5 # Monitor Dell server hardware status using Dell OpenManage Server
6 # Administrator, either locally via NRPE, or remotely via SNMP.
10 # Copyright (C) 2010 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 subroutine 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-beta6';
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) 2010 $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;
1110 my $minstdr = undef; # Minimum required Storport driver version (whats this?)
1111 my $stdr = undef; # Storport driver version (whats this?)
1117 '1.3.6.1.4.1.674.10893.1.20.130.1.1.1' => 'controllerNumber',
1118 '1.3.6.1.4.1.674.10893.1.20.130.1.1.2' => 'controllerName',
1119 '1.3.6.1.4.1.674.10893.1.20.130.1.1.5' => 'controllerState',
1120 '1.3.6.1.4.1.674.10893.1.20.130.1.1.8' => 'controllerFWVersion',
1121 '1.3.6.1.4.1.674.10893.1.20.130.1.1.38' => 'controllerComponentStatus',
1122 '1.3.6.1.4.1.674.10893.1.20.130.1.1.39' => 'controllerNexusID',
1123 '1.3.6.1.4.1.674.10893.1.20.130.1.1.41' => 'controllerDriverVersion',
1124 '1.3.6.1.4.1.674.10893.1.20.130.1.1.44' => 'controllerMinFWVersion',
1125 '1.3.6.1.4.1.674.10893.1.20.130.1.1.45' => 'controllerMinDriverVersion',
1126 '1.3.6.1.4.1.674.10893.1.20.130.1.1.55' => 'FIXME_StorportDriverVersion',
1127 '1.3.6.1.4.1.674.10893.1.20.130.1.1.56' => 'FIXME_StorportMinDriverVersion',
1130 # We use get_table() here for the odd case where a server has
1131 # two or more controllers, and where some OIDs are missing on
1132 # one of the controllers.
1133 my $controllerTable = '1.3.6.1.4.1.674.10893.1.20.130.1';
1134 my $result = $snmp_session->get_table(-baseoid => $controllerTable);
1136 # No controllers is OK
1137 return if !defined $result;
1139 @output = @{ get_snmp_output($result, \%ctrl_oid) };
1142 @output = @{ run_omreport('storage controller') };
1156 foreach my $out (@output) {
1158 $id = $out->{controllerNumber} - 1;
1159 $name = $out->{controllerName};
1160 $state = $ctrl_state{$out->{controllerState}};
1161 $status = $snmp_status{$out->{controllerComponentStatus}};
1162 $minfw = exists $out->{controllerMinFWVersion}
1163 ? $out->{controllerMinFWVersion} : undef;
1164 $mindr = exists $out->{controllerMinDriverVersion}
1165 ? $out->{controllerMinDriverVersion} : undef;
1166 $firmware = exists $out->{controllerFWVersion}
1167 ? $out->{controllerFWVersion} : 'N/A';
1168 $driver = exists $out->{controllerDriverVersion}
1169 ? $out->{controllerDriverVersion} : 'N/A';
1170 $minstdr = exists $out->{'FIXME_StorportMinDriverVersion'}
1171 ? $out->{FIXME_StorportMinDriverVersion} : undef;
1172 $stdr = exists $out->{FIXME_StorportDriverVersion}
1173 ? $out->{FIXME_StorportDriverVersion} : 'N/A';
1174 $nexus = convert_nexus($out->{controllerNexusID});
1178 $name = $out->{Name};
1179 $state = $out->{State};
1180 $status = $out->{Status};
1181 $minfw = $out->{'Minimum Required Firmware Version'} ne 'Not Applicable'
1182 ? $out->{'Minimum Required Firmware Version'} : undef;
1183 $mindr = $out->{'Minimum Required Driver Version'} ne 'Not Applicable'
1184 ? $out->{'Minimum Required Driver Version'} : undef;
1185 $firmware = $out->{'Firmware Version'} ne 'Not Applicable'
1186 ? $out->{'Firmware Version'} : 'N/A';
1187 $driver = $out->{'Driver Version'} ne 'Not Applicable'
1188 ? $out->{'Driver Version'} : 'N/A';
1189 $minstdr = (exists $out->{'Minimum Required Storport Driver Version'}
1190 and $out->{'Minimum Required Storport Driver Version'} ne 'Not Applicable')
1191 ? $out->{'Minimum Required Storport Driver Version'} : undef;
1192 $stdr = (exists $out->{'Storport Driver Version'}
1193 and $out->{'Storport Driver Version'} ne 'Not Applicable')
1194 ? $out->{'Storport Driver Version'} : 'N/A';
1198 $name =~ s{\s+\z}{}xms; # remove trailing whitespace
1199 push @controllers, $id;
1201 # Collecting some storage info
1202 $sysinfo{'controller'}{$id}{'id'} = $nexus;
1203 $sysinfo{'controller'}{$id}{'name'} = $name;
1204 $sysinfo{'controller'}{$id}{'driver'} = $driver;
1205 $sysinfo{'controller'}{$id}{'firmware'} = $firmware;
1207 next CTRL if blacklisted('ctrl', $nexus);
1209 # Special case: old firmware
1210 if (!blacklisted('ctrl_fw', $id) && defined $minfw) {
1212 my $msg = sprintf q{Controller %d [%s]: Firmware '%s' is out of date},
1213 $id, $name, $firmware;
1214 report('storage', $msg, $E_WARNING, $nexus);
1216 # Special case: old driver
1217 if (!blacklisted('ctrl_driver', $id) && defined $mindr) {
1219 my $msg = sprintf q{Controller %d [%s]: Driver '%s' is out of date},
1220 $id, $name, $driver;
1221 report('storage', $msg, $E_WARNING, $nexus);
1223 # Special case: old storport driver
1224 if (!blacklisted('ctrl_stdr', $id) && defined $minstdr) {
1226 my $msg = sprintf q{Controller %d [%s]: Storport driver '%s' is out of date},
1228 report('storage', $msg, $E_WARNING, $nexus);
1231 if ($status eq 'Ok' or ($status eq 'Non-Critical'
1232 and (defined $minfw or defined $mindr or defined $minstdr))) {
1233 my $msg = sprintf 'Controller %d [%s] is %s',
1235 report('storage', $msg, $E_OK, $nexus);
1239 my $msg = sprintf 'Controller %d [%s] needs attention: %s',
1241 report('storage', $msg, $status2nagios{$status}, $nexus);
1248 #-----------------------------------------
1249 # STORAGE: Check physical drives
1250 #-----------------------------------------
1251 sub check_physical_disks {
1252 return if $#controllers == -1;
1262 my $vendor = undef; # disk vendor
1263 my $product = undef; # product ID
1264 my $capacity = undef; # disk length (size) in bytes
1270 '1.3.6.1.4.1.674.10893.1.20.130.4.1.1' => 'arrayDiskNumber',
1271 '1.3.6.1.4.1.674.10893.1.20.130.4.1.2' => 'arrayDiskName',
1272 '1.3.6.1.4.1.674.10893.1.20.130.4.1.3' => 'arrayDiskVendor',
1273 '1.3.6.1.4.1.674.10893.1.20.130.4.1.4' => 'arrayDiskState',
1274 '1.3.6.1.4.1.674.10893.1.20.130.4.1.6' => 'arrayDiskProductID',
1275 '1.3.6.1.4.1.674.10893.1.20.130.4.1.9' => 'arrayDiskEnclosureID',
1276 '1.3.6.1.4.1.674.10893.1.20.130.4.1.10' => 'arrayDiskChannel',
1277 '1.3.6.1.4.1.674.10893.1.20.130.4.1.11' => 'arrayDiskLengthInMB',
1278 '1.3.6.1.4.1.674.10893.1.20.130.4.1.15' => 'arrayDiskTargetID',
1279 '1.3.6.1.4.1.674.10893.1.20.130.4.1.16' => 'arrayDiskLunID',
1280 '1.3.6.1.4.1.674.10893.1.20.130.4.1.24' => 'arrayDiskComponentStatus',
1281 '1.3.6.1.4.1.674.10893.1.20.130.4.1.26' => 'arrayDiskNexusID',
1282 '1.3.6.1.4.1.674.10893.1.20.130.4.1.31' => 'arrayDiskSmartAlertIndication',
1283 '1.3.6.1.4.1.674.10893.1.20.130.5.1.5' => 'arrayDiskEnclosureConnectionEnclosureNumber',
1284 '1.3.6.1.4.1.674.10893.1.20.130.5.1.7' => 'arrayDiskEnclosureConnectionControllerNumber',
1286 my $result = $snmp_session->get_entries(-columns => [keys %pdisk_oid]);
1288 if (!defined $result) {
1289 printf "SNMP ERROR [storage / pdisk]: %s.\n", $snmp_session->error;
1290 $snmp_session->close;
1294 @output = @{ get_snmp_output($result, \%pdisk_oid) };
1297 foreach my $c (@controllers) {
1298 push @output, @{ run_omreport("storage pdisk controller=$c") };
1299 map_item('ctrl', $c, \@output);
1317 28 => 'Diagnostics',
1318 34 => 'Predictive failure',
1319 35 => 'Initializing',
1322 41 => 'Unsupported',
1323 53 => 'Incompatible',
1326 # Check physical disks on each of the controllers
1328 foreach my $out (@output) {
1330 $name = $out->{arrayDiskName};
1331 if ($name =~ m{.*\d+:\d+:\d+\z}xms) {
1332 $id = join q{:}, ($out->{arrayDiskChannel}, $out->{arrayDiskEnclosureID},
1333 $out->{arrayDiskTargetID});
1336 $id = join q{:}, ($out->{arrayDiskChannel}, $out->{arrayDiskTargetID});
1338 $state = $pdisk_state{$out->{arrayDiskState}};
1339 $status = $snmp_status{$out->{arrayDiskComponentStatus}};
1340 $fpred = $out->{arrayDiskSmartAlertIndication} == 2 ? 1 : 0;
1342 $ctrl = exists $out->{arrayDiskEnclosureConnectionControllerNumber}
1343 ? $out->{arrayDiskEnclosureConnectionControllerNumber} - 1
1345 $nexus = convert_nexus($out->{arrayDiskNexusID});
1346 $vendor = $out->{arrayDiskVendor};
1347 $product = $out->{arrayDiskProductID};
1348 $capacity = $out->{arrayDiskLengthInMB} * 1024**2;
1352 $name = $out->{'Name'};
1353 $state = $out->{'State'};
1354 $status = $out->{'Status'};
1355 $fpred = lc($out->{'Failure Predicted'}) eq 'yes' ? 1 : 0;
1356 $progr = ' [' . $out->{'Progress'} . ']';
1357 $ctrl = $out->{'ctrl'};
1358 $nexus = join q{:}, $out->{ctrl}, $id;
1359 $vendor = $out->{'Vendor ID'};
1360 $product = $out->{'Product ID'};
1361 $capacity = $out->{'Capacity'};
1362 $capacity =~ s{\A .*? \((\d+) \s bytes\) \z}{$1}xms;
1365 next PDISK if blacklisted('pdisk', $nexus);
1368 $vendor =~ s{\s+\z}{}xms; # remove trailing whitespace
1369 $product =~ s{\s+\z}{}xms; # remove trailing whitespace
1371 # Calculate human readable capacity
1372 $capacity = ceil($capacity / 1000**3) >= 1000
1373 ? sprintf '%.1fTB', ($capacity / 1000**4)
1374 : sprintf '%.0fGB', ($capacity / 1000**3);
1375 $capacity = '450GB' if $capacity eq '449GB'; # quick fix for 450GB disks
1376 $capacity = '300GB' if $capacity eq '299GB'; # quick fix for 300GB disks
1377 $capacity = '146GB' if $capacity eq '147GB'; # quick fix for 146GB disks
1379 # Capitalize only the first letter of the vendor name
1380 $vendor = (substr $vendor, 0, 1) . lc (substr $vendor, 1, length $vendor);
1382 # Remove unnecessary trademark rubbish from vendor name
1383 $vendor =~ s{\(tm\)\z}{}xms;
1385 # Special case: Failure predicted
1386 if ($status eq 'Non-Critical' and $fpred) {
1387 my $msg = sprintf '%s [%s %s, %s] on ctrl %d needs attention: Failure Predicted',
1388 $name, $vendor, $product, $capacity, $ctrl;
1389 report('storage', $msg, $E_WARNING, $nexus);
1391 # Special case: Rebuilding
1392 elsif ($state eq 'Rebuilding') {
1393 my $msg = sprintf '%s [%s] on ctrl %d is %s%s',
1394 $name, $capacity, $ctrl, $state, $progr;
1395 report('storage', $msg, $E_WARNING, $nexus);
1398 elsif ($status ne 'Ok') {
1399 my $msg = sprintf '%s [%s %s, %s] on ctrl %d needs attention: %s',
1400 $name, $vendor, $product, $capacity, $ctrl, $state;
1401 report('storage', $msg, $status2nagios{$status}, $nexus);
1405 my $msg = sprintf '%s [%s] on ctrl %d is %s',
1406 $name, $capacity, $ctrl, $state;
1407 report('storage', $msg, $E_OK, $nexus);
1414 #-----------------------------------------
1415 # STORAGE: Check logical drives
1416 #-----------------------------------------
1417 sub check_virtual_disks {
1418 return if $#controllers == -1;
1435 '1.3.6.1.4.1.674.10893.1.20.140.1.1.3' => 'virtualDiskDeviceName',
1436 '1.3.6.1.4.1.674.10893.1.20.140.1.1.4' => 'virtualDiskState',
1437 '1.3.6.1.4.1.674.10893.1.20.140.1.1.6' => 'virtualDiskLengthInMB',
1438 '1.3.6.1.4.1.674.10893.1.20.140.1.1.13' => 'virtualDiskLayout',
1439 '1.3.6.1.4.1.674.10893.1.20.140.1.1.17' => 'virtualDiskTargetID',
1440 '1.3.6.1.4.1.674.10893.1.20.140.1.1.20' => 'virtualDiskComponentStatus',
1441 '1.3.6.1.4.1.674.10893.1.20.140.1.1.21' => 'virtualDiskNexusID',
1443 my $result = $snmp_session->get_entries(-columns => [keys %vdisk_oid]);
1445 # No logical drives is OK
1446 return if !defined $result;
1448 @output = @{ get_snmp_output($result, \%vdisk_oid) };
1451 foreach my $c (@controllers) {
1452 push @output, @{ run_omreport("storage vdisk controller=$c") };
1453 map_item('ctrl', $c, \@output);
1466 16 => 'Regenerating',
1469 32 => 'Reconstructing',
1470 35 => 'Initializing',
1471 36 => 'Background Initialization',
1472 38 => 'Resynching Paused',
1473 52 => 'Permanently Degraded',
1474 54 => 'Degraded Redundancy',
1479 1 => 'Concatenated',
1486 19 => 'Concatenated RAID 1',
1490 # Check virtual disks on each of the controllers
1492 foreach my $out (@output) {
1494 $id = $out->{virtualDiskTargetID};
1495 $dev = $out->{virtualDiskDeviceName};
1496 $state = $vdisk_state{$out->{virtualDiskState}};
1497 $status = $snmp_status{$out->{virtualDiskComponentStatus}};
1498 $layout = $vdisk_layout{$out->{virtualDiskLayout}};
1499 $size = sprintf '%.2f GB', $out->{virtualDiskLengthInMB} / 1024;
1500 $progr = q{}; # can't get this from SNMP(?)
