsite{emea} . 'en/details?c=pt&l=pt&ServiceTag=', # Portugal
1029 'ru' => $supportsite{emea} . 'ru/details?c=ru&l=ru&ServiceTag=', # Russia
1030 'se' => $supportsite{emea} . 'sv/details?c=se&l=sv&ServiceTag=', # Sweden
1031 'uk' => $supportsite{emea} . 'en/details?c=uk&l=en&ServiceTag=', # United Kingdom
1032 'za' => $supportsite{emea} . 'en/details?c=za&l=en&ServiceTag=', # South Africa
1034 'br' => $supportsite{glob} . 'c=br&l=pt&ServiceTag=', # Brazil
1035 'ca' => $supportsite{glob} . 'c=ca&l=en&ServiceTag=', # Canada
1036 'mx' => $supportsite{glob} . 'c=mx&l=es&ServiceTag=', # Mexico
1037 'us' => $supportsite{glob} . 'c=us&l=en&ServiceTag=', # USA
1039 'au' => $supportsite{ap} . 'c=au&l=en&ServiceTag=', # Australia
1040 'cn' => $supportsite{ap} . 'c=cn&l=zh&ServiceTag=', # China
1041 'in' => $supportsite{ap} . 'c=in&l=en&ServiceTag=', # India
1043 'XX' => $supportsite{glob} . 'ServiceTag=', # default
1046 if (exists $url{$opt{htmlinfo}}) {
1047 return $url{$opt{htmlinfo}} . $tag;
1050 return $url{XX} . $tag;
1056 #---------------------------------------------------------------------
1058 #---------------------------------------------------------------------
1060 #-----------------------------------------
1061 # Check global health status
1062 #-----------------------------------------
1068 # Checks global status, i.e. both storage and chassis
1070 my $systemStateGlobalSystemStatus = '1.3.6.1.4.1.674.10892.1.200.10.1.2.1';
1071 my $result = $snmp_session->get_request(-varbindlist => [$systemStateGlobalSystemStatus]);
1072 if (!defined $result) {
1073 printf "SNMP ERROR [global]: %s\n", $snmp_error;
1076 $health = $status2nagios{$snmp_status{$result->{$systemStateGlobalSystemStatus}}};
1080 # NB! This does not check storage, only chassis...
1082 foreach (@{ run_command("$omreport $omopt_system -fmt ssv") }) {
1084 next if m/\A SEVERITY;COMPONENT/xms;
1085 if (m/\A (.+?);Main\sSystem(\sChassis)? /xms) {
1086 $health = $status2nagios{$1};
1097 #-----------------------------------------
1098 # STORAGE: Check controllers
1099 #-----------------------------------------
1100 sub check_controllers {
1108 my $firmware = undef;
1115 '1.3.6.1.4.1.674.10893.1.20.130.1.1.1' => 'controllerNumber',
1116 '1.3.6.1.4.1.674.10893.1.20.130.1.1.2' => 'controllerName',
1117 '1.3.6.1.4.1.674.10893.1.20.130.1.1.5' => 'controllerState',
1118 '1.3.6.1.4.1.674.10893.1.20.130.1.1.8' => 'controllerFWVersion',
1119 '1.3.6.1.4.1.674.10893.1.20.130.1.1.38' => 'controllerComponentStatus',
1120 '1.3.6.1.4.1.674.10893.1.20.130.1.1.39' => 'controllerNexusID',
1121 '1.3.6.1.4.1.674.10893.1.20.130.1.1.41' => 'controllerDriverVersion',
1122 '1.3.6.1.4.1.674.10893.1.20.130.1.1.44' => 'controllerMinFWVersion',
1123 '1.3.6.1.4.1.674.10893.1.20.130.1.1.45' => 'controllerMinDriverVersion',
1126 # We use get_table() here for the odd case where a server has
1127 # two or more controllers, and where some OIDs are missing on
1128 # one of the controllers.
1129 my $controllerTable = '1.3.6.1.4.1.674.10893.1.20.130.1';
1130 my $result = $snmp_session->get_table(-baseoid => $controllerTable);
1132 # No controllers is OK
1133 return if !defined $result;
1135 @output = @{ get_snmp_output($result, \%ctrl_oid) };
1138 @output = @{ run_omreport('storage controller') };
1152 foreach my $out (@output) {
1154 $id = $out->{'controllerNumber'} - 1;
1155 $name = $out->{'controllerName'};
1156 $state = $ctrl_state{$out->{'controllerState'}};
1157 $status = $snmp_status{$out->{'controllerComponentStatus'}};
1158 $minfw = exists $out->{'controllerMinFWVersion'}
1159 ? $out->{'controllerMinFWVersion'} : undef;
1160 $mindr = exists $out->{'controllerMinDriverVersion'}
1161 ? $out->{'controllerMinDriverVersion'} : undef;
1162 $firmware = exists $out->{controllerFWVersion}
1163 ? $out->{controllerFWVersion} : 'N/A';
1164 $driver = exists $out->{controllerDriverVersion}
1165 ? $out->{controllerDriverVersion} : 'N/A';
1166 $nexus = convert_nexus($out->{controllerNexusID});
1170 $name = $out->{Name};
1171 $state = $out->{State};
1172 $status = $out->{Status};
1173 $minfw = $out->{'Minimum Required Firmware Version'} ne 'Not Applicable'
1174 ? $out->{'Minimum Required Firmware Version'} : undef;
1175 $mindr = $out->{'Minimum Required Driver Version'} ne 'Not Applicable'
1176 ? $out->{'Minimum Required Driver Version'} : undef;
1177 $firmware = $out->{'Firmware Version'} ne 'Not Applicable'
1178 ? $out->{'Firmware Version'} : 'N/A';
1179 $driver = $out->{'Driver Version'} ne 'Not Applicable'
1180 ? $out->{'Driver Version'} : 'N/A';
1184 $name =~ s{\s+\z}{}xms; # remove trailing whitespace
1185 push @controllers, $id;
1187 # Collecting some storage info
1188 $sysinfo{'controller'}{$id}{'id'} = $nexus;
1189 $sysinfo{'controller'}{$id}{'name'} = $name;
1190 $sysinfo{'controller'}{$id}{'driver'} = $driver;
1191 $sysinfo{'controller'}{$id}{'firmware'} = $firmware;
1193 next CTRL if blacklisted('ctrl', $nexus);
1195 # Special case: old firmware
1196 if (!blacklisted('ctrl_fw', $id) && defined $minfw) {
1198 my $msg = sprintf q{Controller %d [%s]: Firmware '%s' is out of date},
1199 $id, $name, $firmware;
1200 report('storage', $msg, $E_WARNING, $nexus);
1202 # Special case: old driver
1203 if (!blacklisted('ctrl_driver', $id) && defined $mindr) {
1205 my $msg = sprintf q{Controller %d [%s]: Driver '%s' is out of date},
1206 $id, $name, $driver;
1207 report('storage', $msg, $E_WARNING, $nexus);
1210 if ($status eq 'Ok' or ($status eq 'Non-Critical'
1211 and (defined $minfw or defined $mindr))) {
1212 my $msg = sprintf 'Controller %d [%s] is %s',
1214 report('storage', $msg, $E_OK, $nexus);
1218 my $msg = sprintf 'Controller %d [%s] needs attention: %s',
1220 report('storage', $msg, $status2nagios{$status}, $nexus);
1227 #-----------------------------------------
1228 # STORAGE: Check physical drives
1229 #-----------------------------------------
1230 sub check_physical_disks {
1231 return if $#controllers == -1;
1241 my $vendor = undef; # disk vendor
1242 my $product = undef; # product ID
1243 my $capacity = undef; # disk length (size) in bytes
1249 '1.3.6.1.4.1.674.10893.1.20.130.4.1.1' => 'arrayDiskNumber',
1250 '1.3.6.1.4.1.674.10893.1.20.130.4.1.2' => 'arrayDiskName',
1251 '1.3.6.1.4.1.674.10893.1.20.130.4.1.3' => 'arrayDiskVendor',
1252 '1.3.6.1.4.1.674.10893.1.20.130.4.1.4' => 'arrayDiskState',
1253 '1.3.6.1.4.1.674.10893.1.20.130.4.1.6' => 'arrayDiskProductID',
1254 '1.3.6.1.4.1.674.10893.1.20.130.4.1.9' => 'arrayDiskEnclosureID',
1255 '1.3.6.1.4.1.674.10893.1.20.130.4.1.10' => 'arrayDiskChannel',
1256 '1.3.6.1.4.1.674.10893.1.20.130.4.1.11' => 'arrayDiskLengthInMB',
1257 '1.3.6.1.4.1.674.10893.1.20.130.4.1.15' => 'arrayDiskTargetID',
1258 '1.3.6.1.4.1.674.10893.1.20.130.4.1.16' => 'arrayDiskLunID',
1259 '1.3.6.1.4.1.674.10893.1.20.130.4.1.24' => 'arrayDiskComponentStatus',
1260 '1.3.6.1.4.1.674.10893.1.20.130.4.1.26' => 'arrayDiskNexusID',
1261 '1.3.6.1.4.1.674.10893.1.20.130.4.1.31' => 'arrayDiskSmartAlertIndication',
1262 '1.3.6.1.4.1.674.10893.1.20.130.5.1.5' => 'arrayDiskEnclosureConnectionEnclosureNumber',
1263 '1.3.6.1.4.1.674.10893.1.20.130.5.1.7' => 'arrayDiskEnclosureConnectionControllerNumber',
1265 my $result = $snmp_session->get_entries(-columns => [keys %pdisk_oid]);
1267 if (!defined $result) {
1268 printf "SNMP ERROR [storage / pdisk]: %s.\n", $snmp_session->error;
1269 $snmp_session->close;
1273 @output = @{ get_snmp_output($result, \%pdisk_oid) };
1276 foreach my $c (@controllers) {
1277 push @output, @{ run_omreport("storage pdisk controller=$c") };
1278 map_item('ctrl', $c, \@output);
1296 28 => 'Diagnostics',
1297 34 => 'Predictive failure',
1298 35 => 'Initializing',
1301 41 => 'Unsupported',
1302 53 => 'Incompatible',
1305 # Check physical disks on each of the controllers
1307 foreach my $out (@output) {
1309 $name = $out->{arrayDiskName};
1310 if ($name =~ m{.*\d+:\d+:\d+\z}xms) {
1311 $id = join q{:}, ($out->{arrayDiskChannel}, $out->{arrayDiskEnclosureID},
1312 $out->{arrayDiskTargetID});
1315 $id = join q{:}, ($out->{arrayDiskChannel}, $out->{arrayDiskTargetID});
1317 $state = $pdisk_state{$out->{arrayDiskState}};
1318 $status = $snmp_status{$out->{arrayDiskComponentStatus}};
1319 $fpred = $out->{arrayDiskSmartAlertIndication} == 2 ? 1 : 0;
1321 $ctrl = exists $out->{arrayDiskEnclosureConnectionControllerNumber}
1322 ? $out->{arrayDiskEnclosureConnectionControllerNumber} - 1
1324 $nexus = convert_nexus($out->{arrayDiskNexusID});
1325 $vendor = $out->{arrayDiskVendor};
1326 $product = $out->{arrayDiskProductID};
1327 $capacity = $out->{arrayDiskLengthInMB} * 1024**2;
1331 $name = $out->{'Name'};
1332 $state = $out->{'State'};
1333 $status = $out->{'Status'};
1334 $fpred = lc($out->{'Failure Predicted'}) eq 'yes' ? 1 : 0;
1335 $progr = ' [' . $out->{'Progress'} . ']';
1336 $ctrl = $out->{'ctrl'};
1337 $nexus = join q{:}, $out->{ctrl}, $id;
1338 $vendor = $out->{'Vendor ID'};
1339 $product = $out->{'Product ID'};
1340 $capacity = $out->{'Capacity'};
1341 $capacity =~ s{\A .*? \((\d+) \s bytes\) \z}{$1}xms;
1344 next PDISK if blacklisted('pdisk', $nexus);
1347 $vendor =~ s{\s+\z}{}xms; # remove trailing whitespace
1348 $product =~ s{\s+\z}{}xms; # remove trailing whitespace
1350 # Calculate human readable capacity
1351 $capacity = ceil($capacity / 1000**3) >= 1000
1352 ? sprintf '%.1fTB', ($capacity / 1000**4)
1353 : sprintf '%.0fGB', ($capacity / 1000**3);
1354 $capacity = '450GB' if $capacity eq '449GB'; # quick fix for 450GB disks
1355 $capacity = '300GB' if $capacity eq '299GB'; # quick fix for 300GB disks
1356 $capacity = '146GB' if $capacity eq '147GB'; # quick fix for 146GB disks
1358 # Capitalize only the first letter of the vendor name
1359 $vendor = (substr $vendor, 0, 1) . lc (substr $vendor, 1, length $vendor);
1361 # Remove unnecessary trademark rubbish from vendor name
1362 $vendor =~ s{\(tm\)\z}{}xms;
1364 # Special case: Failure predicted
1365 if ($status eq 'Non-Critical' and $fpred) {
1366 my $msg = sprintf '%s [%s %s, %s] on ctrl %d needs attention: Failure Predicted',
1367 $name, $vendor, $product, $capacity, $ctrl;
1368 report('storage', $msg, $E_WARNING, $nexus);
1370 # Special case: Rebuilding
1371 elsif ($state eq 'Rebuilding') {
1372 my $msg = sprintf '%s [%s] on ctrl %d is %s%s',
1373 $name, $capacity, $ctrl, $state, $progr;
1374 report('storage', $msg, $E_WARNING, $nexus);
1377 elsif ($status ne 'Ok') {
1378 my $msg = sprintf '%s [%s %s, %s] on ctrl %d needs attention: %s',
1379 $name, $vendor, $product, $capacity, $ctrl, $state;
1380 report('storage', $msg, $status2nagios{$status}, $nexus);
1384 my $msg = sprintf '%s [%s] on ctrl %d is %s',
1385 $name, $capacity, $ctrl, $state;
1386 report('storage', $msg, $E_OK, $nexus);
1393 #-----------------------------------------
1394 # STORAGE: Check logical drives
1395 #-----------------------------------------
1396 sub check_virtual_disks {
1397 return if $#controllers == -1;
1414 '1.3.6.1.4.1.674.10893.1.20.140.1.1.3' => 'virtualDiskDeviceName',
1415 '1.3.6.1.4.1.674.10893.1.20.140.1.1.4' => 'virtualDiskState',
1416 '1.3.6.1.4.1.674.10893.1.20.140.1.1.6' => 'virtualDiskLengthInMB',
1417 '1.3.6.1.4.1.674.10893.1.20.140.1.1.13' => 'virtualDiskLayout',
1418 '1.3.6.1.4.1.674.10893.1.20.140.1.1.17' => 'virtualDiskTargetID',
1419 '1.3.6.1.4.1.674.10893.1.20.140.1.1.20' => 'virtualDiskComponentStatus',
1420 '1.3.6.1.4.1.674.10893.1.20.140.1.1.21' => 'virtualDiskNexusID',
1422 my $result = $snmp_session->get_entries(-columns => [keys %vdisk_oid]);
1424 # No logical drives is OK
1425 return if !defined $result;
1427 @output = @{ get_snmp_output($result, \%vdisk_oid) };
1430 foreach my $c (@controllers) {
1431 push @output, @{ run_omreport("storage vdisk controller=$c") };
1432 map_item('ctrl', $c, \@output);
1445 16 => 'Regenerating',
1448 32 => 'Reconstructing',
1449 35 => 'Initializing',
1450 36 => 'Background Initialization',
1451 38 => 'Resynching Paused',
1452 52 => 'Permanently Degraded',
1453 54 => 'Degraded Redundancy',
1458 1 => 'Concatenated',
1465 19 => 'Concatenated RAID 1',
1469 # Check virtual disks on each of the controllers
1471 foreach my $out (@output) {
1473 $id = $out->{virtualDiskTargetID};
1474 $dev = $out->{virtualDiskDeviceName};
1475 $state = $vdisk_state{$out->{virtualDiskState}};
1476 $status = $snmp_status{$out->{virtualDiskComponentStatus}};
1477 $layout = $vdisk_layout{$out->{virtualDiskLayout}};
1478 $size = sprintf '%.2f GB', $out->{virtualDiskLengthInMB} / 1024;
1479 $progr = q{}; # can't get this from SNMP(?)
1480 $nexus = convert_nexus($out->{virtualDiskNexusID});
1481 $ctrl = $nexus; # We use the nexus id to get the controller id
1482 $ctrl =~ s{\A (\d+):\d+ \z}{$1}xms;
1486 $dev = $out->{'Device Name'};
1487 $state = $out->{State};
1488 $status = $out->{Status};
1489 $layout = $out->{Layout};
1490 $size = $out->{Size};
1491 $progr = ' [' . $out->{Progress} . ']';
1492 $size =~ s{\A (.*GB).* \z}{$1}xms;
1493 $nexus = join q{:}, $out->{ctrl}, $id;
1494 $ctrl = $out->{ctrl};
1497 next VDISK if blacklisted('vdisk', $nexus);
1500 # The device name is undefined sometimes
1501 $dev = q{} if !defined $dev;
1503 # Special case: Regenerating
1504 if ($state eq 'Regenerating') {
1505 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d is %s%s},
1506 $id, $dev, $layout, $size, $ctrl, $state, $progr;
1507 report('storage', $msg, $E_WARNING, $nexus);
1510 elsif ($status ne 'Ok') {
1511 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d needs attention: %s},
1512 $id, $dev, $layout, $size, $ctrl, $state;
1513 report('storage', $msg, $status2nagios{$status}, $nexus);
1517 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d is %s},
1518 $id, $dev, $layout, $size, $ctrl, $state;
1519 report('storage', $msg, $E_OK, $nexus);
1526 #-----------------------------------------
1527 # STORAGE: Check cache batteries
1528 #-----------------------------------------
1529 sub check_cache_battery {
1530 return if $#controllers == -1;
1537 my $learn = undef; # learn state
1538 my $pred = undef; # battery's ability to be charged
1544 '1.3.6.1.4.1.674.10893.1.20.130.15.1.4' => 'batteryState',
1545 '1.3.6.1.4.1.674.10893.1.20.130.15.1.6' => 'batteryComponentStatus',
1546 '1.3.6.1.4.1.674.10893.1.20.130.15.1.9' => 'batteryNexusID',
1547 '1.3.6.1.4.1.674.10893.1.20.130.15.1.10' => 'batteryPredictedCapacity',
1548 '1.3.6.1.4.1.674.10893.1.20.130.15.1.12' => 'batteryLearnState',
1549 '1.3.6.1.4.1.674.10893.1.20.130.16.1.5' => 'batteryConnectionControllerNumber',
1551 my $result = $snmp_session->get_entries(-columns => [keys %bat_oid]);
1553 # No cache battery is OK
1554 return if !defined $result;
1556 @output = @{ get_snmp_output($result, \%bat_oid) };
1559 foreach my $c (@controllers) {
1560 push @output, @{ run_omreport("storage battery controller=$c") };
1561 map_item('ctrl', $c, \@output);
1571 7 => 'Reconditioning',
1579 # Specifies the learn state activity of the battery
1589 # This property displays the battery's ability to be charged
1592 1 => 'Failed', # The battery cannot be charged and needs to be replaced
1593 2 => 'Ready', # The battery can be charged to full capacity
1594 4 => 'Unknown', # The battery is completing a Learn cycle. The charge capacity of the
1595 # battery cannot be determined until the Learn cycle is complete
1598 # Check battery on each of the controllers
1600 foreach my $out (@output) {
1602 $state = $bat_state{$out->{batteryState}};
1603 $status = $snmp_status{$out->{batteryComponentStatus}};
1604 $learn = exists $out->{batteryLearnState}
1605 ? $bat_learn_state{$out->{batteryLearnState}} : undef;
1606 $pred = exists $out->{batteryPredictedCapacity}
1607 ? $bat_pred_cap{$out->{batteryPredictedCapacity}} : undef;
1608 $ctrl = $out->{batteryConnectionControllerNumber} - 1;
1609 $nexus = convert_nexus($out->{batteryNexusID});
1611 $id =~ s{\A \d+:(\d+) \z}{$1}xms;
1615 $state = $out->{'State'};
1616 $status = $out->{'Status'};
1617 $learn = $out->{'Learn State'};
1618 $pred = $out->{'Predicted Capacity Status'};
1619 $ctrl = $out->{'ctrl'};
1620 $nexus = join q{:}, $out->{ctrl}, $id;
1623 next BATTERY if blacklisted('bat', $nexus);
1625 # Special case: Charging
1626 if ($state eq 'Charging') {
1627 if ($pred eq 'Failed') {
1628 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [replace battery]',
1629 $id, $ctrl, $state, $pred;
1630 report('storage', $msg, $E_CRITICAL, $nexus);
1633 next BATTERY if blacklisted('bat_charge', $nexus);
1634 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1635 $id, $ctrl, $state, $pred;
1636 report('storage', $msg, $E_WARNING, $nexus);
1639 # Special case: Learning (battery learns its capacity)
1640 elsif ($state eq 'Learning') {
1641 if ($learn eq 'Failed') {
1642 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s)',
1643 $id, $ctrl, $state, $learn;
1644 report('storage', $msg, $E_CRITICAL, $nexus);
1647 next BATTERY if blacklisted('bat_charge', $nexus);
1648 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1649 $id, $ctrl, $state, $learn;
1650 report('storage', $msg, $E_WARNING, $nexus);
1653 # Special case: Power Low (first part of recharge cycle)
1654 elsif ($state eq 'Power Low') {
1655 next BATTERY if blacklisted('bat_charge', $nexus);
1656 my $msg = sprintf 'Cache battery %d in controller %d is %s [probably harmless]',
1658 report('storage', $msg, $E_WARNING, $nexus);
1660 # Special case: Degraded and Non-Critical (usually part of recharge cycle)
1661 elsif ($state eq 'Degraded' && $status eq 'Non-Critical') {
1662 next BATTERY if blacklisted('bat_charge', $nexus);
1663 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1664 $id, $ctrl, $state, $status;
1665 report('storage', $msg, $E_WARNING, $nexus);
1668 elsif ($status ne 'Ok') {
1669 my $msg = sprintf 'Cache battery %d in controller %d needs attention: %s (%s)',
1670 $id, $ctrl, $state, $status;
1671 report('storage', $msg, $status2nagios{$status}, $nexus);
1675 my $msg = sprintf 'Cache battery %d in controller %d is %s',
1677 report('storage', $msg, $E_OK, $nexus);
1684 #-----------------------------------------
1685 # STORAGE: Check connectors (channels)
1686 #-----------------------------------------
1687 sub check_connectors {
1688 return if $#controllers == -1;
1702 '1.3.6.1.4.1.674.10893.1.20.130.2.1.1' => 'channelNumber',
1703 '1.3.6.1.4.1.674.10893.1.20.130.2.1.2' => 'channelName',
1704 '1.3.6.1.4.1.674.10893.1.20.130.2.1.3' => 'channelState',
1705 '1.3.6.1.4.1.674.10893.1.20.130.2.1.8' => 'channelComponentStatus',
1706 '1.3.6.1.4.1.674.10893.1.20.130.2.1.9' => 'channelNexusID',
1707 '1.3.6.1.4.1.674.10893.1.20.130.2.1.11' => 'channelBusType',
1709 my $result = $snmp_session->get_entries(-columns => [keys %conn_oid]);
1711 if (!defined $result) {
1712 printf "SNMP ERROR [storage / channel]: %s.\n", $snmp_session->error;
1713 $snmp_session->close;
1717 @output = @{ get_snmp_output($result, \%conn_oid) };
1720 foreach my $c (@controllers) {
1721 push @output, @{ run_omreport("storage connector controller=$c") };
1722 map_item('ctrl', $c, \@output);
1740 3 => 'Fibre Channel',
1747 # Check connectors on each of the controllers
1749 foreach my $out (@output) {
1751 $id = $out->{channelNumber} - 1;
1752 $name = $out->{channelName};
1753 $state = $conn_state{$out->{channelState}};
1754 $status = $snmp_status{$out->{channelComponentStatus}};
1755 $type = $conn_bustype{$out->{channelBusType}};
1756 $nexus = convert_nexus($out->{channelNexusID});
1758 $ctrl =~ s{(\d+):\d+}{$1}xms;
1762 $name = $out->{'Name'};
1763 $state = $out->{'State'};
1764 $status = $out->{'Status'};
1765 $type = $out->{'Connector Type'};
1766 $ctrl = $out->{ctrl};
1767 $nexus = join q{:}, $out->{ctrl}, $id;
1770 next CHANNEL if blacklisted('conn', $nexus);
1772 my $msg = sprintf '%s [%s] on controller %d is %s',
1773 $name, $type, $ctrl, $state;
1774 report('storage', $msg, $status2nagios{$status}, $nexus);
1780 #-----------------------------------------
1781 # STORAGE: Check enclosures
1782 #-----------------------------------------
1783 sub check_enclosures {
1789 my $firmware = undef;
1796 '1.3.6.1.4.1.674.10893.1.20.130.3.1.1' => 'enclosureNumber',
1797 '1.3.6.1.4.1.674.10893.1.20.130.3.1.2' => 'enclosureName',
1798 '1.3.6.1.4.1.674.10893.1.20.130.3.1.4' => 'enclosureState',
1799 '1.3.6.1.4.1.674.10893.1.20.130.3.1.19' => 'enclosureChannelNumber',
1800 '1.3.6.1.4.1.674.10893.1.20.130.3.1.24' => 'enclosureComponentStatus',
1801 '1.3.6.1.4.1.674.10893.1.20.130.3.1.25' => 'enclosureNexusID',
1802 '1.3.6.1.4.1.674.10893.1.20.130.3.1.26' => 'enclosureFirmwareVersion',
1804 my $result = $snmp_session->get_entries(-columns => [keys %encl_oid]);
1806 # No enclosures is OK
1807 return if !defined $result;
1809 @output = @{ get_snmp_output($result, \%encl_oid) };
1812 foreach my $c (@controllers) {
1813 push @output, @{ run_omreport("storage enclosure controller=$c") };
1814 map_item('ctrl', $c, \@output);
1829 foreach my $out (@output) {
1831 $id = $out->{'enclosureNumber'} - 1;
1832 $name = $out->{'enclosureName'};
1833 $state = $encl_state{$out->{'enclosureState'}};
1834 $status = $snmp_status{$out->{'enclosureComponentStatus'}};
1835 $firmware = exists $out->{enclosureFirmwareVersion}
1836 ? $out->{enclosureFirmwareVersion} : 'N/A';
1837 $nexus = convert_nexus($out->{enclosureNexusID});
1839 $ctrl =~ s{\A (\d+):.* \z}{$1}xms;
1843 $name = $out->{Name};
1844 $state = $out->{State};
1845 $status = $out->{Status};
1846 $firmware = $out->{'Firmware Version'} ne 'Not Applicable'
1847 ? $out->{'Firmware Version'} : 'N/A';
1848 $nexus = join q{:}, $out->{ctrl}, $id;
1849 $ctrl = $out->{ctrl};
1852 $name =~ s{\s+\z}{}xms; # remove trailing whitespace
1853 $firmware =~ s{\s+\z}{}xms; # remove trailing whitespace
1855 # store enclosure data for future use
1856 push @enclosures, { 'id' => $id,
1857 'ctrl' => $out->{ctrl},
1860 # Collecting some storage info
1861 $sysinfo{'enclosure'}{$nexus}{'id'} = $nexus;
1862 $sysinfo{'enclosure'}{$nexus}{'name'} = $name;
1863 $sysinfo{'enclosure'}{$nexus}{'firmware'} = $firmware;
1865 next ENCLOSURE if blacklisted('encl', $nexus);
1867 my $msg = sprintf 'Enclosure %s [%s] on controller %d is %s',
1868 $nexus, $name, $ctrl, $state;
1869 report('storage', $msg, $status2nagios{$status}, $nexus);
1875 #-----------------------------------------
1876 # STORAGE: Check enclosure fans
1877 #-----------------------------------------
1878 sub check_enclosure_fans {
1879 return if $#controllers == -1;
1887 my $encl_id = undef;
1888 my $encl_name = undef;
1894 '1.3.6.1.4.1.674.10893.1.20.130.7.1.1' => 'fanNumber',
1895 '1.3.6.1.4.1.674.10893.1.20.130.7.1.2' => 'fanName',
1896 '1.3.6.1.4.1.674.10893.1.20.130.7.1.4' => 'fanState',
1897 '1.3.6.1.4.1.674.10893.1.20.130.7.1.11' => 'fanProbeCurrValue',
1898 '1.3.6.1.4.1.674.10893.1.20.130.7.1.15' => 'fanComponentStatus',
1899 '1.3.6.1.4.1.674.10893.1.20.130.7.1.16' => 'fanNexusID',
1900 '1.3.6.1.4.1.674.10893.1.20.130.8.1.4' => 'fanConnectionEnclosureName',
1901 '1.3.6.1.4.1.674.10893.1.20.130.8.1.5' => 'fanConnectionEnclosureNumber',
1904 my $result = $snmp_session->get_entries(-columns => [keys %fan_oid]);
1906 # No enclosure fans is OK
1907 return if !defined $result;
1909 @output = @{ get_snmp_output($result, \%fan_oid) };
1912 foreach my $enc (@enclosures) {
1913 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=fans") };