1501 $nexus = convert_nexus($out->{virtualDiskNexusID});
1502 $ctrl = $nexus; # We use the nexus id to get the controller id
1503 $ctrl =~ s{\A (\d+):\d+ \z}{$1}xms;
1507 $dev = $out->{'Device Name'};
1508 $state = $out->{State};
1509 $status = $out->{Status};
1510 $layout = $out->{Layout};
1511 $size = $out->{Size};
1512 $progr = ' [' . $out->{Progress} . ']';
1513 $size =~ s{\A (.*GB).* \z}{$1}xms;
1514 $nexus = join q{:}, $out->{ctrl}, $id;
1515 $ctrl = $out->{ctrl};
1518 next VDISK if blacklisted('vdisk', $nexus);
1521 # The device name is undefined sometimes
1522 $dev = q{} if !defined $dev;
1524 # Special case: Regenerating
1525 if ($state eq 'Regenerating') {
1526 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d is %s%s},
1527 $id, $dev, $layout, $size, $ctrl, $state, $progr;
1528 report('storage', $msg, $E_WARNING, $nexus);
1531 elsif ($status ne 'Ok') {
1532 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d needs attention: %s},
1533 $id, $dev, $layout, $size, $ctrl, $state;
1534 report('storage', $msg, $status2nagios{$status}, $nexus);
1538 my $msg = sprintf q{Logical drive %d '%s' [%s, %s] on ctrl %d is %s},
1539 $id, $dev, $layout, $size, $ctrl, $state;
1540 report('storage', $msg, $E_OK, $nexus);
1547 #-----------------------------------------
1548 # STORAGE: Check cache batteries
1549 #-----------------------------------------
1550 sub check_cache_battery {
1551 return if $#controllers == -1;
1558 my $learn = undef; # learn state
1559 my $pred = undef; # battery's ability to be charged
1565 '1.3.6.1.4.1.674.10893.1.20.130.15.1.4' => 'batteryState',
1566 '1.3.6.1.4.1.674.10893.1.20.130.15.1.6' => 'batteryComponentStatus',
1567 '1.3.6.1.4.1.674.10893.1.20.130.15.1.9' => 'batteryNexusID',
1568 '1.3.6.1.4.1.674.10893.1.20.130.15.1.10' => 'batteryPredictedCapacity',
1569 '1.3.6.1.4.1.674.10893.1.20.130.15.1.12' => 'batteryLearnState',
1570 '1.3.6.1.4.1.674.10893.1.20.130.16.1.5' => 'batteryConnectionControllerNumber',
1572 my $result = $snmp_session->get_entries(-columns => [keys %bat_oid]);
1574 # No cache battery is OK
1575 return if !defined $result;
1577 @output = @{ get_snmp_output($result, \%bat_oid) };
1580 foreach my $c (@controllers) {
1581 push @output, @{ run_omreport("storage battery controller=$c") };
1582 map_item('ctrl', $c, \@output);
1592 7 => 'Reconditioning',
1600 # Specifies the learn state activity of the battery
1610 # This property displays the battery's ability to be charged
1613 1 => 'Failed', # The battery cannot be charged and needs to be replaced
1614 2 => 'Ready', # The battery can be charged to full capacity
1615 4 => 'Unknown', # The battery is completing a Learn cycle. The charge capacity of the
1616 # battery cannot be determined until the Learn cycle is complete
1619 # Check battery on each of the controllers
1621 foreach my $out (@output) {
1623 $state = $bat_state{$out->{batteryState}};
1624 $status = $snmp_status{$out->{batteryComponentStatus}};
1625 $learn = exists $out->{batteryLearnState}
1626 ? $bat_learn_state{$out->{batteryLearnState}} : undef;
1627 $pred = exists $out->{batteryPredictedCapacity}
1628 ? $bat_pred_cap{$out->{batteryPredictedCapacity}} : undef;
1629 $ctrl = $out->{batteryConnectionControllerNumber} - 1;
1630 $nexus = convert_nexus($out->{batteryNexusID});
1632 $id =~ s{\A \d+:(\d+) \z}{$1}xms;
1636 $state = $out->{'State'};
1637 $status = $out->{'Status'};
1638 $learn = $out->{'Learn State'};
1639 $pred = $out->{'Predicted Capacity Status'};
1640 $ctrl = $out->{'ctrl'};
1641 $nexus = join q{:}, $out->{ctrl}, $id;
1644 next BATTERY if blacklisted('bat', $nexus);
1646 # Special case: Charging
1647 if ($state eq 'Charging') {
1648 if ($pred eq 'Failed') {
1649 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [replace battery]',
1650 $id, $ctrl, $state, $pred;
1651 report('storage', $msg, $E_CRITICAL, $nexus);
1654 next BATTERY if blacklisted('bat_charge', $nexus);
1655 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1656 $id, $ctrl, $state, $pred;
1657 report('storage', $msg, $E_WARNING, $nexus);
1660 # Special case: Learning (battery learns its capacity)
1661 elsif ($state eq 'Learning') {
1662 if ($learn eq 'Failed') {
1663 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s)',
1664 $id, $ctrl, $state, $learn;
1665 report('storage', $msg, $E_CRITICAL, $nexus);
1668 next BATTERY if blacklisted('bat_charge', $nexus);
1669 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1670 $id, $ctrl, $state, $learn;
1671 report('storage', $msg, $E_WARNING, $nexus);
1674 # Special case: Power Low (first part of recharge cycle)
1675 elsif ($state eq 'Power Low') {
1676 next BATTERY if blacklisted('bat_charge', $nexus);
1677 my $msg = sprintf 'Cache battery %d in controller %d is %s [probably harmless]',
1679 report('storage', $msg, $E_WARNING, $nexus);
1681 # Special case: Degraded and Non-Critical (usually part of recharge cycle)
1682 elsif ($state eq 'Degraded' && $status eq 'Non-Critical') {
1683 next BATTERY if blacklisted('bat_charge', $nexus);
1684 my $msg = sprintf 'Cache battery %d in controller %d is %s (%s) [probably harmless]',
1685 $id, $ctrl, $state, $status;
1686 report('storage', $msg, $E_WARNING, $nexus);
1689 elsif ($status ne 'Ok') {
1690 my $msg = sprintf 'Cache battery %d in controller %d needs attention: %s (%s)',
1691 $id, $ctrl, $state, $status;
1692 report('storage', $msg, $status2nagios{$status}, $nexus);
1696 my $msg = sprintf 'Cache battery %d in controller %d is %s',
1698 report('storage', $msg, $E_OK, $nexus);
1705 #-----------------------------------------
1706 # STORAGE: Check connectors (channels)
1707 #-----------------------------------------
1708 sub check_connectors {
1709 return if $#controllers == -1;
1723 '1.3.6.1.4.1.674.10893.1.20.130.2.1.1' => 'channelNumber',
1724 '1.3.6.1.4.1.674.10893.1.20.130.2.1.2' => 'channelName',
1725 '1.3.6.1.4.1.674.10893.1.20.130.2.1.3' => 'channelState',
1726 '1.3.6.1.4.1.674.10893.1.20.130.2.1.8' => 'channelComponentStatus',
1727 '1.3.6.1.4.1.674.10893.1.20.130.2.1.9' => 'channelNexusID',
1728 '1.3.6.1.4.1.674.10893.1.20.130.2.1.11' => 'channelBusType',
1730 my $result = $snmp_session->get_entries(-columns => [keys %conn_oid]);
1732 if (!defined $result) {
1733 printf "SNMP ERROR [storage / channel]: %s.\n", $snmp_session->error;
1734 $snmp_session->close;
1738 @output = @{ get_snmp_output($result, \%conn_oid) };
1741 foreach my $c (@controllers) {
1742 push @output, @{ run_omreport("storage connector controller=$c") };
1743 map_item('ctrl', $c, \@output);
1761 3 => 'Fibre Channel',
1768 # Check connectors on each of the controllers
1770 foreach my $out (@output) {
1772 $id = $out->{channelNumber} - 1;
1773 $name = $out->{channelName};
1774 $state = $conn_state{$out->{channelState}};
1775 $status = $snmp_status{$out->{channelComponentStatus}};
1776 $type = $conn_bustype{$out->{channelBusType}};
1777 $nexus = convert_nexus($out->{channelNexusID});
1779 $ctrl =~ s{(\d+):\d+}{$1}xms;
1783 $name = $out->{'Name'};
1784 $state = $out->{'State'};
1785 $status = $out->{'Status'};
1786 $type = $out->{'Connector Type'};
1787 $ctrl = $out->{ctrl};
1788 $nexus = join q{:}, $out->{ctrl}, $id;
1791 next CHANNEL if blacklisted('conn', $nexus);
1793 my $msg = sprintf '%s [%s] on controller %d is %s',
1794 $name, $type, $ctrl, $state;
1795 report('storage', $msg, $status2nagios{$status}, $nexus);
1801 #-----------------------------------------
1802 # STORAGE: Check enclosures
1803 #-----------------------------------------
1804 sub check_enclosures {
1810 my $firmware = undef;
1817 '1.3.6.1.4.1.674.10893.1.20.130.3.1.1' => 'enclosureNumber',
1818 '1.3.6.1.4.1.674.10893.1.20.130.3.1.2' => 'enclosureName',
1819 '1.3.6.1.4.1.674.10893.1.20.130.3.1.4' => 'enclosureState',
1820 '1.3.6.1.4.1.674.10893.1.20.130.3.1.19' => 'enclosureChannelNumber',
1821 '1.3.6.1.4.1.674.10893.1.20.130.3.1.24' => 'enclosureComponentStatus',
1822 '1.3.6.1.4.1.674.10893.1.20.130.3.1.25' => 'enclosureNexusID',
1823 '1.3.6.1.4.1.674.10893.1.20.130.3.1.26' => 'enclosureFirmwareVersion',
1825 my $result = $snmp_session->get_entries(-columns => [keys %encl_oid]);
1827 # No enclosures is OK
1828 return if !defined $result;
1830 @output = @{ get_snmp_output($result, \%encl_oid) };
1833 foreach my $c (@controllers) {
1834 push @output, @{ run_omreport("storage enclosure controller=$c") };
1835 map_item('ctrl', $c, \@output);
1850 foreach my $out (@output) {
1852 $id = $out->{'enclosureNumber'} - 1;
1853 $name = $out->{'enclosureName'};
1854 $state = $encl_state{$out->{'enclosureState'}};
1855 $status = $snmp_status{$out->{'enclosureComponentStatus'}};
1856 $firmware = exists $out->{enclosureFirmwareVersion}
1857 ? $out->{enclosureFirmwareVersion} : 'N/A';
1858 $nexus = convert_nexus($out->{enclosureNexusID});
1860 $ctrl =~ s{\A (\d+):.* \z}{$1}xms;
1864 $name = $out->{Name};
1865 $state = $out->{State};
1866 $status = $out->{Status};
1867 $firmware = $out->{'Firmware Version'} ne 'Not Applicable'
1868 ? $out->{'Firmware Version'} : 'N/A';
1869 $nexus = join q{:}, $out->{ctrl}, $id;
1870 $ctrl = $out->{ctrl};
1873 $name =~ s{\s+\z}{}xms; # remove trailing whitespace
1874 $firmware =~ s{\s+\z}{}xms; # remove trailing whitespace
1876 # store enclosure data for future use
1877 push @enclosures, { 'id' => $id,
1878 'ctrl' => $out->{ctrl},
1881 # Collecting some storage info
1882 $sysinfo{'enclosure'}{$nexus}{'id'} = $nexus;
1883 $sysinfo{'enclosure'}{$nexus}{'name'} = $name;
1884 $sysinfo{'enclosure'}{$nexus}{'firmware'} = $firmware;
1886 next ENCLOSURE if blacklisted('encl', $nexus);
1888 my $msg = sprintf 'Enclosure %s [%s] on controller %d is %s',
1889 $nexus, $name, $ctrl, $state;
1890 report('storage', $msg, $status2nagios{$status}, $nexus);
1896 #-----------------------------------------
1897 # STORAGE: Check enclosure fans
1898 #-----------------------------------------
1899 sub check_enclosure_fans {
1900 return if $#controllers == -1;
1908 my $encl_id = undef;
1909 my $encl_name = undef;
1915 '1.3.6.1.4.1.674.10893.1.20.130.7.1.1' => 'fanNumber',
1916 '1.3.6.1.4.1.674.10893.1.20.130.7.1.2' => 'fanName',
1917 '1.3.6.1.4.1.674.10893.1.20.130.7.1.4' => 'fanState',
1918 '1.3.6.1.4.1.674.10893.1.20.130.7.1.11' => 'fanProbeCurrValue',
1919 '1.3.6.1.4.1.674.10893.1.20.130.7.1.15' => 'fanComponentStatus',
1920 '1.3.6.1.4.1.674.10893.1.20.130.7.1.16' => 'fanNexusID',
1921 '1.3.6.1.4.1.674.10893.1.20.130.8.1.4' => 'fanConnectionEnclosureName',
1922 '1.3.6.1.4.1.674.10893.1.20.130.8.1.5' => 'fanConnectionEnclosureNumber',
1925 my $result = $snmp_session->get_entries(-columns => [keys %fan_oid]);
1927 # No enclosure fans is OK
1928 return if !defined $result;
1930 @output = @{ get_snmp_output($result, \%fan_oid) };
1933 foreach my $enc (@enclosures) {
1934 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=fans") };