1914 map_item('ctrl', $enc->{ctrl}, \@output);
1915 map_item('encl_id', $enc->{id}, \@output);
1916 map_item('encl_name', $enc->{name}, \@output);
1931 # Check fans on each of the enclosures
1933 foreach my $out (@output) {
1935 $id = $out->{fanNumber} - 1;
1936 $name = $out->{fanName};
1937 $state = $fan_state{$out->{fanState}};
1938 $status = $snmp_status{$out->{fanComponentStatus}};
1939 $speed = $out->{fanProbeCurrValue};
1940 $encl_id = $out->{fanConnectionEnclosureNumber} - 1;
1941 $encl_name = $out->{fanConnectionEnclosureName};
1942 $nexus = convert_nexus($out->{fanNexusID});
1946 $name = $out->{'Name'};
1947 $state = $out->{'State'};
1948 $status = $out->{'Status'};
1949 $speed = $out->{'Speed'};
1950 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
1951 $encl_name = $out->{encl_name};
1952 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
1955 next FAN if blacklisted('encl_fan', $nexus);
1958 if ($status ne 'Ok') {
1959 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
1960 $name, $encl_id, $encl_name, $state;
1961 report('storage', $msg, $status2nagios{$status}, $nexus);
1965 my $msg = sprintf '%s in enclosure %s [%s] is %s (speed=%s)',
1966 $name, $encl_id, $encl_name, $state, $speed;
1967 report('storage', $msg, $E_OK, $nexus);
1974 #-----------------------------------------
1975 # STORAGE: Check enclosure power supplies
1976 #-----------------------------------------
1977 sub check_enclosure_pwr {
1978 return if $#controllers == -1;
1985 my $encl_id = undef;
1986 my $encl_name = undef;
1992 '1.3.6.1.4.1.674.10893.1.20.130.9.1.1' => 'powerSupplyNumber',
1993 '1.3.6.1.4.1.674.10893.1.20.130.9.1.2' => 'powerSupplyName',
1994 '1.3.6.1.4.1.674.10893.1.20.130.9.1.4' => 'powerSupplyState',
1995 '1.3.6.1.4.1.674.10893.1.20.130.9.1.9' => 'powerSupplyComponentStatus',
1996 '1.3.6.1.4.1.674.10893.1.20.130.9.1.10' => 'powerSupplyNexusID',
1997 '1.3.6.1.4.1.674.10893.1.20.130.10.1.4' => 'powerSupplyConnectionEnclosureName',
1998 '1.3.6.1.4.1.674.10893.1.20.130.10.1.5' => 'powerSupplyConnectionEnclosureNumber',
2000 my $result = $snmp_session->get_entries(-columns => [keys %ps_oid]);
2002 # No enclosure power supplies is OK
2003 return if !defined $result;
2005 @output = @{ get_snmp_output($result, \%ps_oid) };
2008 foreach my $enc (@enclosures) {
2009 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=pwrsupplies") };
2010 map_item('ctrl', $enc->{ctrl}, \@output);
2011 map_item('encl_id', $enc->{id}, \@output);
2012 map_item('encl_name', $enc->{name}, \@output);
2021 5 => 'Not Installed',
2027 # Check power supplies on each of the enclosures
2029 foreach my $out (@output) {
2031 $id = $out->{powerSupplyNumber};
2032 $name = $out->{powerSupplyName};
2033 $state = $ps_state{$out->{powerSupplyState}};
2034 $status = $snmp_status{$out->{powerSupplyComponentStatus}};
2035 $encl_id = $out->{powerSupplyConnectionEnclosureNumber} - 1;
2036 $encl_name = $out->{powerSupplyConnectionEnclosureName};
2037 $nexus = convert_nexus($out->{powerSupplyNexusID});
2041 $name = $out->{'Name'};
2042 $state = $out->{'State'};
2043 $status = $out->{'Status'};
2044 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2045 $encl_name = $out->{encl_name};
2046 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2049 next PS if blacklisted('encl_ps', $nexus);
2052 if ($status ne 'Ok') {
2053 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
2054 $name, $encl_id, $encl_name, $state;
2055 report('storage', $msg, $status2nagios{$status}, $nexus);
2059 my $msg = sprintf '%s in enclosure %s [%s] is %s',
2060 $name, $encl_id, $encl_name, $state;
2061 report('storage', $msg, $E_OK, $nexus);
2068 #-----------------------------------------
2069 # STORAGE: Check enclosure temperatures
2070 #-----------------------------------------
2071 sub check_enclosure_temp {
2072 return if $#controllers == -1;
2079 my $reading = undef;
2081 my $max_warn = undef;
2082 my $max_crit = undef;
2083 my $encl_id = undef;
2084 my $encl_name = undef;
2090 '1.3.6.1.4.1.674.10893.1.20.130.11.1.1' => 'temperatureProbeNumber',
2091 '1.3.6.1.4.1.674.10893.1.20.130.11.1.2' => 'temperatureProbeName',
2092 '1.3.6.1.4.1.674.10893.1.20.130.11.1.4' => 'temperatureProbeState',
2093 '1.3.6.1.4.1.674.10893.1.20.130.11.1.6' => 'temperatureProbeUnit',
2094 '1.3.6.1.4.1.674.10893.1.20.130.11.1.9' => 'temperatureProbeMaxWarning',
2095 '1.3.6.1.4.1.674.10893.1.20.130.11.1.10' => 'temperatureProbeMaxCritical',
2096 '1.3.6.1.4.1.674.10893.1.20.130.11.1.11' => 'temperatureProbeCurValue',
2097 '1.3.6.1.4.1.674.10893.1.20.130.11.1.13' => 'temperatureProbeComponentStatus',
2098 '1.3.6.1.4.1.674.10893.1.20.130.11.1.14' => 'temperatureProbeNexusID',
2099 '1.3.6.1.4.1.674.10893.1.20.130.12.1.4' => 'temperatureConnectionEnclosureName',
2100 '1.3.6.1.4.1.674.10893.1.20.130.12.1.5' => 'temperatureConnectionEnclosureNumber',
2102 my $result = $snmp_session->get_entries(-columns => [keys %temp_oid]);
2104 # No enclosure temperature probes is OK
2105 return if !defined $result;
2107 @output = @{ get_snmp_output($result, \%temp_oid) };
2110 foreach my $enc (@enclosures) {
2111 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=temps") };
2112 map_item('ctrl', $enc->{ctrl}, \@output);
2113 map_item('encl_id', $enc->{id}, \@output);
2114 map_item('encl_name', $enc->{name}, \@output);
2129 # Check temperature probes on each of the enclosures
2131 foreach my $out (@output) {
2133 $id = $out->{temperatureProbeNumber} - 1;
2134 $name = $out->{temperatureProbeName};
2135 $state = $temp_state{$out->{temperatureProbeState}};
2136 $status = $snmp_status{$out->{temperatureProbeComponentStatus}};
2137 $unit = $out->{temperatureProbeUnit};
2138 $reading = $out->{temperatureProbeCurValue};
2139 $max_warn = $out->{temperatureProbeMaxWarning};
2140 $max_crit = $out->{temperatureProbeMaxCritical};
2141 $encl_id = $out->{temperatureConnectionEnclosureNumber} - 1;
2142 $encl_name = $out->{temperatureConnectionEnclosureName};
2143 $nexus = convert_nexus($out->{temperatureProbeNexusID});
2147 $name = $out->{'Name'};
2148 $state = $out->{'State'};
2149 $status = $out->{'Status'};
2151 $reading = $out->{'Reading'}; $reading =~ s{\s*C}{}xms;
2152 $max_warn = $out->{'Maximum Warning Threshold'}; $max_warn =~ s{\s*C}{}xms;
2153 $max_crit = $out->{'Maximum Failure Threshold'}; $max_crit =~ s{\s*C}{}xms;
2154 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2155 $encl_name = $out->{encl_name};
2156 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2159 next TEMP if blacklisted('encl_temp', $nexus);
2162 if ($status ne 'Ok') {
2163 my $msg = sprintf '%s in enclosure %s [%s] is %s C at %s (%s max)',
2164 $name, $encl_id, $encl_name, $state, $reading, $max_crit;
2165 report('storage', $msg, $status2nagios{$status}, $nexus);
2169 my $msg = sprintf '%s in enclosure %s [%s]: %s C (%s max)',
2170 $name, $encl_id, $encl_name, $reading, $max_crit;
2171 report('storage', $msg, $E_OK, $nexus);
2174 # Collect performance data
2175 if (defined $opt{perfdata}) {
2176 $name =~ s{\A Temperature\sProbe\s(\d+) \z}{temp_$1}gxms;
2177 my $pkey = "enclosure_${encl_id}_${name}";
2178 my $pval = join q{;}, "${reading}C", $max_warn, $max_crit;
2179 $perfdata{$pkey} = $pval;
2186 #-----------------------------------------
2187 # STORAGE: Check enclosure management modules (EMM)
2188 #-----------------------------------------
2189 sub check_enclosure_emms {
2190 return if $#controllers == -1;
2197 my $encl_id = undef;
2198 my $encl_name = undef;
2204 '1.3.6.1.4.1.674.10893.1.20.130.13.1.1' => 'enclosureManagementModuleNumber',
2205 '1.3.6.1.4.1.674.10893.1.20.130.13.1.2' => 'enclosureManagementModuleName',
2206 '1.3.6.1.4.1.674.10893.1.20.130.13.1.4' => 'enclosureManagementModuleState',
2207 '1.3.6.1.4.1.674.10893.1.20.130.13.1.11' => 'enclosureManagementModuleComponentStatus',
2208 '1.3.6.1.4.1.674.10893.1.20.130.13.1.12' => 'enclosureManagementModuleNexusID',
2209 '1.3.6.1.4.1.674.10893.1.20.130.14.1.4' => 'enclosureManagementModuleConnectionEnclosureName',
2210 '1.3.6.1.4.1.674.10893.1.20.130.14.1.5' => 'enclosureManagementModuleConnectionEnclosureNumber',
2212 my $result = $snmp_session->get_entries(-columns => [keys %emms_oid]);
2214 # No enclosure EMMs is OK
2215 return if !defined $result;
2217 @output = @{ get_snmp_output($result, \%emms_oid) };
2220 foreach my $enc (@enclosures) {
2221 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=emms") };
2222 map_item('ctrl', $enc->{ctrl}, \@output);
2223 map_item('encl_id', $enc->{id}, \@output);
2224 map_item('encl_name', $enc->{name}, \@output);
2235 5 => 'Not Installed',
2240 # Check temperature probes on each of the enclosures
2242 foreach my $out (@output) {
2244 $id = $out->{enclosureManagementModuleNumber} - 1;
2245 $name = $out->{enclosureManagementModuleName};
2246 $state = $emms_state{$out->{enclosureManagementModuleState}};
2247 $status = $snmp_status{$out->{enclosureManagementModuleComponentStatus}};
2248 $encl_id = $out->{enclosureManagementModuleConnectionEnclosureNumber} - 1;
2249 $encl_name = $out->{enclosureManagementModuleConnectionEnclosureName};
2250 $nexus = convert_nexus($out->{enclosureManagementModuleNexusID});
2254 $name = $out->{'Name'};
2255 $state = $out->{'State'};
2256 $status = $out->{'Status'};
2257 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2258 $encl_name = $out->{encl_name};
2259 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2262 next EMM if blacklisted('encl_emm', $nexus);
2265 if ($status ne 'Ok') {
2266 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
2267 $name, $encl_id, $encl_name, $state;
2268 report('storage', $msg, $status2nagios{$status}, $nexus);
2272 my $msg = sprintf '%s in enclosure %s [%s] is %s',
2273 $name, $encl_id, $encl_name, $state;
2274 report('storage', $msg, $E_OK, $nexus);
2281 #-----------------------------------------
2282 # CHASSIS: Check memory modules
2283 #-----------------------------------------
2287 my $location = undef;
2296 '1.3.6.1.4.1.674.10892.1.1100.50.1.2.1' => 'memoryDeviceIndex',
2297 '1.3.6.1.4.1.674.10892.1.1100.50.1.5.1' => 'memoryDeviceStatus',
2298 '1.3.6.1.4.1.674.10892.1.1100.50.1.8.1' => 'memoryDeviceLocationName',
2299 '1.3.6.1.4.1.674.10892.1.1100.50.1.14.1' => 'memoryDeviceSize',
2300 '1.3.6.1.4.1.674.10892.1.1100.50.1.20.1' => 'memoryDeviceFailureModes',
2302 my $result = $snmp_session->get_entries(-columns => [keys %dimm_oid]);
2304 if (!defined $result) {
2305 printf "SNMP ERROR [memory]: %s.\n", $snmp_session->error;
2306 $snmp_session->close;
2310 @output = @{ get_snmp_output($result, \%dimm_oid) };
2313 @output = @{ run_omreport("$omopt_chassis memory") };
2316 # Note: These values are bit masks, so combination values are
2317 # possible. If value is 0 (zero), memory device has no faults.
2320 1 => 'ECC single bit correction warning rate exceeded',
2321 2 => 'ECC single bit correction failure rate exceeded',
2322 4 => 'ECC multibit fault encountered',
2323 8 => 'ECC single bit correction logging disabled',
2324 16 => 'device disabled because of spare activation',
2328 foreach my $out (@output) {
2329 @failures = (); # Initialize
2331 $index = $out->{memoryDeviceIndex};
2332 $status = $snmp_status{$out->{memoryDeviceStatus}};
2333 $location = $out->{memoryDeviceLocationName};
2334 $size = sprintf '%d MB', $out->{memoryDeviceSize}/1024;
2335 $modes = $out->{memoryDeviceFailureModes};
2337 foreach my $mask (sort keys %failure_mode) {
2338 if (($modes & $mask) != 0) { push @failures, $failure_mode{$mask}; }
2343 $index = $out->{'Type'} eq '[Not Occupied]' ? undef : $out->{'Index'};
2344 $status = $out->{'Status'};
2345 $location = $out->{'Connector Name'};
2346 $size = $out->{'Size'};
2347 if (defined $size) {
2348 $size =~ s{\s\s}{ }gxms;
2350 # Run 'omreport chassis memory index=X' to get the failures
2351 if ($status ne 'Ok' && defined $index) {
2352 foreach (@{ run_command("$omreport $omopt_chassis memory index=$index -fmt ssv") }) {
2353 if (m/\A Failures; (.+?) \z/xms) {
2354 chop(my $fail = $1);
2355 push @failures, split m{\.}xms, $fail;
2360 $location =~ s{\A \s*(.*?)\s* \z}{$1}xms;
2362 next DIMM if blacklisted('dimm', $index);
2364 # Ignore empty memory slots
2365 next DIMM if !defined $index;
2368 if ($status ne 'Ok') {
2370 if (scalar @failures == 0) {
2371 $msg = sprintf 'Memory module %d [%s, %s] needs attention (%s)',
2372 $index, $location, $size, $status;
2375 $msg = sprintf 'Memory module %d [%s, %s] needs attention: %s',
2376 $index, $location, $size, (join q{, }, @failures);
2379 report('chassis', $msg, $status2nagios{$status}, $index);
2383 my $msg = sprintf 'Memory module %d [%s, %s] is %s',
2384 $index, $location, $size, $status;
2385 report('chassis', $msg, $E_OK, $index);
2392 #-----------------------------------------
2393 # CHASSIS: Check fans
2394 #-----------------------------------------
2398 my $reading = undef;
2399 my $location = undef;
2400 my $max_crit = undef;
2401 my $max_warn = undef;
2407 '1.3.6.1.4.1.674.10892.1.700.12.1.2.1' => 'coolingDeviceIndex',
2408 '1.3.6.1.4.1.674.10892.1.700.12.1.5.1' => 'coolingDeviceStatus',
2409 '1.3.6.1.4.1.674.10892.1.700.12.1.6.1' => 'coolingDeviceReading',
2410 '1.3.6.1.4.1.674.10892.1.700.12.1.8.1' => 'coolingDeviceLocationName',
2411 '1.3.6.1.4.1.674.10892.1.700.12.1.10.1' => 'coolingDeviceUpperCriticalThreshold',
2412 '1.3.6.1.4.1.674.10892.1.700.12.1.11.1' => 'coolingDeviceUpperNonCriticalThreshold',
2414 my $result = $snmp_session->get_entries(-columns => [keys %cool_oid]);
2416 if ($blade && !defined $result) {
2419 elsif (!$blade && !defined $result) {
2420 printf "SNMP ERROR [cooling]: %s.\n", $snmp_session->error;
2421 $snmp_session->close;
2425 @output = @{ get_snmp_output($result, \%cool_oid) };
2428 @output = @{ run_omreport("$omopt_chassis fans") };
2432 foreach my $out (@output) {
2434 $index = $out->{coolingDeviceIndex};
2435 $status = $snmp_probestatus{$out->{coolingDeviceStatus}};
2436 $reading = $out->{coolingDeviceReading};
2437 $location = $out->{coolingDeviceLocationName};
2438 $max_crit = exists $out->{coolingDeviceUpperCriticalThreshold}
2439 ? $out->{coolingDeviceUpperCriticalThreshold} : 0;
2440 $max_warn = exists $out->{coolingDeviceUpperNonCriticalThreshold}
2441 ? $out->{coolingDeviceUpperNonCriticalThreshold} : 0;
2444 $index = $out->{'Index'};
2445 $status = $out->{'Status'};
2446 $reading = $out->{'Reading'};
2447 $location = $out->{'Probe Name'};
2448 $max_crit = $out->{'Maximum Failure Threshold'} ne '[N/A]'
2449 ? $out->{'Maximum Failure Threshold'} : 0;
2450 $max_warn = $out->{'Maximum Warning Threshold'} ne '[N/A]'
2451 ? $out->{'Maximum Warning Threshold'} : 0;
2452 $reading =~ s{\A (\d+).* \z}{$1}xms;
2453 $max_warn =~ s{\A (\d+).* \z}{$1}xms;
2454 $max_crit =~ s{\A (\d+).* \z}{$1}xms;
2457 next FAN if blacklisted('fan', $index);
2460 if ($status ne 'Ok') {
2461 my $msg = sprintf 'Chassis fan %d [%s] needs attention: %s',
2462 $index, $location, $status;
2463 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2464 report('chassis', $msg, $err, $index);
2467 my $msg = sprintf 'Chassis fan %d [%s]: %s',
2468 $index, $location, $reading;
2469 report('chassis', $msg, $E_OK, $index);
2472 # Collect performance data
2473 if (defined $opt{perfdata}) {
2474 my $pname = lc $location;
2475 $pname =~ s{\s}{_}gxms;
2476 $pname =~ s{proc_}{cpu#}xms;
2477 my $pkey = join q{_}, 'fan', $index, $pname;
2478 my $pval = join q{;}, "${reading}RPM", $max_warn, $max_crit;
2479 $perfdata{$pkey} = $pval;
2486 #-----------------------------------------
2487 # CHASSIS: Check power supplies
2488 #-----------------------------------------
2489 sub check_powersupplies {
2493 my $err_type = undef;
2501 '1.3.6.1.4.1.674.10892.1.600.12.1.2.1' => 'powerSupplyIndex',
2502 '1.3.6.1.4.1.674.10892.1.600.12.1.5.1' => 'powerSupplyStatus',
2503 '1.3.6.1.4.1.674.10892.1.600.12.1.7.1' => 'powerSupplyType',
2504 '1.3.6.1.4.1.674.10892.1.600.12.1.11.1' => 'powerSupplySensorState',
2505 '1.3.6.1.4.1.674.10892.1.600.12.1.12.1' => 'powerSupplyConfigurationErrorType',
2507 my $result = $snmp_session->get_entries(-columns => [keys %ps_oid]);
2509 # No instrumented PSU is OK (blades, low-end servers)
2510 return 0 if !defined $result;
2512 @output = @{ get_snmp_output($result, \%ps_oid) };
2515 @output = @{ run_omreport("$omopt_chassis pwrsupplies") };
2525 6 => 'Uninterruptible Power Supply',
2535 1 => 'Presence detected',
2536 2 => 'Failure detected',
2537 4 => 'Predictive Failure',
2539 16 => 'AC lost or out-of-range',
2540 32 => 'AC out-of-range but present',
2541 64 => 'Configuration error',
2544 my %ps_config_error_type
2546 1 => 'Vendor mismatch',
2547 2 => 'Revision mismatch',
2548 3 => 'Processor missing',
2552 foreach my $out (@output) {
2554 @states = (); # contains states for the PS
2556 $index = $out->{powerSupplyIndex} - 1;
2557 $status = $snmp_status{$out->{powerSupplyStatus}};
2558 $type = $ps_type{$out->{powerSupplyType}};
2559 $err_type = defined $out->{powerSupplyConfigurationErrorType}
2560 ? $ps_config_error_type{$out->{powerSupplyConfigurationErrorType}} : undef;
2562 # get the combined state from the StatusReading OID
2563 foreach my $mask (sort keys %ps_state) {
2564 if (($out->{powerSupplySensorState} & $mask) != 0) {
2565 push @states, $ps_state{$mask};
2569 # If configuration error, also include the error type
2570 if (defined $err_type) {
2571 push @states, $err_type;
2574 # Finally, construct the state string
2575 $state = join q{, }, @states;
2578 $index = $out->{'Index'};
2579 $status = $out->{'Status'};
2580 $type = $out->{'Type'};
2581 $state = $out->{'Online Status'};
2584 next PS if blacklisted('ps', $index);
2587 if ($status ne 'Ok') {
2588 my $msg = sprintf 'Power Supply %d [%s] needs attention: %s',
2589 $index, $type, $state;
2590 report('chassis', $msg, $status2nagios{$status}, $index);
2593 my $msg = sprintf 'Power Supply %d [%s]: %s',
2594 $index, $type, $state;
2595 report('chassis', $msg, $E_OK, $index);
2602 #-----------------------------------------
2603 # CHASSIS: Check temperatures
2604 #-----------------------------------------
2605 sub check_temperatures {
2608 my $reading = undef;
2609 my $location = undef;
2610 my $max_crit = undef;
2611 my $max_warn = undef;
2612 my $min_warn = undef;
2613 my $min_crit = undef;
2615 my $discrete = undef;
2618 # Getting custom temperature thresholds (user option)
2619 my %warn_threshold = %{ custom_temperature_thresholds('w') };
2620 my %crit_threshold = %{ custom_temperature_thresholds('c') };
2625 '1.3.6.1.4.1.674.10892.1.700.20.1.2.1' => 'temperatureProbeIndex',
2626 '1.3.6.1.4.1.674.10892.1.700.20.1.5.1' => 'temperatureProbeStatus',
2627 '1.3.6.1.4.1.674.10892.1.700.20.1.6.1' => 'temperatureProbeReading',
2628 '1.3.6.1.4.1.674.10892.1.700.20.1.7.1' => 'temperatureProbeType',
2629 '1.3.6.1.4.1.674.10892.1.700.20.1.8.1' => 'temperatureProbeLocationName',
2630 '1.3.6.1.4.1.674.10892.1.700.20.1.10.1' => 'temperatureProbeUpperCriticalThreshold',
2631 '1.3.6.1.4.1.674.10892.1.700.20.1.11.1' => 'temperatureProbeUpperNonCriticalThreshold',
2632 '1.3.6.1.4.1.674.10892.1.700.20.1.12.1' => 'temperatureProbeLowerNonCriticalThreshold',
2633 '1.3.6.1.4.1.674.10892.1.700.20.1.13.1' => 'temperatureProbeLowerCriticalThreshold',
2634 '1.3.6.1.4.1.674.10892.1.700.20.1.16.1' => 'temperatureProbeDiscreteReading',
2636 # this didn't work well for some reason
2637 #my $result = $snmp_session->get_entries(-columns => [keys %temp_oid]);
2639 # Getting values using the table
2640 my $temperatureProbeTable = '1.3.6.1.4.1.674.10892.1.700.20';
2641 my $result = $snmp_session->get_table(-baseoid => $temperatureProbeTable);
2643 if (!defined $result) {
2644 printf "SNMP ERROR [temperatures]: %s.\n", $snmp_session->error;
2645 $snmp_session->close;
2649 @output = @{ get_snmp_output($result, \%temp_oid) };
2652 @output = @{ run_omreport("$omopt_chassis temps") };
2657 1 => 'Other', # type is other than following values
2658 2 => 'Unknown', # type is unknown
2659 3 => 'AmbientESM', # type is Ambient Embedded Systems Management temperature probe
2660 16 => 'Discrete', # type is temperature probe with discrete reading
2664 foreach my $out (@output) {
2666 $index = $out->{temperatureProbeIndex} - 1;
2667 $status = $snmp_probestatus{$out->{temperatureProbeStatus}};
2668 $reading = $out->{temperatureProbeReading} / 10;
2669 $location = $out->{temperatureProbeLocationName};
2670 $max_crit = $out->{temperatureProbeUpperCriticalThreshold} / 10;
2671 $max_warn = $out->{temperatureProbeUpperNonCriticalThreshold} / 10;
2672 $min_crit = exists $out->{temperatureProbeLowerCriticalThreshold}
2673 ? $out->{temperatureProbeLowerCriticalThreshold} / 10 : '[N/A]';
2674 $min_warn = exists $out->{temperatureProbeLowerNonCriticalThreshold}
2675 ? $out->{temperatureProbeLowerNonCriticalThreshold} / 10 : '[N/A]';
2676 $type = $probe_type{$out->{temperatureProbeType}};
2677 $discrete = exists $out->{temperatureProbeDiscreteReading}
2678 ? $out->{temperatureProbeDiscreteReading} : undef;
2681 $index = $out->{'Index'};
2682 $status = $out->{'Status'};
2683 $reading = $out->{'Reading'}; $reading =~ s{\.0\s+C}{}xms;
2684 $location = $out->{'Probe Name'};
2685 $max_crit = $out->{'Maximum Failure Threshold'}; $max_crit =~ s{\.0\s+C}{}xms;
2686 $max_warn = $out->{'Maximum Warning Threshold'}; $max_warn =~ s{\.0\s+C}{}xms;
2687 $min_crit = $out->{'Minimum Failure Threshold'}; $min_crit =~ s{\.0\s+C}{}xms;
2688 $min_warn = $out->{'Minimum Warning Threshold'}; $min_warn =~ s{\.0\s+C}{}xms;
2689 $type = $reading =~ m{\A\d+\z}xms ? 'AmbientESM' : 'Discrete';
2690 $discrete = $reading;
2693 next TEMP if blacklisted('temp', $index);
2696 if ($type eq 'Discrete') {
2697 my $msg = sprintf 'Temperature probe %d (%s): is %s',