1935 map_item('ctrl', $enc->{ctrl}, \@output);
1936 map_item('encl_id', $enc->{id}, \@output);
1937 map_item('encl_name', $enc->{name}, \@output);
1952 # Check fans on each of the enclosures
1954 foreach my $out (@output) {
1956 $id = $out->{fanNumber} - 1;
1957 $name = $out->{fanName};
1958 $state = $fan_state{$out->{fanState}};
1959 $status = $snmp_status{$out->{fanComponentStatus}};
1960 $speed = $out->{fanProbeCurrValue};
1961 $encl_id = $out->{fanConnectionEnclosureNumber} - 1;
1962 $encl_name = $out->{fanConnectionEnclosureName};
1963 $nexus = convert_nexus($out->{fanNexusID});
1967 $name = $out->{'Name'};
1968 $state = $out->{'State'};
1969 $status = $out->{'Status'};
1970 $speed = $out->{'Speed'};
1971 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
1972 $encl_name = $out->{encl_name};
1973 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
1976 next FAN if blacklisted('encl_fan', $nexus);
1979 if ($status ne 'Ok') {
1980 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
1981 $name, $encl_id, $encl_name, $state;
1982 report('storage', $msg, $status2nagios{$status}, $nexus);
1986 my $msg = sprintf '%s in enclosure %s [%s] is %s (speed=%s)',
1987 $name, $encl_id, $encl_name, $state, $speed;
1988 report('storage', $msg, $E_OK, $nexus);
1995 #-----------------------------------------
1996 # STORAGE: Check enclosure power supplies
1997 #-----------------------------------------
1998 sub check_enclosure_pwr {
1999 return if $#controllers == -1;
2006 my $encl_id = undef;
2007 my $encl_name = undef;
2013 '1.3.6.1.4.1.674.10893.1.20.130.9.1.1' => 'powerSupplyNumber',
2014 '1.3.6.1.4.1.674.10893.1.20.130.9.1.2' => 'powerSupplyName',
2015 '1.3.6.1.4.1.674.10893.1.20.130.9.1.4' => 'powerSupplyState',
2016 '1.3.6.1.4.1.674.10893.1.20.130.9.1.9' => 'powerSupplyComponentStatus',
2017 '1.3.6.1.4.1.674.10893.1.20.130.9.1.10' => 'powerSupplyNexusID',
2018 '1.3.6.1.4.1.674.10893.1.20.130.10.1.4' => 'powerSupplyConnectionEnclosureName',
2019 '1.3.6.1.4.1.674.10893.1.20.130.10.1.5' => 'powerSupplyConnectionEnclosureNumber',
2021 my $result = $snmp_session->get_entries(-columns => [keys %ps_oid]);
2023 # No enclosure power supplies is OK
2024 return if !defined $result;
2026 @output = @{ get_snmp_output($result, \%ps_oid) };
2029 foreach my $enc (@enclosures) {
2030 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=pwrsupplies") };
2031 map_item('ctrl', $enc->{ctrl}, \@output);
2032 map_item('encl_id', $enc->{id}, \@output);
2033 map_item('encl_name', $enc->{name}, \@output);
2042 5 => 'Not Installed',
2048 # Check power supplies on each of the enclosures
2050 foreach my $out (@output) {
2052 $id = $out->{powerSupplyNumber};
2053 $name = $out->{powerSupplyName};
2054 $state = $ps_state{$out->{powerSupplyState}};
2055 $status = $snmp_status{$out->{powerSupplyComponentStatus}};
2056 $encl_id = $out->{powerSupplyConnectionEnclosureNumber} - 1;
2057 $encl_name = $out->{powerSupplyConnectionEnclosureName};
2058 $nexus = convert_nexus($out->{powerSupplyNexusID});
2062 $name = $out->{'Name'};
2063 $state = $out->{'State'};
2064 $status = $out->{'Status'};
2065 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2066 $encl_name = $out->{encl_name};
2067 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2070 next PS if blacklisted('encl_ps', $nexus);
2073 if ($status ne 'Ok') {
2074 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
2075 $name, $encl_id, $encl_name, $state;
2076 report('storage', $msg, $status2nagios{$status}, $nexus);
2080 my $msg = sprintf '%s in enclosure %s [%s] is %s',
2081 $name, $encl_id, $encl_name, $state;
2082 report('storage', $msg, $E_OK, $nexus);
2089 #-----------------------------------------
2090 # STORAGE: Check enclosure temperatures
2091 #-----------------------------------------
2092 sub check_enclosure_temp {
2093 return if $#controllers == -1;
2100 my $reading = undef;
2102 my $max_warn = undef;
2103 my $max_crit = undef;
2104 my $encl_id = undef;
2105 my $encl_name = undef;
2111 '1.3.6.1.4.1.674.10893.1.20.130.11.1.1' => 'temperatureProbeNumber',
2112 '1.3.6.1.4.1.674.10893.1.20.130.11.1.2' => 'temperatureProbeName',
2113 '1.3.6.1.4.1.674.10893.1.20.130.11.1.4' => 'temperatureProbeState',
2114 '1.3.6.1.4.1.674.10893.1.20.130.11.1.6' => 'temperatureProbeUnit',
2115 '1.3.6.1.4.1.674.10893.1.20.130.11.1.9' => 'temperatureProbeMaxWarning',
2116 '1.3.6.1.4.1.674.10893.1.20.130.11.1.10' => 'temperatureProbeMaxCritical',
2117 '1.3.6.1.4.1.674.10893.1.20.130.11.1.11' => 'temperatureProbeCurValue',
2118 '1.3.6.1.4.1.674.10893.1.20.130.11.1.13' => 'temperatureProbeComponentStatus',
2119 '1.3.6.1.4.1.674.10893.1.20.130.11.1.14' => 'temperatureProbeNexusID',
2120 '1.3.6.1.4.1.674.10893.1.20.130.12.1.4' => 'temperatureConnectionEnclosureName',
2121 '1.3.6.1.4.1.674.10893.1.20.130.12.1.5' => 'temperatureConnectionEnclosureNumber',
2123 my $result = $snmp_session->get_entries(-columns => [keys %temp_oid]);
2125 # No enclosure temperature probes is OK
2126 return if !defined $result;
2128 @output = @{ get_snmp_output($result, \%temp_oid) };
2131 foreach my $enc (@enclosures) {
2132 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=temps") };
2133 map_item('ctrl', $enc->{ctrl}, \@output);
2134 map_item('encl_id', $enc->{id}, \@output);
2135 map_item('encl_name', $enc->{name}, \@output);
2150 # Check temperature probes on each of the enclosures
2152 foreach my $out (@output) {
2154 $id = $out->{temperatureProbeNumber} - 1;
2155 $name = $out->{temperatureProbeName};
2156 $state = $temp_state{$out->{temperatureProbeState}};
2157 $status = $snmp_status{$out->{temperatureProbeComponentStatus}};
2158 $unit = $out->{temperatureProbeUnit};
2159 $reading = $out->{temperatureProbeCurValue};
2160 $max_warn = $out->{temperatureProbeMaxWarning};
2161 $max_crit = $out->{temperatureProbeMaxCritical};
2162 $encl_id = $out->{temperatureConnectionEnclosureNumber} - 1;
2163 $encl_name = $out->{temperatureConnectionEnclosureName};
2164 $nexus = convert_nexus($out->{temperatureProbeNexusID});
2168 $name = $out->{'Name'};
2169 $state = $out->{'State'};
2170 $status = $out->{'Status'};
2172 $reading = $out->{'Reading'}; $reading =~ s{\s*C}{}xms;
2173 $max_warn = $out->{'Maximum Warning Threshold'}; $max_warn =~ s{\s*C}{}xms;
2174 $max_crit = $out->{'Maximum Failure Threshold'}; $max_crit =~ s{\s*C}{}xms;
2175 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2176 $encl_name = $out->{encl_name};
2177 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2180 next TEMP if blacklisted('encl_temp', $nexus);
2183 if ($status ne 'Ok') {
2184 my $msg = sprintf '%s in enclosure %s [%s] is %s C at %s (%s max)',
2185 $name, $encl_id, $encl_name, $state, $reading, $max_crit;
2186 report('storage', $msg, $status2nagios{$status}, $nexus);
2190 my $msg = sprintf '%s in enclosure %s [%s]: %s C (%s max)',
2191 $name, $encl_id, $encl_name, $reading, $max_crit;
2192 report('storage', $msg, $E_OK, $nexus);
2195 # Collect performance data
2196 if (defined $opt{perfdata}) {
2197 $name =~ s{\A Temperature\sProbe\s(\d+) \z}{temp_$1}gxms;
2198 my $pkey = "enclosure_${encl_id}_${name}";
2199 my $pval = join q{;}, "${reading}C", $max_warn, $max_crit;
2200 $perfdata{$pkey} = $pval;
2207 #-----------------------------------------
2208 # STORAGE: Check enclosure management modules (EMM)
2209 #-----------------------------------------
2210 sub check_enclosure_emms {
2211 return if $#controllers == -1;
2218 my $encl_id = undef;
2219 my $encl_name = undef;
2225 '1.3.6.1.4.1.674.10893.1.20.130.13.1.1' => 'enclosureManagementModuleNumber',
2226 '1.3.6.1.4.1.674.10893.1.20.130.13.1.2' => 'enclosureManagementModuleName',
2227 '1.3.6.1.4.1.674.10893.1.20.130.13.1.4' => 'enclosureManagementModuleState',
2228 '1.3.6.1.4.1.674.10893.1.20.130.13.1.11' => 'enclosureManagementModuleComponentStatus',
2229 '1.3.6.1.4.1.674.10893.1.20.130.13.1.12' => 'enclosureManagementModuleNexusID',
2230 '1.3.6.1.4.1.674.10893.1.20.130.14.1.4' => 'enclosureManagementModuleConnectionEnclosureName',
2231 '1.3.6.1.4.1.674.10893.1.20.130.14.1.5' => 'enclosureManagementModuleConnectionEnclosureNumber',
2233 my $result = $snmp_session->get_entries(-columns => [keys %emms_oid]);
2235 # No enclosure EMMs is OK
2236 return if !defined $result;
2238 @output = @{ get_snmp_output($result, \%emms_oid) };
2241 foreach my $enc (@enclosures) {
2242 push @output, @{ run_omreport("storage enclosure controller=$enc->{ctrl} enclosure=$enc->{id} info=emms") };
2243 map_item('ctrl', $enc->{ctrl}, \@output);
2244 map_item('encl_id', $enc->{id}, \@output);
2245 map_item('encl_name', $enc->{name}, \@output);
2256 5 => 'Not Installed',
2261 # Check temperature probes on each of the enclosures
2263 foreach my $out (@output) {
2265 $id = $out->{enclosureManagementModuleNumber} - 1;
2266 $name = $out->{enclosureManagementModuleName};
2267 $state = $emms_state{$out->{enclosureManagementModuleState}};
2268 $status = $snmp_status{$out->{enclosureManagementModuleComponentStatus}};
2269 $encl_id = $out->{enclosureManagementModuleConnectionEnclosureNumber} - 1;
2270 $encl_name = $out->{enclosureManagementModuleConnectionEnclosureName};
2271 $nexus = convert_nexus($out->{enclosureManagementModuleNexusID});
2275 $name = $out->{'Name'};
2276 $state = $out->{'State'};
2277 $status = $out->{'Status'};
2278 $encl_id = join q{:}, $out->{ctrl}, $out->{'encl_id'};
2279 $encl_name = $out->{encl_name};
2280 $nexus = join q{:}, $out->{ctrl}, $out->{'encl_id'}, $id;
2283 next EMM if blacklisted('encl_emm', $nexus);
2286 if ($status ne 'Ok') {
2287 my $msg = sprintf '%s in enclosure %s [%s] needs attention: %s',
2288 $name, $encl_id, $encl_name, $state;
2289 report('storage', $msg, $status2nagios{$status}, $nexus);
2293 my $msg = sprintf '%s in enclosure %s [%s] is %s',
2294 $name, $encl_id, $encl_name, $state;
2295 report('storage', $msg, $E_OK, $nexus);
2302 #-----------------------------------------
2303 # CHASSIS: Check memory modules
2304 #-----------------------------------------
2308 my $location = undef;
2317 '1.3.6.1.4.1.674.10892.1.1100.50.1.2.1' => 'memoryDeviceIndex',
2318 '1.3.6.1.4.1.674.10892.1.1100.50.1.5.1' => 'memoryDeviceStatus',
2319 '1.3.6.1.4.1.674.10892.1.1100.50.1.8.1' => 'memoryDeviceLocationName',
2320 '1.3.6.1.4.1.674.10892.1.1100.50.1.14.1' => 'memoryDeviceSize',
2321 '1.3.6.1.4.1.674.10892.1.1100.50.1.20.1' => 'memoryDeviceFailureModes',
2323 my $result = $snmp_session->get_entries(-columns => [keys %dimm_oid]);
2325 if (!defined $result) {
2326 printf "SNMP ERROR [memory]: %s.\n", $snmp_session->error;
2327 $snmp_session->close;
2331 @output = @{ get_snmp_output($result, \%dimm_oid) };
2334 @output = @{ run_omreport("$omopt_chassis memory") };
2337 # Note: These values are bit masks, so combination values are
2338 # possible. If value is 0 (zero), memory device has no faults.