2698 $index, $location, $discrete;
2699 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2700 report('chassis', $msg, $err, $index);
2703 # First check according to custom thresholds
2704 if (exists $crit_threshold{$index}{max} and $reading > $crit_threshold{$index}{max}) {
2705 # Custom critical MAX
2706 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom max=%d)',
2707 $index, $location, $reading, $crit_threshold{$index}{max};
2708 report('chassis', $msg, $E_CRITICAL, $index);
2710 elsif (exists $warn_threshold{$index}{max} and $reading > $warn_threshold{$index}{max}) {
2711 # Custom warning MAX
2712 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom max=%d)',
2713 $index, $location, $reading, $warn_threshold{$index}{max};
2714 report('chassis', $msg, $E_WARNING, $index);
2716 elsif (exists $crit_threshold{$index}{min} and $reading < $crit_threshold{$index}{min}) {
2717 # Custom critical MIN
2718 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom min=%d)',
2719 $index, $location, $reading, $crit_threshold{$index}{min};
2720 report('chassis', $msg, $E_CRITICAL, $index);
2722 elsif (exists $warn_threshold{$index}{min} and $reading < $warn_threshold{$index}{min}) {
2723 # Custom warning MIN
2724 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom min=%d)',
2725 $index, $location, $reading, $warn_threshold{$index}{min};
2726 report('chassis', $msg, $E_WARNING, $index);
2728 elsif ($status ne 'Ok' and $max_crit ne '[N/A]' and $reading > $max_crit) {
2729 my $msg = sprintf 'Temperature Probe %d [%s] is critically high at %d C',
2730 $index, $location, $reading;
2731 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2732 report('chassis', $msg, $err, $index);
2734 elsif ($status ne 'Ok' and $max_warn ne '[N/A]' and $reading > $max_warn) {
2735 my $msg = sprintf 'Temperature Probe %d [%s] is too high at %d C',
2736 $index, $location, $reading;
2737 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2738 report('chassis', $msg, $err, $index);
2740 elsif ($status ne 'Ok' and $min_crit ne '[N/A]' and $reading < $min_crit) {
2741 my $msg = sprintf 'Temperature Probe %d [%s] is critically low at %d C',
2742 $index, $location, $reading;
2743 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2744 report('chassis', $msg, $err, $index);
2746 elsif ($status ne 'Ok' and $min_warn ne '[N/A]' and $reading < $min_warn) {
2747 my $msg = sprintf 'Temperature Probe %d [%s] is too low at %d C',
2748 $index, $location, $reading;
2749 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2750 report('chassis', $msg, $err, $index);
2754 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C',
2755 $index, $location, $reading;
2756 if ($min_warn eq '[N/A]' and $min_crit eq '[N/A]') {
2757 $msg .= sprintf ' (max=%s/%s)', $max_warn, $max_crit;
2760 $msg .= sprintf ' (min=%s/%s, max=%s/%s)',
2761 $min_warn, $min_crit, $max_warn, $max_crit;
2763 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2764 report('chassis', $msg, $err, $index);
2767 # Collect performance data
2768 if (defined $opt{perfdata}) {
2769 my $pname = lc $location;
2770 $pname =~ s{\s}{_}gxms;
2771 $pname =~ s{_temp\z}{}xms;
2772 $pname =~ s{proc_}{cpu#}xms;
2773 my $pkey = join q{_}, 'temp', $index, $pname;
2774 my $pval = join q{;}, "${reading}C", $max_warn, $max_crit;
2775 $perfdata{$pkey} = $pval;
2783 #-----------------------------------------
2784 # CHASSIS: Check processors
2785 #-----------------------------------------
2786 sub check_processors {
2798 # NOTE: For some reason, older models don't have the
2799 # "Processor Device Status" OIDs. We check both the newer
2800 # (preferred) OIDs and the old ones.
2804 '1.3.6.1.4.1.674.10892.1.1100.30.1.2.1' => 'processorDeviceIndex',
2805 '1.3.6.1.4.1.674.10892.1.1100.30.1.5.1' => 'processorDeviceStatus',
2806 '1.3.6.1.4.1.674.10892.1.1100.30.1.8.1' => 'processorDeviceManufacturerName',
2807 '1.3.6.1.4.1.674.10892.1.1100.30.1.9.1' => 'processorDeviceStatusState',
2808 '1.3.6.1.4.1.674.10892.1.1100.30.1.10.1' => 'processorDeviceFamily',
2809 '1.3.6.1.4.1.674.10892.1.1100.30.1.12.1' => 'processorDeviceCurrentSpeed',
2810 '1.3.6.1.4.1.674.10892.1.1100.30.1.23.1' => 'processorDeviceBrandName',
2811 '1.3.6.1.4.1.674.10892.1.1100.32.1.2.1' => 'processorDeviceStatusIndex',
2812 '1.3.6.1.4.1.674.10892.1.1100.32.1.5.1' => 'processorDeviceStatusStatus',
2813 '1.3.6.1.4.1.674.10892.1.1100.32.1.6.1' => 'processorDeviceStatusReading',
2816 my $result = $snmp_session->get_entries(-columns => [keys %cpu_oid]);
2818 if (!defined $result) {
2819 printf "SNMP ERROR [processors]: %s.\n", $snmp_session->error;
2820 $snmp_session->close;
2824 @output = @{ get_snmp_output($result, \%cpu_oid) };
2827 @output = @{ run_omreport("$omopt_chassis processors") };
2832 1 => 'Other', # other than following values
2833 2 => 'Unknown', # unknown
2834 3 => 'Enabled', # enabled
2835 4 => 'User Disabled', # disabled by user via BIOS setup
2836 5 => 'BIOS Disabled', # disabled by BIOS (POST error)
2842 1 => 'Internal Error', # Internal Error
2843 2 => 'Thermal Trip', # Thermal Trip
2844 32 => 'Configuration Error', # Configuration Error
2845 128 => 'Present', # Processor Present
2846 256 => 'Disabled', # Processor Disabled
2847 512 => 'Terminator Present', # Terminator Present
2848 1024 => 'Throttled', # Processor Throttled
2851 # Mapping between family numbers from SNMP and actual CPU family
2854 1 => 'Other', 2 => 'Unknown', 3 => '8086',
2855 4 => '80286', 5 => '386', 6 => '486',
2856 7 => '8087', 8 => '80287', 9 => '80387',
2857 10 => '80487', 11 => 'Pentium', 12 => 'Pentium Pro',
2858 13 => 'Pentium II', 14 => 'Pentium with MMX', 15 => 'Celeron',
2859 16 => 'Pentium II Xeon', 17 => 'Pentium III', 18 => 'Pentium III Xeon',
2860 19 => 'Pentium III', 20 => 'Itanium', 21 => 'Xeon',
2861 22 => 'Pentium 4', 23 => 'Xeon MP', 24 => 'Itanium 2',
2862 25 => 'K5', 26 => 'K6', 27 => 'K6-2',
2863 28 => 'K6-3', 29 => 'Athlon', 30 => 'AMD2900',
2864 31 => 'K6-2+', 32 => 'Power PC', 33 => 'Power PC 601',
2865 34 => 'Power PC 603', 35 => 'Power PC 603+', 36 => 'Power PC 604',
2866 37 => 'Power PC 620', 38 => 'Power PC x704', 39 => 'Power PC 750',
2867 48 => 'Alpha', 49 => 'Alpha 21064', 50 => 'Alpha 21066',
2868 51 => 'Alpha 21164', 52 => 'Alpha 21164PC', 53 => 'Alpha 21164a',
2869 54 => 'Alpha 21264', 55 => 'Alpha 21364', 64 => 'MIPS',
2870 65 => 'MIPS R4000', 66 => 'MIPS R4200', 67 => 'MIPS R4400',
2871 68 => 'MIPS R4600', 69 => 'MIPS R10000', 80 => 'SPARC',
2872 81 => 'SuperSPARC', 82 => 'microSPARC II', 83 => 'microSPARC IIep',
2873 84 => 'UltraSPARC', 85 => 'UltraSPARC II', 86 => 'UltraSPARC IIi',
2874 87 => 'UltraSPARC III', 88 => 'UltraSPARC IIIi', 96 => '68040',
2875 97 => '68xxx', 98 => '68000', 99 => '68010',
2876 100 => '68020', 101 => '68030', 112 => 'Hobbit',
2877 120 => 'Crusoe TM5000', 121 => 'Crusoe TM3000', 122 => 'Efficeon TM8000',
2878 128 => 'Weitek', 131 => 'Athlon 64', 132 => 'Opteron',
2879 133 => 'Sempron', 134 => 'Turion 64 Mobile', 135 => 'Dual-Core Opteron',
2880 136 => 'Athlon 64 X2 DC', 137 => 'Turion 64 X2 M', 138 => 'Quad-Core Opteron',
2881 139 => '3rd gen Opteron', 144 => 'PA-RISC', 145 => 'PA-RISC 8500',
2882 146 => 'PA-RISC 8000', 147 => 'PA-RISC 7300LC', 148 => 'PA-RISC 7200',
2883 149 => 'PA-RISC 7100LC', 150 => 'PA-RISC 7100', 160 => 'V30',
2884 171 => 'Dual-Core Xeon 5200', 172 => 'Dual-Core Xeon 7200', 173 => 'Quad-Core Xeon 7300',
2885 174 => 'Quad-Core Xeon 7400', 175 => 'Multi-Core Xeon 7400', 176 => 'M1',
2886 177 => 'M2', 180 => 'AS400', 182 => 'Athlon XP',
2887 183 => 'Athlon MP', 184 => 'Duron', 185 => 'Pentium M',
2888 186 => 'Celeron D', 187 => 'Pentium D', 188 => 'Pentium Extreme',
2889 189 => 'Core Solo', 190 => 'Core2', 191 => 'Core2 Duo',
2890 198 => 'Core i7', 199 => 'Dual-Core Celeron', 200 => 'IBM390',
2891 201 => 'G4', 202 => 'G5', 203 => 'ESA/390 G6',
2892 204 => 'z/Architectur', 210 => 'C7-M', 211 => 'C7-D',
2893 212 => 'C7', 213 => 'Eden', 214 => 'Multi-Core Xeon',
2894 215 => 'Dual-Core Xeon 3xxx', 216 => 'Quad-Core Xeon 3xxx', 218 => 'Dual-Core Xeon 5xxx',
2895 219 => 'Quad-Core Xeon 5xxx', 221 => 'Dual-Core Xeon 7xxx', 222 => 'Quad-Core Xeon 7xxx',
2896 223 => 'Multi-Core Xeon 7xxx', 250 => 'i860', 251 => 'i960',
2900 foreach my $out (@output) {
2902 $index = exists $out->{processorDeviceStatusIndex}
2903 ? $out->{processorDeviceStatusIndex} - 1
2904 : $out->{processorDeviceIndex} - 1;
2905 $status = exists $out->{processorDeviceStatusStatus}
2906 ? $snmp_status{$out->{processorDeviceStatusStatus}}
2907 : $snmp_status{$out->{processorDeviceStatus}};
2908 if (exists $out->{processorDeviceStatusReading}) {
2909 my @states = (); # contains states for the CPU
2911 # get the combined state from the StatusReading OID
2912 foreach my $mask (sort keys %cpu_reading) {
2913 if (($out->{processorDeviceStatusReading} & $mask) != 0) {
2914 push @states, $cpu_reading{$mask};
2918 # Finally, create the state string
2919 $state = join q{, }, @states;
2922 $state = $cpu_state{$out->{processorDeviceStatusState}};
2924 $man = $out->{processorDeviceManufacturerName};
2925 $family = (exists $out->{processorDeviceFamily}
2926 and exists $cpu_family{$out->{processorDeviceFamily}})
2927 ? $cpu_family{$out->{processorDeviceFamily}} : undef;
2928 $speed = $out->{processorDeviceCurrentSpeed};
2929 $brand = $out->{processorDeviceBrandName};
2932 $index = $out->{'Index'};
2933 $status = $out->{'Status'};
2934 $state = $out->{'State'};
2935 $brand = exists $out->{'Processor Brand'} ? $out->{'Processor Brand'} : undef;
2936 $family = exists $out->{'Processor Family'} ? $out->{'Processor Family'} : undef;
2937 $man = exists $out->{'Processor Manufacturer'} ? $out->{'Processor Manufacturer'} : undef;
2938 $speed = exists $out->{'Current Speed'} ? $out->{'Current Speed'} : undef;
2941 next CPU if blacklisted('cpu', $index);
2943 # Ignore unoccupied CPU slots (omreport)
2944 next CPU if (defined $out->{'Processor Manufacturer'}
2945 and $out->{'Processor Manufacturer'} eq '[Not Occupied]')
2946 or (defined $out->{'Processor Brand'} and $out->{'Processor Brand'} eq '[Not Occupied]');
2948 # Ignore unoccupied CPU slots (snmp)
2949 if ($snmp and exists $out->{processorDeviceStatusReading}
2950 and $out->{processorDeviceStatusReading} == 0) {
2956 if (defined $brand) {
2957 $brand =~ s{\s\s+}{ }gxms;
2958 $brand =~ s{\((R|tm)\)}{}gxms;
2959 $brand =~ s{\s(CPU|Processor)}{}xms;
2960 $brand =~ s{\s\@}{}xms;
2962 elsif (defined $family and defined $man and defined $speed) {
2963 $speed =~ s{\A (\d+) .*}{$1}xms;
2964 $brand = sprintf '%s %s %.2fGHz', $man, $family, $speed / 1000;
2971 if ($status ne 'Ok') {
2972 my $msg = sprintf 'Processor %d [%s] needs attention: %s',
2973 $index, $brand, $state;
2974 report('chassis', $msg, $status2nagios{$status}, $index);
2978 my $msg = sprintf 'Processor %d [%s] is %s',
2979 $index, $brand, $state;
2980 report('chassis', $msg, $E_OK, $index);
2987 #-----------------------------------------
2988 # CHASSIS: Check voltage probes
2989 #-----------------------------------------
2993 my $reading = undef;
2994 my $location = undef;
3000 '1.3.6.1.4.1.674.10892.1.600.20.1.2.1' => 'voltageProbeIndex',
3001 '1.3.6.1.4.1.674.10892.1.600.20.1.5.1' => 'voltageProbeStatus',
3002 '1.3.6.1.4.1.674.10892.1.600.20.1.6.1' => 'voltageProbeReading',
3003 '1.3.6.1.4.1.674.10892.1.600.20.1.8.1' => 'voltageProbeLocationName',
3004 '1.3.6.1.4.1.674.10892.1.600.20.1.16.1' => 'voltageProbeDiscreteReading',
3007 my $voltageProbeTable = '1.3.6.1.4.1.674.10892.1.600.20.1';
3008 my $result = $snmp_session->get_table(-baseoid => $voltageProbeTable);
3010 if (!defined $result) {
3011 printf "SNMP ERROR [voltage]: %s.\n", $snmp_session->error;
3012 $snmp_session->close;
3016 @output = @{ get_snmp_output($result, \%volt_oid) };
3019 @output = @{ run_omreport("$omopt_chassis volts") };
3022 my %volt_discrete_reading
3029 foreach my $out (@output) {
3031 $index = $out->{voltageProbeIndex} - 1;
3032 $status = $snmp_status{$out->{voltageProbeStatus}};
3033 $reading = exists $out->{voltageProbeReading}
3034 ? sprintf('%.3f V', $out->{voltageProbeReading}/1000)
3035 : $volt_discrete_reading{$out->{voltageProbeDiscreteReading}};
3036 $location = $out->{voltageProbeLocationName};
3039 $index = $out->{'Index'};
3040 $status = $out->{'Status'};
3041 $reading = $out->{'Reading'};
3042 $location = $out->{'Probe Name'};
3045 next VOLT if blacklisted('volt', $index);
3048 my $msg = sprintf 'Voltage sensor %d [%s] is %s',
3049 $index, $location, $reading;
3050 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
3051 report('chassis', $msg, $err, $index);
3057 #-----------------------------------------
3058 # CHASSIS: Check batteries
3059 #-----------------------------------------
3060 sub check_batteries {
3063 my $reading = undef;
3064 my $location = undef;
3070 '1.3.6.1.4.1.674.10892.1.600.50.1.2.1' => 'batteryIndex',
3071 '1.3.6.1.4.1.674.10892.1.600.50.1.5.1' => 'batteryStatus',
3072 '1.3.6.1.4.1.674.10892.1.600.50.1.6.1' => 'batteryReading',
3073 '1.3.6.1.4.1.674.10892.1.600.50.1.7.1' => 'batteryLocationName',
3075 my $result = $snmp_session->get_entries(-columns => [keys %bat_oid]);
3077 # No batteries is OK
3078 return 0 if !defined $result;
3080 @output = @{ get_snmp_output($result, \%bat_oid) };
3083 @output = @{ run_omreport("$omopt_chassis batteries") };
3088 1 => 'Predictive Failure',
3090 4 => 'Presence Detected',
3094 foreach my $out (@output) {
3096 $index = $out->{batteryIndex} - 1;
3097 $status = $snmp_status{$out->{batteryStatus}};
3098 $reading = $bat_reading{$out->{batteryReading}};
3099 $location = $out->{batteryLocationName};
3102 $index = $out->{'Index'};
3103 $status = $out->{'Status'};
3104 $reading = $out->{'Reading'};
3105 $location = $out->{'Probe Name'};
3108 next BATTERY if blacklisted('bp', $index);
3111 my $msg = sprintf 'Battery probe %d [%s] is %s',
3112 $index, $location, $reading;
3113 report('chassis', $msg, $status2nagios{$status}, $index);
3119 #-----------------------------------------
3120 # CHASSIS: Check amperage probes (power monitoring)
3121 #-----------------------------------------
3122 sub check_pwrmonitoring {
3125 my $reading = undef;
3126 my $location = undef;
3127 my $max_crit = undef;
3128 my $max_warn = undef;
3135 '1.3.6.1.4.1.674.10892.1.600.30.1.2.1' => 'amperageProbeIndex',
3136 '1.3.6.1.4.1.674.10892.1.600.30.1.5.1' => 'amperageProbeStatus',
3137 '1.3.6.1.4.1.674.10892.1.600.30.1.6.1' => 'amperageProbeReading',
3138 '1.3.6.1.4.1.674.10892.1.600.30.1.7.1' => 'amperageProbeType',
3139 '1.3.6.1.4.1.674.10892.1.600.30.1.8.1' => 'amperageProbeLocationName',
3140 '1.3.6.1.4.1.674.10892.1.600.30.1.10.1' => 'amperageProbeUpperCriticalThreshold',
3141 '1.3.6.1.4.1.674.10892.1.600.30.1.11.1' => 'amperageProbeUpperNonCriticalThreshold',
3142 '1.3.6.1.4.1.674.10892.1.600.30.1.16.1' => 'amperageProbeDiscreteReading',
3144 my $result = $snmp_session->get_entries(-columns => [keys %amp_oid]);
3146 # No pwrmonitoring is OK
3147 return 0 if !defined $result;
3149 @output = @{ get_snmp_output($result, \%amp_oid) };
3152 @output = @{ run_omreport("$omopt_chassis pwrmonitoring") };
3155 my %amp_type # Amperage probe types
3157 1 => 'amperageProbeTypeIsOther', # other than following values
3158 2 => 'amperageProbeTypeIsUnknown', # unknown
3159 3 => 'amperageProbeTypeIs1Point5Volt', # 1.5 amperage probe
3160 4 => 'amperageProbeTypeIs3Point3volt', # 3.3 amperage probe
3161 5 => 'amperageProbeTypeIs5Volt', # 5 amperage probe
3162 6 => 'amperageProbeTypeIsMinus5Volt', # -5 amperage probe
3163 7 => 'amperageProbeTypeIs12Volt', # 12 amperage probe
3164 8 => 'amperageProbeTypeIsMinus12Volt', # -12 amperage probe
3165 9 => 'amperageProbeTypeIsIO', # I/O probe
3166 10 => 'amperageProbeTypeIsCore', # Core probe
3167 11 => 'amperageProbeTypeIsFLEA', # FLEA (standby) probe
3168 12 => 'amperageProbeTypeIsBattery', # Battery probe
3169 13 => 'amperageProbeTypeIsTerminator', # SCSI Termination probe
3170 14 => 'amperageProbeTypeIs2Point5Volt', # 2.5 amperage probe
3171 15 => 'amperageProbeTypeIsGTL', # GTL (ground termination logic) probe
3172 16 => 'amperageProbeTypeIsDiscrete', # amperage probe with discrete reading
3173 23 => 'amperageProbeTypeIsPowerSupplyAmps', # Power Supply probe with reading in Amps
3174 24 => 'amperageProbeTypeIsPowerSupplyWatts', # Power Supply probe with reading in Watts
3175 25 => 'amperageProbeTypeIsSystemAmps', # System probe with reading in Amps
3176 26 => 'amperageProbeTypeIsSystemWatts', # System probe with reading in Watts
3187 'amperageProbeTypeIsPowerSupplyAmps' => 'hA', # tenths of Amps
3188 'amperageProbeTypeIsSystemAmps' => 'hA', # tenths of Amps
3189 'amperageProbeTypeIsPowerSupplyWatts' => 'W', # Watts
3190 'amperageProbeTypeIsSystemWatts' => 'W', # Watts
3191 'amperageProbeTypeIsDiscrete' => q{}, # discrete reading, no unit
3195 foreach my $out (@output) {
3197 $index = $out->{amperageProbeIndex} - 1;
3198 $status = $snmp_status{$out->{amperageProbeStatus}};
3199 $reading = $amp_type{$out->{amperageProbeType}} eq 'amperageProbeTypeIsDiscrete'
3200 ? $amp_discrete{$out->{amperageProbeDiscreteReading}}
3201 : $out->{amperageProbeReading};
3202 $location = $out->{amperageProbeLocationName};
3203 $max_crit = exists $out->{amperageProbeUpperCriticalThreshold}
3204 ? $out->{amperageProbeUpperCriticalThreshold} : 0;
3205 $max_warn = exists $out->{amperageProbeUpperNonCriticalThreshold}
3206 ? $out->{amperageProbeUpperNonCriticalThreshold} : 0;
3207 $unit = exists $amp_unit{$amp_type{$out->{amperageProbeType}}}
3208 ? $amp_unit{$amp_type{$out->{amperageProbeType}}} : 'mA';
3209 if ($unit eq 'hA') {
3217 $index = $out->{'Index'};
3218 next AMP if (!defined $index || $index !~ m/^\d+$/x);
3219 $status = $out->{'Status'};
3220 $reading = $out->{'Reading'};
3221 $location = $out->{'Probe Name'};
3222 $max_crit = $out->{'Failure Threshold'} ne '[N/A]'
3223 ? $out->{'Failure Threshold'} : 0;
3224 $max_warn = $out->{'Warning Threshold'} ne '[N/A]'
3225 ? $out->{'Warning Threshold'} : 0;
3226 $reading =~ s{\A (\d+.*?)\s+([a-zA-Z]+) \s*\z}{$1}xms;
3228 $max_warn =~ s{\A (\d+.*?)\s+[a-zA-Z]+ \s*\z}{$1}xms;
3229 $max_crit =~ s{\A (\d+.*?)\s+[a-zA-Z]+ \s*\z}{$1}xms;
3232 next AMP if blacklisted('amp', $index);
3233 next AMP if $index !~ m{\A \d+ \z}xms;
3236 my $msg = sprintf 'Amperage probe %d [%s] reads %s %s',
3237 $index, $location, $reading, $unit, $status;
3238 report('chassis', $msg, $status2nagios{$status}, $index);
3240 # Collect performance data
3241 if (defined $opt{perfdata}) {
3242 next AMP if $reading !~ m{\A \d+(\.\d+)? \z}xms; # discrete reading (not number)
3243 my $pname = lc $location;
3244 $pname =~ s{\s}{_}gxms;
3245 my $pkey = join q{_}, 'pwr_mon', $index, $pname;
3246 my $pval = join q{;}, "$reading$unit", $max_warn, $max_crit;
3247 $perfdata{$pkey} = $pval;
3251 # Collect EXTRA performance data not found at first run. This is a
3253 if (defined $opt{perfdata} && !$snmp) {
3259 foreach (keys %perfdata) {
3260 if (m/\A pwr_mon_(\d+)/xms) {
3266 foreach my $line (@{ run_command("$omreport $omopt_chassis pwrmonitoring -fmt ssv") }) {
3268 if ($line eq 'Location;Reading') {
3276 if ($found and $line =~ m/\A ([^;]+?) ; (\d*\.\d+) \s ([AW]) \z/xms) {
3280 $aname =~ s{\s}{_}gxms;
3282 # don't use an existing index
3283 while (exists $used{$index}) { ++$index; }
3285 $perfdata{"pwr_mon_${index}_${aname}"} = "$aval$aunit;0;0";
3295 #-----------------------------------------
3296 # CHASSIS: Check intrusion
3297 #-----------------------------------------
3298 sub check_intrusion {
3301 my $reading = undef;
3307 '1.3.6.1.4.1.674.10892.1.300.70.1.2.1' => 'intrusionIndex',
3308 '1.3.6.1.4.1.674.10892.1.300.70.1.5.1' => 'intrusionStatus',
3309 '1.3.6.1.4.1.674.10892.1.300.70.1.6.1' => 'intrusionReading',
3311 my $result = $snmp_session->get_entries(-columns => [keys %int_oid]);
3313 # No intrusion is OK
3314 return 0 if !defined $result;
3316 @output = @{ get_snmp_output($result, \%int_oid) };
3319 @output = @{ run_omreport("$omopt_chassis intrusion") };
3324 1 => 'Not Breached', # chassis not breached and no uncleared breaches
3325 2 => 'Breached', # chassis currently breached
3326 3 => 'Breached Prior', # chassis breached prior to boot and has not been cleared
3327 4 => 'Breach Sensor Failure', # intrusion sensor has failed
3331 foreach my $out (@output) {
3333 $index = $out->{intrusionIndex} - 1;
3334 $status = $snmp_status{$out->{intrusionStatus}};
3335 $reading = $int_reading{$out->{intrusionReading}};
3338 $index = $out->{'Index'};
3339 $status = $out->{'Status'};
3340 $reading = $out->{'State'};
3343 next INTRUSION if blacklisted('intr', $index);
3346 if ($status ne 'Ok') {
3347 my $msg = sprintf 'Chassis intrusion %d detected: %s',
3349 report('chassis', $msg, $E_WARNING, $index);
3353 my $msg = sprintf 'Chassis intrusion %d detection: %s (%s)',
3354 $index, $status, $reading;
3355 report('chassis', $msg, $E_OK, $index);
3362 #-----------------------------------------
3363 # CHASSIS: Check alert log
3364 #-----------------------------------------
3365 sub check_alertlog {
3366 return if $snmp; # Not supported with SNMP
3368 my @output = @{ run_omreport("$omopt_system alertlog") };
3369 foreach my $out (@output) {
3370 ++$count{alert}{$out->{Severity}};
3373 # Create error messages and set exit value if appropriate
3375 if ($count{alert}{'Critical'} > 0) { $err = $E_CRITICAL; }
3376 elsif ($count{alert}{'Non-Critical'} > 0) { $err = $E_WARNING; }
3378 my $msg = sprintf 'Alert log content: %d critical, %d non-critical, %d ok',
3379 $count{alert}{'Critical'}, $count{alert}{'Non-Critical'}, $count{alert}{'Ok'};
3380 report('other', $msg, $err);
3385 #-----------------------------------------
3386 # CHASSIS: Check ESM log overall health
3387 #-----------------------------------------
3388 sub check_esmlog_health {