2341 1 => 'ECC single bit correction warning rate exceeded',
2342 2 => 'ECC single bit correction failure rate exceeded',
2343 4 => 'ECC multibit fault encountered',
2344 8 => 'ECC single bit correction logging disabled',
2345 16 => 'device disabled because of spare activation',
2349 foreach my $out (@output) {
2350 @failures = (); # Initialize
2352 $index = $out->{memoryDeviceIndex};
2353 $status = $snmp_status{$out->{memoryDeviceStatus}};
2354 $location = $out->{memoryDeviceLocationName};
2355 $size = sprintf '%d MB', $out->{memoryDeviceSize}/1024;
2356 $modes = $out->{memoryDeviceFailureModes};
2358 foreach my $mask (sort keys %failure_mode) {
2359 if (($modes & $mask) != 0) { push @failures, $failure_mode{$mask}; }
2364 $index = $out->{'Type'} eq '[Not Occupied]' ? undef : $out->{'Index'};
2365 $status = $out->{'Status'};
2366 $location = $out->{'Connector Name'};
2367 $size = $out->{'Size'};
2368 if (defined $size) {
2369 $size =~ s{\s\s}{ }gxms;
2371 # Run 'omreport chassis memory index=X' to get the failures
2372 if ($status ne 'Ok' && defined $index) {
2373 foreach (@{ run_command("$omreport $omopt_chassis memory index=$index -fmt ssv") }) {
2374 if (m/\A Failures; (.+?) \z/xms) {
2375 chop(my $fail = $1);
2376 push @failures, split m{\.}xms, $fail;
2381 $location =~ s{\A \s*(.*?)\s* \z}{$1}xms;
2383 next DIMM if blacklisted('dimm', $index);
2385 # Ignore empty memory slots
2386 next DIMM if !defined $index;
2389 if ($status ne 'Ok') {
2391 if (scalar @failures == 0) {
2392 $msg = sprintf 'Memory module %d [%s, %s] needs attention (%s)',
2393 $index, $location, $size, $status;
2396 $msg = sprintf 'Memory module %d [%s, %s] needs attention: %s',
2397 $index, $location, $size, (join q{, }, @failures);
2400 report('chassis', $msg, $status2nagios{$status}, $index);
2404 my $msg = sprintf 'Memory module %d [%s, %s] is %s',
2405 $index, $location, $size, $status;
2406 report('chassis', $msg, $E_OK, $index);
2413 #-----------------------------------------
2414 # CHASSIS: Check fans
2415 #-----------------------------------------
2419 my $reading = undef;
2420 my $location = undef;
2421 my $max_crit = undef;
2422 my $max_warn = undef;
2428 '1.3.6.1.4.1.674.10892.1.700.12.1.2.1' => 'coolingDeviceIndex',
2429 '1.3.6.1.4.1.674.10892.1.700.12.1.5.1' => 'coolingDeviceStatus',
2430 '1.3.6.1.4.1.674.10892.1.700.12.1.6.1' => 'coolingDeviceReading',
2431 '1.3.6.1.4.1.674.10892.1.700.12.1.8.1' => 'coolingDeviceLocationName',
2432 '1.3.6.1.4.1.674.10892.1.700.12.1.10.1' => 'coolingDeviceUpperCriticalThreshold',
2433 '1.3.6.1.4.1.674.10892.1.700.12.1.11.1' => 'coolingDeviceUpperNonCriticalThreshold',
2435 my $result = $snmp_session->get_entries(-columns => [keys %cool_oid]);
2437 if ($blade && !defined $result) {
2440 elsif (!$blade && !defined $result) {
2441 printf "SNMP ERROR [cooling]: %s.\n", $snmp_session->error;
2442 $snmp_session->close;
2446 @output = @{ get_snmp_output($result, \%cool_oid) };
2449 @output = @{ run_omreport("$omopt_chassis fans") };
2453 foreach my $out (@output) {
2455 $index = $out->{coolingDeviceIndex};
2456 $status = $snmp_probestatus{$out->{coolingDeviceStatus}};
2457 $reading = $out->{coolingDeviceReading};
2458 $location = $out->{coolingDeviceLocationName};
2459 $max_crit = exists $out->{coolingDeviceUpperCriticalThreshold}
2460 ? $out->{coolingDeviceUpperCriticalThreshold} : 0;
2461 $max_warn = exists $out->{coolingDeviceUpperNonCriticalThreshold}
2462 ? $out->{coolingDeviceUpperNonCriticalThreshold} : 0;
2465 $index = $out->{'Index'};
2466 $status = $out->{'Status'};
2467 $reading = $out->{'Reading'};
2468 $location = $out->{'Probe Name'};
2469 $max_crit = $out->{'Maximum Failure Threshold'} ne '[N/A]'
2470 ? $out->{'Maximum Failure Threshold'} : 0;
2471 $max_warn = $out->{'Maximum Warning Threshold'} ne '[N/A]'
2472 ? $out->{'Maximum Warning Threshold'} : 0;
2473 $reading =~ s{\A (\d+).* \z}{$1}xms;
2474 $max_warn =~ s{\A (\d+).* \z}{$1}xms;
2475 $max_crit =~ s{\A (\d+).* \z}{$1}xms;
2478 next FAN if blacklisted('fan', $index);
2481 if ($status ne 'Ok') {
2482 my $msg = sprintf 'Chassis fan %d [%s] needs attention: %s',
2483 $index, $location, $status;
2484 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2485 report('chassis', $msg, $err, $index);
2488 my $msg = sprintf 'Chassis fan %d [%s]: %s',
2489 $index, $location, $reading;
2490 report('chassis', $msg, $E_OK, $index);
2493 # Collect performance data
2494 if (defined $opt{perfdata}) {
2495 my $pname = lc $location;
2496 $pname =~ s{\s}{_}gxms;
2497 $pname =~ s{proc_}{cpu#}xms;
2498 my $pkey = join q{_}, 'fan', $index, $pname;
2499 my $pval = join q{;}, "${reading}RPM", $max_warn, $max_crit;
2500 $perfdata{$pkey} = $pval;
2507 #-----------------------------------------
2508 # CHASSIS: Check power supplies
2509 #-----------------------------------------
2510 sub check_powersupplies {
2514 my $err_type = undef;
2522 '1.3.6.1.4.1.674.10892.1.600.12.1.2.1' => 'powerSupplyIndex',
2523 '1.3.6.1.4.1.674.10892.1.600.12.1.5.1' => 'powerSupplyStatus',
2524 '1.3.6.1.4.1.674.10892.1.600.12.1.7.1' => 'powerSupplyType',
2525 '1.3.6.1.4.1.674.10892.1.600.12.1.11.1' => 'powerSupplySensorState',
2526 '1.3.6.1.4.1.674.10892.1.600.12.1.12.1' => 'powerSupplyConfigurationErrorType',
2528 my $result = $snmp_session->get_entries(-columns => [keys %ps_oid]);
2530 # No instrumented PSU is OK (blades, low-end servers)
2531 return 0 if !defined $result;
2533 @output = @{ get_snmp_output($result, \%ps_oid) };
2536 @output = @{ run_omreport("$omopt_chassis pwrsupplies") };
2546 6 => 'Uninterruptible Power Supply',
2556 1 => 'Presence detected',
2557 2 => 'Failure detected',
2558 4 => 'Predictive Failure',
2560 16 => 'AC lost or out-of-range',
2561 32 => 'AC out-of-range but present',
2562 64 => 'Configuration error',
2565 my %ps_config_error_type
2567 1 => 'Vendor mismatch',
2568 2 => 'Revision mismatch',
2569 3 => 'Processor missing',
2573 foreach my $out (@output) {
2575 @states = (); # contains states for the PS
2577 $index = $out->{powerSupplyIndex} - 1;
2578 $status = $snmp_status{$out->{powerSupplyStatus}};
2579 $type = $ps_type{$out->{powerSupplyType}};
2580 $err_type = defined $out->{powerSupplyConfigurationErrorType}
2581 ? $ps_config_error_type{$out->{powerSupplyConfigurationErrorType}} : undef;
2583 # get the combined state from the StatusReading OID
2584 foreach my $mask (sort keys %ps_state) {
2585 if (($out->{powerSupplySensorState} & $mask) != 0) {
2586 push @states, $ps_state{$mask};
2590 # If configuration error, also include the error type
2591 if (defined $err_type) {
2592 push @states, $err_type;
2595 # Finally, construct the state string
2596 $state = join q{, }, @states;
2599 $index = $out->{'Index'};
2600 $status = $out->{'Status'};
2601 $type = $out->{'Type'};
2602 $state = $out->{'Online Status'};
2605 next PS if blacklisted('ps', $index);
2608 if ($status ne 'Ok') {
2609 my $msg = sprintf 'Power Supply %d [%s] needs attention: %s',
2610 $index, $type, $state;
2611 report('chassis', $msg, $status2nagios{$status}, $index);
2614 my $msg = sprintf 'Power Supply %d [%s]: %s',
2615 $index, $type, $state;
2616 report('chassis', $msg, $E_OK, $index);
2623 #-----------------------------------------
2624 # CHASSIS: Check temperatures
2625 #-----------------------------------------
2626 sub check_temperatures {
2629 my $reading = undef;
2630 my $location = undef;
2631 my $max_crit = undef;
2632 my $max_warn = undef;
2633 my $min_warn = undef;
2634 my $min_crit = undef;
2636 my $discrete = undef;
2639 # Getting custom temperature thresholds (user option)
2640 my %warn_threshold = %{ custom_temperature_thresholds('w') };
2641 my %crit_threshold = %{ custom_temperature_thresholds('c') };
2646 '1.3.6.1.4.1.674.10892.1.700.20.1.2.1' => 'temperatureProbeIndex',
2647 '1.3.6.1.4.1.674.10892.1.700.20.1.5.1' => 'temperatureProbeStatus',
2648 '1.3.6.1.4.1.674.10892.1.700.20.1.6.1' => 'temperatureProbeReading',
2649 '1.3.6.1.4.1.674.10892.1.700.20.1.7.1' => 'temperatureProbeType',
2650 '1.3.6.1.4.1.674.10892.1.700.20.1.8.1' => 'temperatureProbeLocationName',
2651 '1.3.6.1.4.1.674.10892.1.700.20.1.10.1' => 'temperatureProbeUpperCriticalThreshold',
2652 '1.3.6.1.4.1.674.10892.1.700.20.1.11.1' => 'temperatureProbeUpperNonCriticalThreshold',
2653 '1.3.6.1.4.1.674.10892.1.700.20.1.12.1' => 'temperatureProbeLowerNonCriticalThreshold',
2654 '1.3.6.1.4.1.674.10892.1.700.20.1.13.1' => 'temperatureProbeLowerCriticalThreshold',
2655 '1.3.6.1.4.1.674.10892.1.700.20.1.16.1' => 'temperatureProbeDiscreteReading',
2657 # this didn't work well for some reason
2658 #my $result = $snmp_session->get_entries(-columns => [keys %temp_oid]);
2660 # Getting values using the table
2661 my $temperatureProbeTable = '1.3.6.1.4.1.674.10892.1.700.20';
2662 my $result = $snmp_session->get_table(-baseoid => $temperatureProbeTable);
2664 if (!defined $result) {
2665 printf "SNMP ERROR [temperatures]: %s.\n", $snmp_session->error;
2666 $snmp_session->close;
2670 @output = @{ get_snmp_output($result, \%temp_oid) };
2673 @output = @{ run_omreport("$omopt_chassis temps") };
2678 1 => 'Other', # type is other than following values
2679 2 => 'Unknown', # type is unknown
2680 3 => 'AmbientESM', # type is Ambient Embedded Systems Management temperature probe
2681 16 => 'Discrete', # type is temperature probe with discrete reading
2685 foreach my $out (@output) {
2687 $index = $out->{temperatureProbeIndex} - 1;
2688 $status = $snmp_probestatus{$out->{temperatureProbeStatus}};
2689 $reading = $out->{temperatureProbeReading} / 10;
2690 $location = $out->{temperatureProbeLocationName};
2691 $max_crit = $out->{temperatureProbeUpperCriticalThreshold} / 10;
2692 $max_warn = $out->{temperatureProbeUpperNonCriticalThreshold} / 10;
2693 $min_crit = exists $out->{temperatureProbeLowerCriticalThreshold}
2694 ? $out->{temperatureProbeLowerCriticalThreshold} / 10 : '[N/A]';
2695 $min_warn = exists $out->{temperatureProbeLowerNonCriticalThreshold}
2696 ? $out->{temperatureProbeLowerNonCriticalThreshold} / 10 : '[N/A]';
2697 $type = $probe_type{$out->{temperatureProbeType}};
2698 $discrete = exists $out->{temperatureProbeDiscreteReading}
2699 ? $out->{temperatureProbeDiscreteReading} : undef;
2702 $index = $out->{'Index'};
2703 $status = $out->{'Status'};
2704 $reading = $out->{'Reading'}; $reading =~ s{\.0\s+C}{}xms;
2705 $location = $out->{'Probe Name'};
2706 $max_crit = $out->{'Maximum Failure Threshold'}; $max_crit =~ s{\.0\s+C}{}xms;
2707 $max_warn = $out->{'Maximum Warning Threshold'}; $max_warn =~ s{\.0\s+C}{}xms;
2708 $min_crit = $out->{'Minimum Failure Threshold'}; $min_crit =~ s{\.0\s+C}{}xms;
2709 $min_warn = $out->{'Minimum Warning Threshold'}; $min_warn =~ s{\.0\s+C}{}xms;
2710 $type = $reading =~ m{\A\d+\z}xms ? 'AmbientESM' : 'Discrete';
2711 $discrete = $reading;
2714 next TEMP if blacklisted('temp', $index);
2717 if ($type eq 'Discrete') {
2718 my $msg = sprintf 'Temperature probe %d (%s): is %s',
2719 $index, $location, $discrete;
2720 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2721 report('chassis', $msg, $err, $index);
2724 # First check according to custom thresholds
2725 if (exists $crit_threshold{$index}{max} and $reading > $crit_threshold{$index}{max}) {
2726 # Custom critical MAX
2727 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom max=%d)',
2728 $index, $location, $reading, $crit_threshold{$index}{max};
2729 report('chassis', $msg, $E_CRITICAL, $index);
2731 elsif (exists $warn_threshold{$index}{max} and $reading > $warn_threshold{$index}{max}) {
2732 # Custom warning MAX
2733 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom max=%d)',
2734 $index, $location, $reading, $warn_threshold{$index}{max};
2735 report('chassis', $msg, $E_WARNING, $index);
2737 elsif (exists $crit_threshold{$index}{min} and $reading < $crit_threshold{$index}{min}) {
2738 # Custom critical MIN
2739 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom min=%d)',
2740 $index, $location, $reading, $crit_threshold{$index}{min};
2741 report('chassis', $msg, $E_CRITICAL, $index);
2743 elsif (exists $warn_threshold{$index}{min} and $reading < $warn_threshold{$index}{min}) {
2744 # Custom warning MIN
2745 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C (custom min=%d)',
2746 $index, $location, $reading, $warn_threshold{$index}{min};
2747 report('chassis', $msg, $E_WARNING, $index);
2749 elsif ($status ne 'Ok' and $max_crit ne '[N/A]' and $reading > $max_crit) {
2750 my $msg = sprintf 'Temperature Probe %d [%s] is critically high at %d C',
2751 $index, $location, $reading;
2752 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2753 report('chassis', $msg, $err, $index);
2755 elsif ($status ne 'Ok' and $max_warn ne '[N/A]' and $reading > $max_warn) {
2756 my $msg = sprintf 'Temperature Probe %d [%s] is too high at %d C',
2757 $index, $location, $reading;
2758 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2759 report('chassis', $msg, $err, $index);
2761 elsif ($status ne 'Ok' and $min_crit ne '[N/A]' and $reading < $min_crit) {
2762 my $msg = sprintf 'Temperature Probe %d [%s] is critically low at %d C',
2763 $index, $location, $reading;
2764 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2765 report('chassis', $msg, $err, $index);
2767 elsif ($status ne 'Ok' and $min_warn ne '[N/A]' and $reading < $min_warn) {
2768 my $msg = sprintf 'Temperature Probe %d [%s] is too low at %d C',
2769 $index, $location, $reading;
2770 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2771 report('chassis', $msg, $err, $index);
2775 my $msg = sprintf 'Temperature Probe %d [%s] reads %d C',
2776 $index, $location, $reading;
2777 if ($min_warn eq '[N/A]' and $min_crit eq '[N/A]') {
2778 $msg .= sprintf ' (max=%s/%s)', $max_warn, $max_crit;
2781 $msg .= sprintf ' (min=%s/%s, max=%s/%s)',
2782 $min_warn, $min_crit, $max_warn, $max_crit;
2784 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
2785 report('chassis', $msg, $err, $index);
2788 # Collect performance data
2789 if (defined $opt{perfdata}) {
2790 my $pname = lc $location;
2791 $pname =~ s{\s}{_}gxms;
2792 $pname =~ s{_temp\z}{}xms;
2793 $pname =~ s{proc_}{cpu#}xms;
2794 my $pkey = join q{_}, 'temp', $index, $pname;
2795 my $pval = join q{;}, "${reading}C", $max_warn, $max_crit;
2796 $perfdata{$pkey} = $pval;
2804 #-----------------------------------------
2805 # CHASSIS: Check processors
2806 #-----------------------------------------
2807 sub check_processors {
2819 # NOTE: For some reason, older models don't have the
2820 # "Processor Device Status" OIDs. We check both the newer
2821 # (preferred) OIDs and the old ones.
2825 '1.3.6.1.4.1.674.10892.1.1100.30.1.2.1' => 'processorDeviceIndex',
2826 '1.3.6.1.4.1.674.10892.1.1100.30.1.5.1' => 'processorDeviceStatus',
2827 '1.3.6.1.4.1.674.10892.1.1100.30.1.8.1' => 'processorDeviceManufacturerName',
2828 '1.3.6.1.4.1.674.10892.1.1100.30.1.9.1' => 'processorDeviceStatusState',
2829 '1.3.6.1.4.1.674.10892.1.1100.30.1.10.1' => 'processorDeviceFamily',
2830 '1.3.6.1.4.1.674.10892.1.1100.30.1.12.1' => 'processorDeviceCurrentSpeed',
2831 '1.3.6.1.4.1.674.10892.1.1100.30.1.23.1' => 'processorDeviceBrandName',
2832 '1.3.6.1.4.1.674.10892.1.1100.32.1.2.1' => 'processorDeviceStatusIndex',
2833 '1.3.6.1.4.1.674.10892.1.1100.32.1.5.1' => 'processorDeviceStatusStatus',
2834 '1.3.6.1.4.1.674.10892.1.1100.32.1.6.1' => 'processorDeviceStatusReading',
2837 my $result = $snmp_session->get_entries(-columns => [keys %cpu_oid]);
2839 if (!defined $result) {
2840 printf "SNMP ERROR [processors]: %s.\n", $snmp_session->error;
2841 $snmp_session->close;
2845 @output = @{ get_snmp_output($result, \%cpu_oid) };
2848 @output = @{ run_omreport("$omopt_chassis processors") };
2853 1 => 'Other', # other than following values
2854 2 => 'Unknown', # unknown
2855 3 => 'Enabled', # enabled
2856 4 => 'User Disabled', # disabled by user via BIOS setup
2857 5 => 'BIOS Disabled', # disabled by BIOS (POST error)
2863 1 => 'Internal Error', # Internal Error
2864 2 => 'Thermal Trip', # Thermal Trip
2865 32 => 'Configuration Error', # Configuration Error
2866 128 => 'Present', # Processor Present
2867 256 => 'Disabled', # Processor Disabled
2868 512 => 'Terminator Present', # Terminator Present
2869 1024 => 'Throttled', # Processor Throttled
2872 # Mapping between family numbers from SNMP and actual CPU family
2875 1 => 'Other', 2 => 'Unknown', 3 => '8086',
2876 4 => '80286', 5 => '386', 6 => '486',
2877 7 => '8087', 8 => '80287', 9 => '80387',
2878 10 => '80487', 11 => 'Pentium', 12 => 'Pentium Pro',
2879 13 => 'Pentium II', 14 => 'Pentium with MMX', 15 => 'Celeron',
2880 16 => 'Pentium II Xeon', 17 => 'Pentium III', 18 => 'Pentium III Xeon',
2881 19 => 'Pentium III', 20 => 'Itanium', 21 => 'Xeon',
2882 22 => 'Pentium 4', 23 => 'Xeon MP', 24 => 'Itanium 2',
2883 25 => 'K5', 26 => 'K6', 27 => 'K6-2',
2884 28 => 'K6-3', 29 => 'Athlon', 30 => 'AMD2900',
2885 31 => 'K6-2+', 32 => 'Power PC', 33 => 'Power PC 601',
2886 34 => 'Power PC 603', 35 => 'Power PC 603+', 36 => 'Power PC 604',
2887 37 => 'Power PC 620', 38 => 'Power PC x704', 39 => 'Power PC 750',
2888 48 => 'Alpha', 49 => 'Alpha 21064', 50 => 'Alpha 21066',
2889 51 => 'Alpha 21164', 52 => 'Alpha 21164PC', 53 => 'Alpha 21164a',
2890 54 => 'Alpha 21264', 55 => 'Alpha 21364', 64 => 'MIPS',
2891 65 => 'MIPS R4000', 66 => 'MIPS R4200', 67 => 'MIPS R4400',
2892 68 => 'MIPS R4600', 69 => 'MIPS R10000', 80 => 'SPARC',
2893 81 => 'SuperSPARC', 82 => 'microSPARC II', 83 => 'microSPARC IIep',
2894 84 => 'UltraSPARC', 85 => 'UltraSPARC II', 86 => 'UltraSPARC IIi',
2895 87 => 'UltraSPARC III', 88 => 'UltraSPARC IIIi', 96 => '68040',
2896 97 => '68xxx', 98 => '68000', 99 => '68010',
2897 100 => '68020', 101 => '68030', 112 => 'Hobbit',
2898 120 => 'Crusoe TM5000', 121 => 'Crusoe TM3000', 122 => 'Efficeon TM8000',
2899 128 => 'Weitek', 131 => 'Athlon 64', 132 => 'Opteron',
2900 133 => 'Sempron', 134 => 'Turion 64 Mobile', 135 => 'Dual-Core Opteron',
2901 136 => 'Athlon 64 X2 DC', 137 => 'Turion 64 X2 M', 138 => 'Quad-Core Opteron',
2902 139 => '3rd gen Opteron', 144 => 'PA-RISC', 145 => 'PA-RISC 8500',
2903 146 => 'PA-RISC 8000', 147 => 'PA-RISC 7300LC', 148 => 'PA-RISC 7200',
2904 149 => 'PA-RISC 7100LC', 150 => 'PA-RISC 7100', 160 => 'V30',
2905 171 => 'Dual-Core Xeon 5200', 172 => 'Dual-Core Xeon 7200', 173 => 'Quad-Core Xeon 7300',
2906 174 => 'Quad-Core Xeon 7400', 175 => 'Multi-Core Xeon 7400', 176 => 'M1',
2907 177 => 'M2', 180 => 'AS400', 182 => 'Athlon XP',
2908 183 => 'Athlon MP', 184 => 'Duron', 185 => 'Pentium M',
2909 186 => 'Celeron D', 187 => 'Pentium D', 188 => 'Pentium Extreme',
2910 189 => 'Core Solo', 190 => 'Core2', 191 => 'Core2 Duo',
2911 198 => 'Core i7', 199 => 'Dual-Core Celeron', 200 => 'IBM390',
2912 201 => 'G4', 202 => 'G5', 203 => 'ESA/390 G6',
2913 204 => 'z/Architectur', 210 => 'C7-M', 211 => 'C7-D',
2914 212 => 'C7', 213 => 'Eden', 214 => 'Multi-Core Xeon',
2915 215 => 'Dual-Core Xeon 3xxx', 216 => 'Quad-Core Xeon 3xxx', 218 => 'Dual-Core Xeon 5xxx',
2916 219 => 'Quad-Core Xeon 5xxx', 221 => 'Dual-Core Xeon 7xxx', 222 => 'Quad-Core Xeon 7xxx',
2917 223 => 'Multi-Core Xeon 7xxx', 250 => 'i860', 251 => 'i960',
2921 foreach my $out (@output) {
2923 $index = exists $out->{processorDeviceStatusIndex}
2924 ? $out->{processorDeviceStatusIndex} - 1
2925 : $out->{processorDeviceIndex} - 1;
2926 $status = exists $out->{processorDeviceStatusStatus}
2927 ? $snmp_status{$out->{processorDeviceStatusStatus}}
2928 : $snmp_status{$out->{processorDeviceStatus}};
2929 if (exists $out->{processorDeviceStatusReading}) {
2930 my @states = (); # contains states for the CPU
2932 # get the combined state from the StatusReading OID
2933 foreach my $mask (sort keys %cpu_reading) {
2934 if (($out->{processorDeviceStatusReading} & $mask) != 0) {
2935 push @states, $cpu_reading{$mask};
2939 # Finally, create the state string
2940 $state = join q{, }, @states;
2943 $state = $cpu_state{$out->{processorDeviceStatusState}};
2945 $man = $out->{processorDeviceManufacturerName};
2946 $family = (exists $out->{processorDeviceFamily}
2947 and exists $cpu_family{$out->{processorDeviceFamily}})
2948 ? $cpu_family{$out->{processorDeviceFamily}} : undef;
2949 $speed = $out->{processorDeviceCurrentSpeed};
2950 $brand = $out->{processorDeviceBrandName};
2953 $index = $out->{'Index'};
2954 $status = $out->{'Status'};
2955 $state = $out->{'State'};
2956 $brand = exists $out->{'Processor Brand'} ? $out->{'Processor Brand'} : undef;
2957 $family = exists $out->{'Processor Family'} ? $out->{'Processor Family'} : undef;
2958 $man = exists $out->{'Processor Manufacturer'} ? $out->{'Processor Manufacturer'} : undef;
2959 $speed = exists $out->{'Current Speed'} ? $out->{'Current Speed'} : undef;
2962 next CPU if blacklisted('cpu', $index);
2964 # Ignore unoccupied CPU slots (omreport)
2965 next CPU if (defined $out->{'Processor Manufacturer'}
2966 and $out->{'Processor Manufacturer'} eq '[Not Occupied]')
2967 or (defined $out->{'Processor Brand'} and $out->{'Processor Brand'} eq '[Not Occupied]');
2969 # Ignore unoccupied CPU slots (snmp)
2970 if ($snmp and exists $out->{processorDeviceStatusReading}
2971 and $out->{processorDeviceStatusReading} == 0) {
2977 if (defined $brand) {
2978 $brand =~ s{\s\s+}{ }gxms;
2979 $brand =~ s{\((R|tm)\)}{}gxms;
2980 $brand =~ s{\s(CPU|Processor)}{}xms;
2981 $brand =~ s{\s\@}{}xms;
2983 elsif (defined $family and defined $man and defined $speed) {
2984 $speed =~ s{\A (\d+) .*}{$1}xms;
2985 $brand = sprintf '%s %s %.2fGHz', $man, $family, $speed / 1000;
2992 if ($status ne 'Ok') {
2993 my $msg = sprintf 'Processor %d [%s] needs attention: %s',
2994 $index, $brand, $state;
2995 report('chassis', $msg, $status2nagios{$status}, $index);
2999 my $msg = sprintf 'Processor %d [%s] is %s',
3000 $index, $brand, $state;
3001 report('chassis', $msg, $E_OK, $index);