3392 my $systemStateEventLogStatus = '1.3.6.1.4.1.674.10892.1.200.10.1.41.1';
3393 my $result = $snmp_session->get_request(-varbindlist => [$systemStateEventLogStatus]);
3394 if (!defined $result) {
3395 my $msg = sprintf 'SNMP ERROR [esmhealth]: %s',
3396 $snmp_session->error;
3397 report('other', $msg, $E_UNKNOWN);
3399 $health = $snmp_status{$result->{$systemStateEventLogStatus}};
3402 foreach (@{ run_command("$omreport $omopt_system esmlog -fmt ssv") }) {
3403 if (m/\A Health;(.+) \z/xms) {
3411 # If the overall health of the ESM log is other than "Ok", the
3412 # fill grade of the log is more than 80% and the log should be
3414 if ($health eq 'Ok') {
3415 my $msg = sprintf 'ESM log health is Ok (less than 80%% full)';
3416 report('other', $msg, $E_OK);
3418 elsif ($health eq 'Critical') {
3419 my $msg = sprintf 'ESM log is 100%% full';
3420 report('other', $msg, $status2nagios{$health});
3423 my $msg = sprintf 'ESM log is more than 80%% full';
3424 report('other', $msg, $status2nagios{$health});
3430 #-----------------------------------------
3431 # CHASSIS: Check ESM log
3432 #-----------------------------------------
3439 '1.3.6.1.4.1.674.10892.1.300.40.1.7.1' => 'eventLogSeverityStatus',
3441 my $result = $snmp_session->get_entries(-columns => [keys %esm_oid]);
3444 return if !defined $result;
3446 @output = @{ get_snmp_output($result, \%esm_oid) };
3447 foreach my $out (@output) {
3448 ++$count{esm}{$snmp_status{$out->{eventLogSeverityStatus}}};
3452 @output = @{ run_omreport("$omopt_system esmlog") };
3453 foreach my $out (@output) {
3454 ++$count{esm}{$out->{Severity}};
3458 # Create error messages and set exit value if appropriate
3460 if ($count{esm}{'Critical'} > 0) { $err = $E_CRITICAL; }
3461 elsif ($count{esm}{'Non-Critical'} > 0) { $err = $E_WARNING; }
3463 my $msg = sprintf 'ESM log content: %d critical, %d non-critical, %d ok',
3464 $count{esm}{'Critical'}, $count{esm}{'Non-Critical'}, $count{esm}{'Ok'};
3465 report('other', $msg, $err);
3471 # Handy function for checking all storage components
3474 check_controllers();
3475 check_physical_disks();
3476 check_virtual_disks();
3477 check_cache_battery();
3480 check_enclosure_fans();
3481 check_enclosure_pwr();
3482 check_enclosure_temp();
3483 check_enclosure_emms();
3489 #---------------------------------------------------------------------
3491 #---------------------------------------------------------------------
3494 # Fetch output from 'omreport chassis info', put in sysinfo hash
3496 sub get_omreport_chassis_info {
3497 if (open my $INFO, '-|', "$omreport $omopt_chassis info -fmt ssv") {
3498 my @lines = <$INFO>;
3501 next if !m/\A (Chassis\sModel|Chassis\sService\sTag|Model|Service\sTag)/xms;
3502 my ($key, $val) = split /;/xms;
3503 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3504 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3505 if ($key eq 'Chassis Model' or $key eq 'Model') {
3506 $sysinfo{model} = $val;
3508 if ($key eq 'Chassis Service Tag' or $key eq 'Service Tag') {
3509 $sysinfo{serial} = $val;
3517 # Fetch output from 'omreport chassis bios', put in sysinfo hash
3519 sub get_omreport_chassis_bios {
3520 if (open my $BIOS, '-|', "$omreport $omopt_chassis bios -fmt ssv") {
3521 my @lines = <$BIOS>;
3525 my ($key, $val) = split /;/xms;
3526 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3527 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3528 $sysinfo{bios} = $val if $key eq 'Version';
3529 $sysinfo{biosdate} = $val if $key eq 'Release Date';
3536 # Fetch output from 'omreport system operatingsystem', put in sysinfo hash
3538 sub get_omreport_system_operatingsystem {
3539 if (open my $VER, '-|', "$omreport $omopt_system operatingsystem -fmt ssv") {
3544 my ($key, $val) = split /;/xms;
3545 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3546 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3547 if ($key eq 'Operating System') {
3548 $sysinfo{osname} = $val;
3550 elsif ($key eq 'Operating System Version') {
3551 $sysinfo{osver} = $val;
3559 # Fetch output from 'omreport about', put in sysinfo hash
3561 sub get_omreport_about {
3562 if (open my $OM, '-|', "$omreport about -fmt ssv") {
3566 if (m/\A Version;(.+) \z/xms) {
3576 # Fetch chassis info via SNMP, put in sysinfo hash
3578 sub get_snmp_chassis_info {
3581 '1.3.6.1.4.1.674.10892.1.300.10.1.9.1' => 'chassisModelName',
3582 '1.3.6.1.4.1.674.10892.1.300.10.1.11.1' => 'chassisServiceTagName',
3585 my $chassisInformationTable = '1.3.6.1.4.1.674.10892.1.300.10.1';
3586 my $result = $snmp_session->get_table(-baseoid => $chassisInformationTable);
3588 if (defined $result) {
3589 foreach my $oid (keys %{ $result }) {
3590 if (exists $chassis_oid{$oid} and $chassis_oid{$oid} eq 'chassisModelName') {
3591 $sysinfo{model} = $result->{$oid};
3592 $sysinfo{model} =~ s{\s+\z}{}xms; # remove trailing whitespace
3594 elsif (exists $chassis_oid{$oid} and $chassis_oid{$oid} eq 'chassisServiceTagName') {
3595 $sysinfo{serial} = $result->{$oid};
3600 my $msg = sprintf 'SNMP ERROR getting chassis info: %s',
3601 $snmp_session->error;
3602 report('other', $msg, $E_UNKNOWN);
3608 # Fetch BIOS info via SNMP, put in sysinfo hash
3610 sub get_snmp_chassis_bios {
3613 '1.3.6.1.4.1.674.10892.1.300.50.1.7.1.1' => 'systemBIOSReleaseDateName',
3614 '1.3.6.1.4.1.674.10892.1.300.50.1.8.1.1' => 'systemBIOSVersionName',
3617 my $systemBIOSTable = '1.3.6.1.4.1.674.10892.1.300.50.1';
3618 my $result = $snmp_session->get_table(-baseoid => $systemBIOSTable);
3620 if (defined $result) {
3621 foreach my $oid (keys %{ $result }) {
3622 if (exists $bios_oid{$oid} and $bios_oid{$oid} eq 'systemBIOSReleaseDateName') {
3623 $sysinfo{biosdate} = $result->{$oid};
3624 $sysinfo{biosdate} =~ s{\A (\d{4})(\d{2})(\d{2}).*}{$2/$3/$1}xms;
3626 elsif (exists $bios_oid{$oid} and $bios_oid{$oid} eq 'systemBIOSVersionName') {
3627 $sysinfo{bios} = $result->{$oid};
3632 my $msg = sprintf 'SNMP ERROR getting BIOS info: %s',
3633 $snmp_session->error;
3634 report('other', $msg, $E_UNKNOWN);
3640 # Fetch OS info via SNMP, put in sysinfo hash
3642 sub get_snmp_system_operatingsystem {
3645 '1.3.6.1.4.1.674.10892.1.400.10.1.6.1' => 'operatingSystemOperatingSystemName',
3646 '1.3.6.1.4.1.674.10892.1.400.10.1.7.1' => 'operatingSystemOperatingSystemVersionName',
3649 my $operatingSystemTable = '1.3.6.1.4.1.674.10892.1.400.10.1';
3650 my $result = $snmp_session->get_table(-baseoid => $operatingSystemTable);
3652 if (defined $result) {
3653 foreach my $oid (keys %{ $result }) {
3654 if (exists $os_oid{$oid} and $os_oid{$oid} eq 'operatingSystemOperatingSystemName') {
3655 $sysinfo{osname} = ($result->{$oid});
3657 elsif (exists $os_oid{$oid} and $os_oid{$oid} eq 'operatingSystemOperatingSystemVersionName') {
3658 $sysinfo{osver} = $result->{$oid};
3663 my $msg = sprintf 'SNMP ERROR getting OS info: %s',
3664 $snmp_session->error;
3665 report('other', $msg, $E_UNKNOWN);
3671 # Fetch OMSA version via SNMP, put in sysinfo hash
3673 sub get_snmp_about {
3676 '1.3.6.1.4.1.674.10892.1.100.10.0' => 'systemManagementSoftwareGlobalVersionName',
3678 my $systemManagementSoftwareGroup = '1.3.6.1.4.1.674.10892.1.100';
3679 my $result = $snmp_session->get_table(-baseoid => $systemManagementSoftwareGroup);
3680 if (defined $result) {
3681 foreach my $oid (keys %{ $result }) {
3682 if (exists $omsa_oid{$oid} and $omsa_oid{$oid} eq 'systemManagementSoftwareGlobalVersionName') {
3683 $sysinfo{om} = ($result->{$oid});
3688 my $msg = sprintf 'SNMP ERROR getting OMSA info: %s',
3689 $snmp_session->error;
3690 report('other', $msg, $E_UNKNOWN);
3696 # Collects some information about the system
3700 # Get system model and serial number
3701 $snmp ? get_snmp_chassis_info() : get_omreport_chassis_info();
3703 # Get BIOS information. Only if needed
3704 if ( $opt{okinfo} >= 1
3706 or (defined $opt{postmsg} and $opt{postmsg} =~ m/[%][bd]/xms) ) {
3707 $snmp ? get_snmp_chassis_bios() : get_omreport_chassis_bios();
3710 # Get OMSA information. Only if needed
3711 if ($opt{okinfo} >= 3 or $opt{debug}) {
3712 $snmp ? get_snmp_about() : get_omreport_about();
3715 # Return now if debug
3716 return if $opt{debug};
3718 # Get OS information. Only if needed
3719 if (defined $opt{postmsg} and $opt{postmsg} =~ m/[%][or]/xms) {
3720 $snmp ? get_snmp_system_operatingsystem() : get_omreport_system_operatingsystem();
3727 # Helper function for running omreport when the results are strictly
3729 sub run_omreport_info {
3730 my $command = shift;
3734 # Run omreport and fetch output
3735 my $rawtext = slurp_command("$omreport $command -fmt ssv 2>&1");
3737 # Parse output, store in array
3738 for ((split /\n/xms, $rawtext)) {
3739 if (m/\A Error/xms) {
3740 my $msg = "Problem running 'omreport $command': $_";
3741 report('other', $msg, $E_UNKNOWN);
3743 next if !m/;/xms; # ignore lines with less than two fields
3744 my @vals = split m/;/xms;
3745 $output{$vals[0]} = $vals[1];
3748 # Finally, return the collected information
3752 # Get various firmware information (BMC, RAC)
3753 sub get_firmware_info {
3754 my @snmp_output = ();
3755 my %nrpe_output = ();
3760 '1.3.6.1.4.1.674.10892.1.300.60.1.7.1' => 'firmwareType',
3761 '1.3.6.1.4.1.674.10892.1.300.60.1.8.1' => 'firmwareTypeName',
3762 '1.3.6.1.4.1.674.10892.1.300.60.1.11.1' => 'firmwareVersionName',
3765 my $firmwareTable = '1.3.6.1.4.1.674.10892.1.300.60.1';
3766 my $result = $snmp_session->get_table(-baseoid => $firmwareTable);
3768 # Some don't have this OID, this is ok
3769 if (!defined $result) {
3773 @snmp_output = @{ get_snmp_output($result, \%fw_oid) };
3776 %nrpe_output = %{ run_omreport_info("$omopt_chassis info") };
3779 my %fw_type # Firmware types
3781 1 => 'other', # other than following values
3782 2 => 'unknown', # unknown
3783 3 => 'systemBIOS', # System BIOS
3784 4 => 'embeddedSystemManagementController', # Embedded System Management Controller
3785 5 => 'powerSupplyParallelingBoard', # Power Supply Paralleling Board
3786 6 => 'systemBackPlane', # System (Primary) Backplane
3787 7 => 'powerVault2XXSKernel', # PowerVault 2XXS Kernel
3788 8 => 'powerVault2XXSApplication', # PowerVault 2XXS Application
3789 9 => 'frontPanel', # Front Panel Controller
3790 10 => 'baseboardManagementController', # Baseboard Management Controller
3791 11 => 'hotPlugPCI', # Hot Plug PCI Controller
3792 12 => 'sensorData', # Sensor Data Records
3793 13 => 'peripheralBay', # Peripheral Bay Backplane
3794 14 => 'secondaryBackPlane', # Secondary Backplane for ESM 2 systems
3795 15 => 'secondaryBackPlaneESM3And4', # Secondary Backplane for ESM 3 and 4 systems
3796 16 => 'rac', # Remote Access Controller
3797 17 => 'imc' # Integrated Management Controller
3802 foreach my $out (@snmp_output) {
3803 if ($fw_type{$out->{firmwareType}} eq 'baseboardManagementController') {
3804 $sysinfo{'bmc'} = 1;
3805 $sysinfo{'bmc_fw'} = $out->{firmwareVersionName};
3807 elsif ($fw_type{$out->{firmwareType}} =~ m{\A rac|imc \z}xms) {