3008 #-----------------------------------------
3009 # CHASSIS: Check voltage probes
3010 #-----------------------------------------
3014 my $reading = undef;
3015 my $location = undef;
3021 '1.3.6.1.4.1.674.10892.1.600.20.1.2.1' => 'voltageProbeIndex',
3022 '1.3.6.1.4.1.674.10892.1.600.20.1.5.1' => 'voltageProbeStatus',
3023 '1.3.6.1.4.1.674.10892.1.600.20.1.6.1' => 'voltageProbeReading',
3024 '1.3.6.1.4.1.674.10892.1.600.20.1.8.1' => 'voltageProbeLocationName',
3025 '1.3.6.1.4.1.674.10892.1.600.20.1.16.1' => 'voltageProbeDiscreteReading',
3028 my $voltageProbeTable = '1.3.6.1.4.1.674.10892.1.600.20.1';
3029 my $result = $snmp_session->get_table(-baseoid => $voltageProbeTable);
3031 if (!defined $result) {
3032 printf "SNMP ERROR [voltage]: %s.\n", $snmp_session->error;
3033 $snmp_session->close;
3037 @output = @{ get_snmp_output($result, \%volt_oid) };
3040 @output = @{ run_omreport("$omopt_chassis volts") };
3043 my %volt_discrete_reading
3050 foreach my $out (@output) {
3052 $index = $out->{voltageProbeIndex} - 1;
3053 $status = $snmp_status{$out->{voltageProbeStatus}};
3054 $reading = exists $out->{voltageProbeReading}
3055 ? sprintf('%.3f V', $out->{voltageProbeReading}/1000)
3056 : $volt_discrete_reading{$out->{voltageProbeDiscreteReading}};
3057 $location = $out->{voltageProbeLocationName};
3060 $index = $out->{'Index'};
3061 $status = $out->{'Status'};
3062 $reading = $out->{'Reading'};
3063 $location = $out->{'Probe Name'};
3066 next VOLT if blacklisted('volt', $index);
3069 my $msg = sprintf 'Voltage sensor %d [%s] is %s',
3070 $index, $location, $reading;
3071 my $err = $snmp ? $probestatus2nagios{$status} : $status2nagios{$status};
3072 report('chassis', $msg, $err, $index);
3078 #-----------------------------------------
3079 # CHASSIS: Check batteries
3080 #-----------------------------------------
3081 sub check_batteries {
3084 my $reading = undef;
3085 my $location = undef;
3091 '1.3.6.1.4.1.674.10892.1.600.50.1.2.1' => 'batteryIndex',
3092 '1.3.6.1.4.1.674.10892.1.600.50.1.5.1' => 'batteryStatus',
3093 '1.3.6.1.4.1.674.10892.1.600.50.1.6.1' => 'batteryReading',
3094 '1.3.6.1.4.1.674.10892.1.600.50.1.7.1' => 'batteryLocationName',
3096 my $result = $snmp_session->get_entries(-columns => [keys %bat_oid]);
3098 # No batteries is OK
3099 return 0 if !defined $result;
3101 @output = @{ get_snmp_output($result, \%bat_oid) };
3104 @output = @{ run_omreport("$omopt_chassis batteries") };
3109 1 => 'Predictive Failure',
3111 4 => 'Presence Detected',
3115 foreach my $out (@output) {
3117 $index = $out->{batteryIndex} - 1;
3118 $status = $snmp_status{$out->{batteryStatus}};
3119 $reading = $bat_reading{$out->{batteryReading}};
3120 $location = $out->{batteryLocationName};
3123 $index = $out->{'Index'};
3124 $status = $out->{'Status'};
3125 $reading = $out->{'Reading'};
3126 $location = $out->{'Probe Name'};
3129 next BATTERY if blacklisted('bp', $index);
3132 my $msg = sprintf 'Battery probe %d [%s] is %s',
3133 $index, $location, $reading;
3134 report('chassis', $msg, $status2nagios{$status}, $index);
3140 #-----------------------------------------
3141 # CHASSIS: Check amperage probes (power monitoring)
3142 #-----------------------------------------
3143 sub check_pwrmonitoring {
3146 my $reading = undef;
3147 my $location = undef;
3148 my $max_crit = undef;
3149 my $max_warn = undef;
3156 '1.3.6.1.4.1.674.10892.1.600.30.1.2.1' => 'amperageProbeIndex',
3157 '1.3.6.1.4.1.674.10892.1.600.30.1.5.1' => 'amperageProbeStatus',
3158 '1.3.6.1.4.1.674.10892.1.600.30.1.6.1' => 'amperageProbeReading',
3159 '1.3.6.1.4.1.674.10892.1.600.30.1.7.1' => 'amperageProbeType',
3160 '1.3.6.1.4.1.674.10892.1.600.30.1.8.1' => 'amperageProbeLocationName',
3161 '1.3.6.1.4.1.674.10892.1.600.30.1.10.1' => 'amperageProbeUpperCriticalThreshold',
3162 '1.3.6.1.4.1.674.10892.1.600.30.1.11.1' => 'amperageProbeUpperNonCriticalThreshold',
3163 '1.3.6.1.4.1.674.10892.1.600.30.1.16.1' => 'amperageProbeDiscreteReading',
3165 my $result = $snmp_session->get_entries(-columns => [keys %amp_oid]);
3167 # No pwrmonitoring is OK
3168 return 0 if !defined $result;
3170 @output = @{ get_snmp_output($result, \%amp_oid) };
3173 @output = @{ run_omreport("$omopt_chassis pwrmonitoring") };
3176 my %amp_type # Amperage probe types
3178 1 => 'amperageProbeTypeIsOther', # other than following values
3179 2 => 'amperageProbeTypeIsUnknown', # unknown
3180 3 => 'amperageProbeTypeIs1Point5Volt', # 1.5 amperage probe
3181 4 => 'amperageProbeTypeIs3Point3volt', # 3.3 amperage probe
3182 5 => 'amperageProbeTypeIs5Volt', # 5 amperage probe
3183 6 => 'amperageProbeTypeIsMinus5Volt', # -5 amperage probe
3184 7 => 'amperageProbeTypeIs12Volt', # 12 amperage probe
3185 8 => 'amperageProbeTypeIsMinus12Volt', # -12 amperage probe
3186 9 => 'amperageProbeTypeIsIO', # I/O probe
3187 10 => 'amperageProbeTypeIsCore', # Core probe
3188 11 => 'amperageProbeTypeIsFLEA', # FLEA (standby) probe
3189 12 => 'amperageProbeTypeIsBattery', # Battery probe
3190 13 => 'amperageProbeTypeIsTerminator', # SCSI Termination probe
3191 14 => 'amperageProbeTypeIs2Point5Volt', # 2.5 amperage probe
3192 15 => 'amperageProbeTypeIsGTL', # GTL (ground termination logic) probe
3193 16 => 'amperageProbeTypeIsDiscrete', # amperage probe with discrete reading
3194 23 => 'amperageProbeTypeIsPowerSupplyAmps', # Power Supply probe with reading in Amps
3195 24 => 'amperageProbeTypeIsPowerSupplyWatts', # Power Supply probe with reading in Watts
3196 25 => 'amperageProbeTypeIsSystemAmps', # System probe with reading in Amps
3197 26 => 'amperageProbeTypeIsSystemWatts', # System probe with reading in Watts
3208 'amperageProbeTypeIsPowerSupplyAmps' => 'hA', # tenths of Amps
3209 'amperageProbeTypeIsSystemAmps' => 'hA', # tenths of Amps
3210 'amperageProbeTypeIsPowerSupplyWatts' => 'W', # Watts
3211 'amperageProbeTypeIsSystemWatts' => 'W', # Watts
3212 'amperageProbeTypeIsDiscrete' => q{}, # discrete reading, no unit
3216 foreach my $out (@output) {
3218 $index = $out->{amperageProbeIndex} - 1;
3219 $status = $snmp_status{$out->{amperageProbeStatus}};
3220 $reading = $amp_type{$out->{amperageProbeType}} eq 'amperageProbeTypeIsDiscrete'
3221 ? $amp_discrete{$out->{amperageProbeDiscreteReading}}
3222 : $out->{amperageProbeReading};
3223 $location = $out->{amperageProbeLocationName};
3224 $max_crit = exists $out->{amperageProbeUpperCriticalThreshold}
3225 ? $out->{amperageProbeUpperCriticalThreshold} : 0;
3226 $max_warn = exists $out->{amperageProbeUpperNonCriticalThreshold}
3227 ? $out->{amperageProbeUpperNonCriticalThreshold} : 0;
3228 $unit = exists $amp_unit{$amp_type{$out->{amperageProbeType}}}
3229 ? $amp_unit{$amp_type{$out->{amperageProbeType}}} : 'mA';
3230 if ($unit eq 'hA') {
3238 $index = $out->{'Index'};
3239 next AMP if (!defined $index || $index !~ m/^\d+$/x);
3240 $status = $out->{'Status'};
3241 $reading = $out->{'Reading'};
3242 $location = $out->{'Probe Name'};
3243 $max_crit = $out->{'Failure Threshold'} ne '[N/A]'
3244 ? $out->{'Failure Threshold'} : 0;
3245 $max_warn = $out->{'Warning Threshold'} ne '[N/A]'
3246 ? $out->{'Warning Threshold'} : 0;
3247 $reading =~ s{\A (\d+.*?)\s+([a-zA-Z]+) \s*\z}{$1}xms;
3249 $max_warn =~ s{\A (\d+.*?)\s+[a-zA-Z]+ \s*\z}{$1}xms;
3250 $max_crit =~ s{\A (\d+.*?)\s+[a-zA-Z]+ \s*\z}{$1}xms;
3253 next AMP if blacklisted('amp', $index);
3254 next AMP if $index !~ m{\A \d+ \z}xms;
3257 my $msg = sprintf 'Amperage probe %d [%s] reads %s %s',
3258 $index, $location, $reading, $unit, $status;
3259 report('chassis', $msg, $status2nagios{$status}, $index);
3261 # Collect performance data
3262 if (defined $opt{perfdata}) {
3263 next AMP if $reading !~ m{\A \d+(\.\d+)? \z}xms; # discrete reading (not number)
3264 my $pname = lc $location;
3265 $pname =~ s{\s}{_}gxms;
3266 my $pkey = join q{_}, 'pwr_mon', $index, $pname;
3267 my $pval = join q{;}, "$reading$unit", $max_warn, $max_crit;
3268 $perfdata{$pkey} = $pval;
3272 # Collect EXTRA performance data not found at first run. This is a
3274 if (defined $opt{perfdata} && !$snmp) {
3280 foreach (keys %perfdata) {
3281 if (m/\A pwr_mon_(\d+)/xms) {
3287 foreach my $line (@{ run_command("$omreport $omopt_chassis pwrmonitoring -fmt ssv") }) {
3289 if ($line eq 'Location;Reading') {
3297 if ($found and $line =~ m/\A ([^;]+?) ; (\d*\.\d+) \s ([AW]) \z/xms) {
3301 $aname =~ s{\s}{_}gxms;
3303 # don't use an existing index
3304 while (exists $used{$index}) { ++$index; }
3306 $perfdata{"pwr_mon_${index}_${aname}"} = "$aval$aunit;0;0";
3316 #-----------------------------------------
3317 # CHASSIS: Check intrusion
3318 #-----------------------------------------
3319 sub check_intrusion {
3322 my $reading = undef;
3328 '1.3.6.1.4.1.674.10892.1.300.70.1.2.1' => 'intrusionIndex',
3329 '1.3.6.1.4.1.674.10892.1.300.70.1.5.1' => 'intrusionStatus',
3330 '1.3.6.1.4.1.674.10892.1.300.70.1.6.1' => 'intrusionReading',
3332 my $result = $snmp_session->get_entries(-columns => [keys %int_oid]);
3334 # No intrusion is OK
3335 return 0 if !defined $result;
3337 @output = @{ get_snmp_output($result, \%int_oid) };
3340 @output = @{ run_omreport("$omopt_chassis intrusion") };
3345 1 => 'Not Breached', # chassis not breached and no uncleared breaches
3346 2 => 'Breached', # chassis currently breached
3347 3 => 'Breached Prior', # chassis breached prior to boot and has not been cleared
3348 4 => 'Breach Sensor Failure', # intrusion sensor has failed
3352 foreach my $out (@output) {
3354 $index = $out->{intrusionIndex} - 1;
3355 $status = $snmp_status{$out->{intrusionStatus}};
3356 $reading = $int_reading{$out->{intrusionReading}};
3359 $index = $out->{'Index'};
3360 $status = $out->{'Status'};
3361 $reading = $out->{'State'};
3364 next INTRUSION if blacklisted('intr', $index);
3367 if ($status ne 'Ok') {
3368 my $msg = sprintf 'Chassis intrusion %d detected: %s',
3370 report('chassis', $msg, $E_WARNING, $index);
3374 my $msg = sprintf 'Chassis intrusion %d detection: %s (%s)',
3375 $index, $status, $reading;
3376 report('chassis', $msg, $E_OK, $index);
3383 #-----------------------------------------
3384 # CHASSIS: Check alert log
3385 #-----------------------------------------
3386 sub check_alertlog {
3387 return if $snmp; # Not supported with SNMP
3389 my @output = @{ run_omreport("$omopt_system alertlog") };
3390 foreach my $out (@output) {
3391 ++$count{alert}{$out->{Severity}};
3394 # Create error messages and set exit value if appropriate
3396 if ($count{alert}{'Critical'} > 0) { $err = $E_CRITICAL; }
3397 elsif ($count{alert}{'Non-Critical'} > 0) { $err = $E_WARNING; }
3399 my $msg = sprintf 'Alert log content: %d critical, %d non-critical, %d ok',
3400 $count{alert}{'Critical'}, $count{alert}{'Non-Critical'}, $count{alert}{'Ok'};
3401 report('other', $msg, $err);
3406 #-----------------------------------------
3407 # CHASSIS: Check ESM log overall health
3408 #-----------------------------------------
3409 sub check_esmlog_health {
3413 my $systemStateEventLogStatus = '1.3.6.1.4.1.674.10892.1.200.10.1.41.1';
3414 my $result = $snmp_session->get_request(-varbindlist => [$systemStateEventLogStatus]);
3415 if (!defined $result) {
3416 my $msg = sprintf 'SNMP ERROR [esmhealth]: %s',
3417 $snmp_session->error;
3418 report('other', $msg, $E_UNKNOWN);
3420 $health = $snmp_status{$result->{$systemStateEventLogStatus}};
3423 foreach (@{ run_command("$omreport $omopt_system esmlog -fmt ssv") }) {
3424 if (m/\A Health;(.+) \z/xms) {
3432 # If the overall health of the ESM log is other than "Ok", the
3433 # fill grade of the log is more than 80% and the log should be
3435 if ($health eq 'Ok') {
3436 my $msg = sprintf 'ESM log health is Ok (less than 80%% full)';
3437 report('other', $msg, $E_OK);
3439 elsif ($health eq 'Critical') {
3440 my $msg = sprintf 'ESM log is 100%% full';
3441 report('other', $msg, $status2nagios{$health});
3444 my $msg = sprintf 'ESM log is more than 80%% full';
3445 report('other', $msg, $status2nagios{$health});
3451 #-----------------------------------------
3452 # CHASSIS: Check ESM log
3453 #-----------------------------------------
3460 '1.3.6.1.4.1.674.10892.1.300.40.1.7.1' => 'eventLogSeverityStatus',
3462 my $result = $snmp_session->get_entries(-columns => [keys %esm_oid]);
3465 return if !defined $result;
3467 @output = @{ get_snmp_output($result, \%esm_oid) };
3468 foreach my $out (@output) {
3469 ++$count{esm}{$snmp_status{$out->{eventLogSeverityStatus}}};
3473 @output = @{ run_omreport("$omopt_system esmlog") };
3474 foreach my $out (@output) {
3475 ++$count{esm}{$out->{Severity}};
3479 # Create error messages and set exit value if appropriate
3481 if ($count{esm}{'Critical'} > 0) { $err = $E_CRITICAL; }
3482 elsif ($count{esm}{'Non-Critical'} > 0) { $err = $E_WARNING; }
3484 my $msg = sprintf 'ESM log content: %d critical, %d non-critical, %d ok',
3485 $count{esm}{'Critical'}, $count{esm}{'Non-Critical'}, $count{esm}{'Ok'};
3486 report('other', $msg, $err);
3492 # Handy function for checking all storage components
3495 check_controllers();
3496 check_physical_disks();
3497 check_virtual_disks();
3498 check_cache_battery();
3501 check_enclosure_fans();
3502 check_enclosure_pwr();
3503 check_enclosure_temp();
3504 check_enclosure_emms();
3510 #---------------------------------------------------------------------
3512 #---------------------------------------------------------------------
3515 # Fetch output from 'omreport chassis info', put in sysinfo hash
3517 sub get_omreport_chassis_info {
3518 if (open my $INFO, '-|', "$omreport $omopt_chassis info -fmt ssv") {
3519 my @lines = <$INFO>;
3522 next if !m/\A (Chassis\sModel|Chassis\sService\sTag|Model|Service\sTag)/xms;
3523 my ($key, $val) = split /;/xms;
3524 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3525 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3526 if ($key eq 'Chassis Model' or $key eq 'Model') {
3527 $sysinfo{model} = $val;
3529 if ($key eq 'Chassis Service Tag' or $key eq 'Service Tag') {
3530 $sysinfo{serial} = $val;
3538 # Fetch output from 'omreport chassis bios', put in sysinfo hash
3540 sub get_omreport_chassis_bios {
3541 if (open my $BIOS, '-|', "$omreport $omopt_chassis bios -fmt ssv") {
3542 my @lines = <$BIOS>;
3546 my ($key, $val) = split /;/xms;
3547 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3548 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3549 $sysinfo{bios} = $val if $key eq 'Version';
3550 $sysinfo{biosdate} = $val if $key eq 'Release Date';
3557 # Fetch output from 'omreport system operatingsystem', put in sysinfo hash
3559 sub get_omreport_system_operatingsystem {
3560 if (open my $VER, '-|', "$omreport $omopt_system operatingsystem -fmt ssv") {
3565 my ($key, $val) = split /;/xms;
3566 $key =~ s{\s+\z}{}xms; # remove trailing whitespace
3567 $val =~ s{\s+\z}{}xms; # remove trailing whitespace
3568 if ($key eq 'Operating System') {
3569 $sysinfo{osname} = $val;
3571 elsif ($key eq 'Operating System Version') {
3572 $sysinfo{osver} = $val;
3580 # Fetch output from 'omreport about', put in sysinfo hash
3582 sub get_omreport_about {
3583 if (open my $OM, '-|', "$omreport about -fmt ssv") {
3587 if (m/\A Version;(.+) \z/xms) {
3597 # Fetch chassis info via SNMP, put in sysinfo hash
3599 sub get_snmp_chassis_info {
3602 '1.3.6.1.4.1.674.10892.1.300.10.1.9.1' => 'chassisModelName',
3603 '1.3.6.1.4.1.674.10892.1.300.10.1.11.1' => 'chassisServiceTagName',
3606 my $chassisInformationTable = '1.3.6.1.4.1.674.10892.1.300.10.1';
3607 my $result = $snmp_session->get_table(-baseoid => $chassisInformationTable);
3609 if (defined $result) {
3610 foreach my $oid (keys %{ $result }) {
3611 if (exists $chassis_oid{$oid} and $chassis_oid{$oid} eq 'chassisModelName') {
3612 $sysinfo{model} = $result->{$oid};
3613 $sysinfo{model} =~ s{\s+\z}{}xms; # remove trailing whitespace
3615 elsif (exists $chassis_oid{$oid} and $chassis_oid{$oid} eq 'chassisServiceTagName') {
3616 $sysinfo{serial} = $result->{$oid};
3621 my $msg = sprintf 'SNMP ERROR getting chassis info: %s',
3622 $snmp_session->error;
3623 report('other', $msg, $E_UNKNOWN);
3629 # Fetch BIOS info via SNMP, put in sysinfo hash
3631 sub get_snmp_chassis_bios {
3634 '1.3.6.1.4.1.674.10892.1.300.50.1.7.1.1' => 'systemBIOSReleaseDateName',
3635 '1.3.6.1.4.1.674.10892.1.300.50.1.8.1.1' => 'systemBIOSVersionName',
3638 my $systemBIOSTable = '1.3.6.1.4.1.674.10892.1.300.50.1';
3639 my $result = $snmp_session->get_table(-baseoid => $systemBIOSTable);
3641 if (defined $result) {
3642 foreach my $oid (keys %{ $result }) {
3643 if (exists $bios_oid{$oid} and $bios_oid{$oid} eq 'systemBIOSReleaseDateName') {
3644 $sysinfo{biosdate} = $result->{$oid};
3645 $sysinfo{biosdate} =~ s{\A (\d{4})(\d{2})(\d{2}).*}{$2/$3/$1}xms;
3647 elsif (exists $bios_oid{$oid} and $bios_oid{$oid} eq 'systemBIOSVersionName') {
3648 $sysinfo{bios} = $result->{$oid};
3653 my $msg = sprintf 'SNMP ERROR getting BIOS info: %s',
3654 $snmp_session->error;
3655 report('other', $msg, $E_UNKNOWN);
3661 # Fetch OS info via SNMP, put in sysinfo hash
3663 sub get_snmp_system_operatingsystem {
3666 '1.3.6.1.4.1.674.10892.1.400.10.1.6.1' => 'operatingSystemOperatingSystemName',
3667 '1.3.6.1.4.1.674.10892.1.400.10.1.7.1' => 'operatingSystemOperatingSystemVersionName',
3670 my $operatingSystemTable = '1.3.6.1.4.1.674.10892.1.400.10.1';
3671 my $result = $snmp_session->get_table(-baseoid => $operatingSystemTable);
3673 if (defined $result) {
3674 foreach my $oid (keys %{ $result }) {
3675 if (exists $os_oid{$oid} and $os_oid{$oid} eq 'operatingSystemOperatingSystemName') {
3676 $sysinfo{osname} = ($result->{$oid});
3678 elsif (exists $os_oid{$oid} and $os_oid{$oid} eq 'operatingSystemOperatingSystemVersionName') {
3679 $sysinfo{osver} = $result->{$oid};
3684 my $msg = sprintf 'SNMP ERROR getting OS info: %s',
3685 $snmp_session->error;
3686 report('other', $msg, $E_UNKNOWN);
3692 # Fetch OMSA version via SNMP, put in sysinfo hash
3694 sub get_snmp_about {
3697 '1.3.6.1.4.1.674.10892.1.100.10.0' => 'systemManagementSoftwareGlobalVersionName',
3699 my $systemManagementSoftwareGroup = '1.3.6.1.4.1.674.10892.1.100';
3700 my $result = $snmp_session->get_table(-baseoid => $systemManagementSoftwareGroup);
3701 if (defined $result) {
3702 foreach my $oid (keys %{ $result }) {
3703 if (exists $omsa_oid{$oid} and $omsa_oid{$oid} eq 'systemManagementSoftwareGlobalVersionName') {
3704 $sysinfo{om} = ($result->{$oid});
3709 my $msg = sprintf 'SNMP ERROR getting OMSA info: %s',
3710 $snmp_session->error;
3711 report('other', $msg, $E_UNKNOWN);
3717 # Collects some information about the system
3721 # Get system model and serial number
3722 $snmp ? get_snmp_chassis_info() : get_omreport_chassis_info();
3724 # Get BIOS information. Only if needed
3725 if ( $opt{okinfo} >= 1
3727 or (defined $opt{postmsg} and $opt{postmsg} =~ m/[%][bd]/xms) ) {
3728 $snmp ? get_snmp_chassis_bios() : get_omreport_chassis_bios();
3731 # Get OMSA information. Only if needed
3732 if ($opt{okinfo} >= 3 or $opt{debug}) {
3733 $snmp ? get_snmp_about() : get_omreport_about();
3736 # Return now if debug
3737 return if $opt{debug};
3739 # Get OS information. Only if needed
3740 if (defined $opt{postmsg} and $opt{postmsg} =~ m/[%][or]/xms) {
3741 $snmp ? get_snmp_system_operatingsystem() : get_omreport_system_operatingsystem();
3748 # Helper function for running omreport when the results are strictly
3750 sub run_omreport_info {
3751 my $command = shift;
3755 # Run omreport and fetch output
3756 my $rawtext = slurp_command("$omreport $command -fmt ssv 2>&1");
3758 # Parse output, store in array
3759 for ((split /\n/xms, $rawtext)) {
3760 if (m/\A Error/xms) {
3761 my $msg = "Problem running 'omreport $command': $_";
3762 report('other', $msg, $E_UNKNOWN);
3764 next if !m/;/xms; # ignore lines with less than two fields
3765 my @vals = split m/;/xms;
3766 $output{$vals[0]} = $vals[1];
3769 # Finally, return the collected information
3773 # Get various firmware information (BMC, RAC)
3774 sub get_firmware_info {
3775 my @snmp_output = ();
3776 my %nrpe_output = ();
3781 '1.3.6.1.4.1.674.10892.1.300.60.1.7.1' => 'firmwareType',
3782 '1.3.6.1.4.1.674.10892.1.300.60.1.8.1' => 'firmwareTypeName',
3783 '1.3.6.1.4.1.674.10892.1.300.60.1.11.1' => 'firmwareVersionName',
3786 my $firmwareTable = '1.3.6.1.4.1.674.10892.1.300.60.1';
3787 my $result = $snmp_session->get_table(-baseoid => $firmwareTable);
3789 # Some don't have this OID, this is ok
3790 if (!defined $result) {
3794 @snmp_output = @{ get_snmp_output($result, \%fw_oid) };
3797 %nrpe_output = %{ run_omreport_info("$omopt_chassis info") };
3800 my %fw_type # Firmware types
3802 1 => 'other', # other than following values
3803 2 => 'unknown', # unknown
3804 3 => 'systemBIOS', # System BIOS
3805 4 => 'embeddedSystemManagementController', # Embedded System Management Controller
3806 5 => 'powerSupplyParallelingBoard', # Power Supply Paralleling Board
3807 6 => 'systemBackPlane', # System (Primary) Backplane
3808 7 => 'powerVault2XXSKernel', # PowerVault 2XXS Kernel
3809 8 => 'powerVault2XXSApplication', # PowerVault 2XXS Application
3810 9 => 'frontPanel', # Front Panel Controller
3811 10 => 'baseboardManagementController', # Baseboard Management Controller
3812 11 => 'hotPlugPCI', # Hot Plug PCI Controller
3813 12 => 'sensorData', # Sensor Data Records
3814 13 => 'peripheralBay', # Peripheral Bay Backplane
3815 14 => 'secondaryBackPlane', # Secondary Backplane for ESM 2 systems
3816 15 => 'secondaryBackPlaneESM3And4', # Secondary Backplane for ESM 3 and 4 systems
3817 16 => 'rac', # Remote Access Controller
3818 17 => 'imc' # Integrated Management Controller
3823 foreach my $out (@snmp_output) {
3824 if ($fw_type{$out->{firmwareType}} eq 'baseboardManagementController') {
3825 $sysinfo{'bmc'} = 1;
3826 $sysinfo{'bmc_fw'} = $out->{firmwareVersionName};