3808 my $name = $out->{firmwareTypeName}; $name =~ s/\s//gxms;
3809 $sysinfo{'rac'} = 1;
3810 $sysinfo{'rac_name'} = $name;
3811 $sysinfo{'rac_fw'} = $out->{firmwareVersionName};
3816 foreach my $key (keys %nrpe_output) {
3817 next if !defined $nrpe_output{$key};
3818 if ($key eq 'BMC Version' or $key eq 'Baseboard Management Controller Version') {
3819 $sysinfo{'bmc'} = 1;
3820 $sysinfo{'bmc_fw'} = $nrpe_output{$key};
3822 elsif ($key =~ m{\A (i?DRAC)\s*(\d?)\s+Version}xms) {
3824 $sysinfo{'rac'} = 1;
3825 $sysinfo{'rac_fw'} = $nrpe_output{$key};
3826 $sysinfo{'rac_name'} = $name;
3836 #=====================================================================
3838 #=====================================================================
3840 # Here we do the actual checking of components
3841 # Check global status if applicable
3843 $globalstatus = check_global();
3846 # Do multiple selected checks
3847 if ($check{storage}) { check_storage(); }
3848 if ($check{memory}) { check_memory(); }
3849 if ($check{fans}) { check_fans(); }
3850 if ($check{power}) { check_powersupplies(); }
3851 if ($check{temp}) { check_temperatures(); }
3852 if ($check{cpu}) { check_processors(); }
3853 if ($check{voltage}) { check_volts(); }
3854 if ($check{batteries}) { check_batteries(); }
3855 if ($check{amperage}) { check_pwrmonitoring(); }
3856 if ($check{intrusion}) { check_intrusion(); }
3857 if ($check{alertlog}) { check_alertlog(); }
3858 if ($check{esmlog}) { check_esmlog(); }
3859 if ($check{esmhealth}) { check_esmlog_health(); }
3862 #---------------------------------------------------------------------
3864 #---------------------------------------------------------------------
3875 # Get system information
3878 # Get firmware info if requested via option
3879 if ($opt{okinfo} >= 1) {
3880 get_firmware_info();
3883 # Close SNMP session
3885 $snmp_session->close;
3890 print " System: $sysinfo{model}\n";
3891 print " ServiceTag: $sysinfo{serial}";
3892 print q{ } x (25 - length $sysinfo{serial}), "OMSA version: $sysinfo{om}\n";
3893 print " BIOS/date: $sysinfo{bios} $sysinfo{biosdate}";
3894 print q{ } x (25 - length "$sysinfo{bios} $sysinfo{biosdate}"), "Plugin version: $VERSION\n";
3895 if ($#report_storage >= 0) {
3896 print "-----------------------------------------------------------------------------\n";
3897 print " Storage Components \n";
3898 print "=============================================================================\n";
3899 print " STATE | ID | MESSAGE TEXT \n";
3900 print "---------+----------+--------------------------------------------------------\n";
3901 foreach (@report_storage) {
3902 my ($msg, $level, $nexus) = @{$_};
3903 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | "
3904 . q{ } x (8 - length $nexus) . "$nexus | $msg\n";
3905 $nagios_alert_count{$reverse_exitcode{$level}}++;
3908 if ($#report_chassis >= 0) {
3909 print "-----------------------------------------------------------------------------\n";
3910 print " Chassis Components \n";
3911 print "=============================================================================\n";
3912 print " STATE | ID | MESSAGE TEXT \n";
3913 print "---------+------+------------------------------------------------------------\n";
3914 foreach (@report_chassis) {
3915 my ($msg, $level, $nexus) = @{$_};
3916 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | "
3917 . q{ } x (4 - length $nexus) . "$nexus | $msg\n";
3918 $nagios_alert_count{$reverse_exitcode{$level}}++;
3921 if ($#report_other >= 0) {
3922 print "-----------------------------------------------------------------------------\n";
3923 print " Other messages \n";
3924 print "=============================================================================\n";
3925 print " STATE | MESSAGE TEXT \n";
3926 print "---------+-------------------------------------------------------------------\n";
3927 foreach (@report_other) {
3928 my ($msg, $level, $nexus) = @{$_};
3929 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | $msg\n";
3930 $nagios_alert_count{$reverse_exitcode{$level}}++;
3935 my $c = 0; # counter to determine linebreaks
3937 # Run through each message, sorted by severity level
3939 foreach (sort {$a->[1] < $b->[1]} (@report_storage, @report_chassis, @report_other)) {
3940 my ($msg, $level, $nexus) = @{ $_ };
3941 next ALERT if $level == $E_OK;
3943 if (defined $opt{only}) {
3944 # If user wants only critical alerts
3945 next ALERT if ($opt{only} eq 'critical' and $level == $E_WARNING);
3947 # If user wants only warning alerts
3948 next ALERT if ($opt{only} eq 'warning' and $level == $E_CRITICAL);
3951 # Prefix with service tag if specified with option '-i|--info'
3953 if (defined $opt{htmlinfo}) {
3954 $msg = '[<a href="' . warranty_url($sysinfo{serial})
3955 . "\">$sysinfo{serial}</a>] " . $msg;
3958 $msg = "[$sysinfo{serial}] " . $msg;
3962 # Prefix with nagios level if specified with option '--state'
3963 $msg = $reverse_exitcode{$level} . ": $msg" if $opt{state};
3965 # Prefix with one-letter nagios level if specified with option '--short-state'
3966 $msg = (substr $reverse_exitcode{$level}, 0, 1) . ": $msg" if $opt{shortstate};
3968 ($c++ == 0) ? print $msg : print $linebreak, $msg;
3970 $nagios_alert_count{$reverse_exitcode{$level}}++;
3974 # Determine our exit code
3976 $exit_code = $E_UNKNOWN if $nagios_alert_count{'UNKNOWN'} > 0;
3977 $exit_code = $E_WARNING if $nagios_alert_count{'WARNING'} > 0;
3978 $exit_code = $E_CRITICAL if $nagios_alert_count{'CRITICAL'} > 0;
3980 # Global status via SNMP.. extra safety check
3981 if ($globalstatus != $E_OK && $exit_code == $E_OK && !defined $opt{only}) {
3982 print "OOPS! Something is wrong with this server, but I don't know what. ";
3983 print "The global system health status is $reverse_exitcode{$globalstatus}, ";
3984 print "but every component check is OK. This may be a bug in the Nagios plugin, ";
3985 print "please file a bug report.\n";
3990 if ($exit_code == $E_OK && defined $opt{only} && $opt{only} !~ m{\A critical|warning|chassis \z}xms && !$opt{debug}) {
3992 = ( 'storage' => "STORAGE OK - $count{pdisk} physical drives, $count{vdisk} logical drives",
3993 'fans' => $count{fan} == 0 && $blade ? 'OK - blade system with no fan probes' : "FANS OK - $count{fan} fan probes checked",
3994 'temp' => "TEMPERATURES OK - $count{temp} temperature probes checked",
3995 'memory' => "MEMORY OK - $count{dimm} memory modules checked",
3996 'power' => $count{power} == 0 ? 'OK - no instrumented power supplies found' : "POWER OK - $count{power} power supplies checked",
3997 'cpu' => "PROCESSORS OK - $count{cpu} processors checked",
3998 'voltage' => "VOLTAGE OK - $count{volt} voltage probes checked",
3999 'batteries' => $count{bat} == 0 ? 'OK - no batteries found' : "BATTERIES OK - $count{bat} batteries checked",
4000 'amperage' => $count{amp} == 0 ? 'OK - no power monitoring probes found' : "AMPERAGE OK - $count{amp} amperage (power monitoring) probes checked",
4001 'intrusion' => $count{intr} == 0 ? 'OK - no intrusion detection probes found' : "INTRUSION OK - $count{intr} intrusion detection probes checked",
4002 'alertlog' => $snmp ? 'OK - not supported via snmp' : "OK - Alert Log content: $count{alert}{Ok} ok, $count{alert}{'Non-Critical'} warning and $count{alert}{Critical} critical",
4003 'esmlog' => "OK - ESM Log content: $count{esm}{Ok} ok, $count{esm}{'Non-Critical'} warning and $count{esm}{Critical} critical",
4004 'esmhealth' => "ESM LOG OK - less than 80% used",
4007 print $okmsg{$opt{only}};
4009 elsif ($exit_code == $E_OK && !$opt{debug}) {
4010 if (defined $opt{htmlinfo}) {
4011 printf q{OK - System: '<a href="%s">%s</a>', SN: '<a href="%s">%s</a>', hardware working fine},
4012 documentation_url($sysinfo{model}), $sysinfo{model},
4013 warranty_url($sysinfo{serial}), $sysinfo{serial};
4016 printf q{OK - System: '%s', SN: '%s', hardware working fine},
4017 $sysinfo{model}, $sysinfo{serial};
4020 if ($check{storage}) {
4021 printf ', %d logical drives, %d physical drives',
4022 $count{vdisk}, $count{pdisk};
4025 print ', not checking storage';
4028 if ($opt{okinfo} >= 1) {
4030 printf q{----- BIOS='%s %s'}, $sysinfo{bios}, $sysinfo{biosdate};
4032 if ($sysinfo{rac}) {
4033 printf q{, %s='%s'}, $sysinfo{rac_name}, $sysinfo{rac_fw};
4035 if ($sysinfo{bmc}) {
4036 printf q{, BMC='%s'}, $sysinfo{bmc_fw};
4040 if ($opt{okinfo} >= 2) {
4041 if ($check{storage}) {
4042 my @storageprint = ();
4043 foreach my $id (sort keys %{ $sysinfo{controller} }) {
4044 chomp $sysinfo{controller}{$id}{driver};
4045 push @storageprint, sprintf q{----- CTRL %s (%s): FW='%s', DR='%s'},
4046 $sysinfo{controller}{$id}{id}, $sysinfo{controller}{$id}{name},
4047 $sysinfo{controller}{$id}{firmware}, $sysinfo{controller}{$id}{driver};
4049 foreach my $id (sort keys %{ $sysinfo{enclosure} }) {
4050 push @storageprint, sprintf q{----- ENCL %s (%s): FW='%s'},
4051 $sysinfo{enclosure}{$id}->{id}, $sysinfo{enclosure}{$id}->{name},
4052 $sysinfo{enclosure}{$id}->{firmware};
4056 foreach my $line (@storageprint) {
4057 print $linebreak, $line;
4062 if ($opt{okinfo} >= 3) {
4063 print "$linebreak----- OpenManage Server Administrator (OMSA) version: '$sysinfo{om}'";
4068 if ($opt{extinfo}) {
4070 if (defined $opt{htmlinfo}) {
4071 printf '------ SYSTEM: <a href="%s">%s</a>, SN: <a href="%s">%s</a>',
4072 documentation_url($sysinfo{model}), $sysinfo{model},
4073 warranty_url($sysinfo{serial}), $sysinfo{serial};
4076 printf '------ SYSTEM: %s, SN: %s',
4077 $sysinfo{model}, $sysinfo{serial};
4080 if (defined $opt{postmsg}) {
4082 if (-f $opt{postmsg}) {
4083 open my $POST, '<', $opt{postmsg}
4084 or ( print $linebreak
4085 and print "ERROR: Couldn't open post message file $opt{postmsg}: $!\n"
4086 and exit $E_UNKNOWN );
4092 $post = $opt{postmsg};
4094 if (defined $post) {
4096 $post =~ s{[%]s}{$sysinfo{serial}}gxms;
4097 $post =~ s{[%]m}{$sysinfo{model}}gxms;
4098 $post =~ s{[%]b}{$sysinfo{bios}}gxms;
4099 $post =~ s{[%]d}{$sysinfo{biosdate}}gxms;
4100 $post =~ s{[%]o}{$sysinfo{osname}}gxms;
4101 $post =~ s{[%]r}{$sysinfo{osver}}gxms;
4102 $post =~ s{[%]p}{$count{pdisk}}gxms;
4103 $post =~ s{[%]l}{$count{vdisk}}gxms;
4104 $post =~ s{[%]n}{$linebreak}gxms;
4105 $post =~ s{[%]{2}}{%}gxms;
4111 # Print any perl warnings that have occured
4112 if (@perl_warnings) {
4113 foreach (@perl_warnings) {
4115 print "${linebreak}INTERNAL ERROR: @$_";
4117 $exit_code = $E_UNKNOWN;
4120 # Reset the WARN signal
4121 $SIG{__WARN__} = $original_sigwarn;
4123 # Print performance data
4124 if (defined $opt{perfdata} && !$opt{debug} && %perfdata) {
4125 my $lb = $opt{perfdata} eq 'multiline' ? "\n" : q{ }; # line break for perfdata
4136 return ($order{(split /_/, $a, 2)[0]} cmp $order{(split /_/, $b, 2)[0]}) || $a cmp $b;
4139 print join $lb, map { "'$_'=$perfdata{$_}" } sort perfdata keys %perfdata;
4142 # Print a linebreak at the end
4143 print "\n" if !$opt{debug};
4145 # Exit with proper exit code