3828 elsif ($fw_type{$out->{firmwareType}} =~ m{\A rac|imc \z}xms) {
3829 my $name = $out->{firmwareTypeName}; $name =~ s/\s//gxms;
3830 $sysinfo{'rac'} = 1;
3831 $sysinfo{'rac_name'} = $name;
3832 $sysinfo{'rac_fw'} = $out->{firmwareVersionName};
3837 foreach my $key (keys %nrpe_output) {
3838 next if !defined $nrpe_output{$key};
3839 if ($key eq 'BMC Version' or $key eq 'Baseboard Management Controller Version') {
3840 $sysinfo{'bmc'} = 1;
3841 $sysinfo{'bmc_fw'} = $nrpe_output{$key};
3843 elsif ($key =~ m{\A (i?DRAC)\s*(\d?)\s+Version}xms) {
3845 $sysinfo{'rac'} = 1;
3846 $sysinfo{'rac_fw'} = $nrpe_output{$key};
3847 $sysinfo{'rac_name'} = $name;
3857 #=====================================================================
3859 #=====================================================================
3861 # Here we do the actual checking of components
3862 # Check global status if applicable
3864 $globalstatus = check_global();
3867 # Do multiple selected checks
3868 if ($check{storage}) { check_storage(); }
3869 if ($check{memory}) { check_memory(); }
3870 if ($check{fans}) { check_fans(); }
3871 if ($check{power}) { check_powersupplies(); }
3872 if ($check{temp}) { check_temperatures(); }
3873 if ($check{cpu}) { check_processors(); }
3874 if ($check{voltage}) { check_volts(); }
3875 if ($check{batteries}) { check_batteries(); }
3876 if ($check{amperage}) { check_pwrmonitoring(); }
3877 if ($check{intrusion}) { check_intrusion(); }
3878 if ($check{alertlog}) { check_alertlog(); }
3879 if ($check{esmlog}) { check_esmlog(); }
3880 if ($check{esmhealth}) { check_esmlog_health(); }
3883 #---------------------------------------------------------------------
3885 #---------------------------------------------------------------------
3896 # Get system information
3899 # Get firmware info if requested via option
3900 if ($opt{okinfo} >= 1) {
3901 get_firmware_info();
3904 # Close SNMP session
3906 $snmp_session->close;
3911 print " System: $sysinfo{model}\n";
3912 print " ServiceTag: $sysinfo{serial}";
3913 print q{ } x (25 - length $sysinfo{serial}), "OMSA version: $sysinfo{om}\n";
3914 print " BIOS/date: $sysinfo{bios} $sysinfo{biosdate}";
3915 print q{ } x (25 - length "$sysinfo{bios} $sysinfo{biosdate}"), "Plugin version: $VERSION\n";
3916 if ($#report_storage >= 0) {
3917 print "-----------------------------------------------------------------------------\n";
3918 print " Storage Components \n";
3919 print "=============================================================================\n";
3920 print " STATE | ID | MESSAGE TEXT \n";
3921 print "---------+----------+--------------------------------------------------------\n";
3922 foreach (@report_storage) {
3923 my ($msg, $level, $nexus) = @{$_};
3924 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | "
3925 . q{ } x (8 - length $nexus) . "$nexus | $msg\n";
3926 $nagios_alert_count{$reverse_exitcode{$level}}++;
3929 if ($#report_chassis >= 0) {
3930 print "-----------------------------------------------------------------------------\n";
3931 print " Chassis Components \n";
3932 print "=============================================================================\n";
3933 print " STATE | ID | MESSAGE TEXT \n";
3934 print "---------+------+------------------------------------------------------------\n";
3935 foreach (@report_chassis) {
3936 my ($msg, $level, $nexus) = @{$_};
3937 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | "
3938 . q{ } x (4 - length $nexus) . "$nexus | $msg\n";
3939 $nagios_alert_count{$reverse_exitcode{$level}}++;
3942 if ($#report_other >= 0) {
3943 print "-----------------------------------------------------------------------------\n";
3944 print " Other messages \n";
3945 print "=============================================================================\n";
3946 print " STATE | MESSAGE TEXT \n";
3947 print "---------+-------------------------------------------------------------------\n";
3948 foreach (@report_other) {
3949 my ($msg, $level, $nexus) = @{$_};
3950 print q{ } x (8 - length $reverse_exitcode{$level}) . "$reverse_exitcode{$level} | $msg\n";
3951 $nagios_alert_count{$reverse_exitcode{$level}}++;
3956 my $c = 0; # counter to determine linebreaks
3958 # Run through each message, sorted by severity level
3960 foreach (sort {$a->[1] < $b->[1]} (@report_storage, @report_chassis, @report_other)) {
3961 my ($msg, $level, $nexus) = @{ $_ };
3962 next ALERT if $level == $E_OK;
3964 if (defined $opt{only}) {
3965 # If user wants only critical alerts
3966 next ALERT if ($opt{only} eq 'critical' and $level == $E_WARNING);
3968 # If user wants only warning alerts
3969 next ALERT if ($opt{only} eq 'warning' and $level == $E_CRITICAL);
3972 # Prefix with service tag if specified with option '-i|--info'
3974 if (defined $opt{htmlinfo}) {
3975 $msg = '[<a href="' . warranty_url($sysinfo{serial})
3976 . "\">$sysinfo{serial}</a>] " . $msg;
3979 $msg = "[$sysinfo{serial}] " . $msg;
3983 # Prefix with nagios level if specified with option '--state'
3984 $msg = $reverse_exitcode{$level} . ": $msg" if $opt{state};
3986 # Prefix with one-letter nagios level if specified with option '--short-state'
3987 $msg = (substr $reverse_exitcode{$level}, 0, 1) . ": $msg" if $opt{shortstate};
3989 ($c++ == 0) ? print $msg : print $linebreak, $msg;
3991 $nagios_alert_count{$reverse_exitcode{$level}}++;
3995 # Determine our exit code
3997 $exit_code = $E_UNKNOWN if $nagios_alert_count{'UNKNOWN'} > 0;
3998 $exit_code = $E_WARNING if $nagios_alert_count{'WARNING'} > 0;
3999 $exit_code = $E_CRITICAL if $nagios_alert_count{'CRITICAL'} > 0;
4001 # Global status via SNMP.. extra safety check
4002 if ($globalstatus != $E_OK && $exit_code == $E_OK && !defined $opt{only}) {
4003 print "OOPS! Something is wrong with this server, but I don't know what. ";
4004 print "The global system health status is $reverse_exitcode{$globalstatus}, ";
4005 print "but every component check is OK. This may be a bug in the Nagios plugin, ";
4006 print "please file a bug report.\n";
4011 if ($exit_code == $E_OK && defined $opt{only} && $opt{only} !~ m{\A critical|warning|chassis \z}xms && !$opt{debug}) {
4013 = ( 'storage' => "STORAGE OK - $count{pdisk} physical drives, $count{vdisk} logical drives",
4014 'fans' => $count{fan} == 0 && $blade ? 'OK - blade system with no fan probes' : "FANS OK - $count{fan} fan probes checked",
4015 'temp' => "TEMPERATURES OK - $count{temp} temperature probes checked",
4016 'memory' => "MEMORY OK - $count{dimm} memory modules checked",
4017 'power' => $count{power} == 0 ? 'OK - no instrumented power supplies found' : "POWER OK - $count{power} power supplies checked",
4018 'cpu' => "PROCESSORS OK - $count{cpu} processors checked",
4019 'voltage' => "VOLTAGE OK - $count{volt} voltage probes checked",
4020 'batteries' => $count{bat} == 0 ? 'OK - no batteries found' : "BATTERIES OK - $count{bat} batteries checked",
4021 'amperage' => $count{amp} == 0 ? 'OK - no power monitoring probes found' : "AMPERAGE OK - $count{amp} amperage (power monitoring) probes checked",
4022 'intrusion' => $count{intr} == 0 ? 'OK - no intrusion detection probes found' : "INTRUSION OK - $count{intr} intrusion detection probes checked",
4023 '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",
4024 'esmlog' => "OK - ESM Log content: $count{esm}{Ok} ok, $count{esm}{'Non-Critical'} warning and $count{esm}{Critical} critical",
4025 'esmhealth' => "ESM LOG OK - less than 80% used",
4028 print $okmsg{$opt{only}};
4030 elsif ($exit_code == $E_OK && !$opt{debug}) {
4031 if (defined $opt{htmlinfo}) {
4032 printf q{OK - System: '<a href="%s">%s</a>', SN: '<a href="%s">%s</a>', hardware working fine},
4033 documentation_url($sysinfo{model}), $sysinfo{model},
4034 warranty_url($sysinfo{serial}), $sysinfo{serial};
4037 printf q{OK - System: '%s', SN: '%s', hardware working fine},
4038 $sysinfo{model}, $sysinfo{serial};
4041 if ($check{storage}) {
4042 printf ', %d logical drives, %d physical drives',
4043 $count{vdisk}, $count{pdisk};
4046 print ', not checking storage';
4049 if ($opt{okinfo} >= 1) {
4051 printf q{----- BIOS='%s %s'}, $sysinfo{bios}, $sysinfo{biosdate};
4053 if ($sysinfo{rac}) {
4054 printf q{, %s='%s'}, $sysinfo{rac_name}, $sysinfo{rac_fw};
4056 if ($sysinfo{bmc}) {
4057 printf q{, BMC='%s'}, $sysinfo{bmc_fw};
4061 if ($opt{okinfo} >= 2) {
4062 if ($check{storage}) {
4063 my @storageprint = ();
4064 foreach my $id (sort keys %{ $sysinfo{controller} }) {
4065 chomp $sysinfo{controller}{$id}{driver};
4066 push @storageprint, sprintf q{----- CTRL %s (%s): FW='%s', DR='%s'},
4067 $sysinfo{controller}{$id}{id}, $sysinfo{controller}{$id}{name},
4068 $sysinfo{controller}{$id}{firmware}, $sysinfo{controller}{$id}{driver};
4070 foreach my $id (sort keys %{ $sysinfo{enclosure} }) {
4071 push @storageprint, sprintf q{----- ENCL %s (%s): FW='%s'},
4072 $sysinfo{enclosure}{$id}->{id}, $sysinfo{enclosure}{$id}->{name},
4073 $sysinfo{enclosure}{$id}->{firmware};
4077 foreach my $line (@storageprint) {
4078 print $linebreak, $line;
4083 if ($opt{okinfo} >= 3) {
4084 print "$linebreak----- OpenManage Server Administrator (OMSA) version: '$sysinfo{om}'";
4089 if ($opt{extinfo}) {
4091 if (defined $opt{htmlinfo}) {
4092 printf '------ SYSTEM: <a href="%s">%s</a>, SN: <a href="%s">%s</a>',
4093 documentation_url($sysinfo{model}), $sysinfo{model},
4094 warranty_url($sysinfo{serial}), $sysinfo{serial};
4097 printf '------ SYSTEM: %s, SN: %s',
4098 $sysinfo{model}, $sysinfo{serial};
4101 if (defined $opt{postmsg}) {
4103 if (-f $opt{postmsg}) {
4104 open my $POST, '<', $opt{postmsg}
4105 or ( print $linebreak
4106 and print "ERROR: Couldn't open post message file $opt{postmsg}: $!\n"
4107 and exit $E_UNKNOWN );
4113 $post = $opt{postmsg};
4115 if (defined $post) {
4117 $post =~ s{[%]s}{$sysinfo{serial}}gxms;
4118 $post =~ s{[%]m}{$sysinfo{model}}gxms;
4119 $post =~ s{[%]b}{$sysinfo{bios}}gxms;
4120 $post =~ s{[%]d}{$sysinfo{biosdate}}gxms;
4121 $post =~ s{[%]o}{$sysinfo{osname}}gxms;
4122 $post =~ s{[%]r}{$sysinfo{osver}}gxms;
4123 $post =~ s{[%]p}{$count{pdisk}}gxms;
4124 $post =~ s{[%]l}{$count{vdisk}}gxms;
4125 $post =~ s{[%]n}{$linebreak}gxms;
4126 $post =~ s{[%]{2}}{%}gxms;
4132 # Print any perl warnings that have occured
4133 if (@perl_warnings) {
4134 foreach (@perl_warnings) {
4136 print "${linebreak}INTERNAL ERROR: @$_";
4138 $exit_code = $E_UNKNOWN;
4141 # Reset the WARN signal
4142 $SIG{__WARN__} = $original_sigwarn;
4144 # Print performance data
4145 if (defined $opt{perfdata} && !$opt{debug} && %perfdata) {
4146 my $lb = $opt{perfdata} eq 'multiline' ? "\n" : q{ }; # line break for perfdata
4157 return ($order{(split /_/, $a, 2)[0]} cmp $order{(split /_/, $b, 2)[0]}) || $a cmp $b;
4160 print join $lb, map { "'$_'=$perfdata{$_}" } sort perfdata keys %perfdata;
4163 # Print a linebreak at the end
4164 print "\n" if !$opt{debug};
4166 # Exit with proper exit code