From 87533869e98944b116344f0650fc392d149dbf5d Mon Sep 17 00:00:00 2001 From: Jan Wagner Date: Fri, 8 Mar 2024 13:06:59 +0000 Subject: [PATCH] check_smart: Update to 6.14.1 --- check_smart/check_smart | 51 +++++++++++++++++++++++++++++++++++------ check_smart/control | 2 +- 2 files changed, 45 insertions(+), 8 deletions(-) diff --git a/check_smart/check_smart b/check_smart/check_smart index 7a392d6..f2b10dc 100644 --- a/check_smart/check_smart +++ b/check_smart/check_smart @@ -55,13 +55,18 @@ # Dec 16, 2021: Lorenz Kaestle - Bugfix when interface parameter was missing in combination with -g (6.12.2) # Apr 27, 2022: Claudio Kuenzler - Allow skip temperature check (--skip-temp-check) (6.13.0) # Apr 27, 2022: Peter Newman - Better handling of missing or non-executable smartctl command (6.13.0) +# Apr 29, 2023: Nick Bertrand - Show drive(s) causing UNKNOWN status using -g/--global check (6.14.0) +# Apr 29, 2023: Claudio Kuenzler - Add possibility to hide serial number (--hide-sn) (6.14.0) +# Apr 29, 2023: Claudio Kuenzler - Add default check on Load Cycle Count (ignore using --skip-load-cycles) (6.14.0) +# Sep 20, 2023: Yannick Martin - Fix default Percent_Lifetime_Remain threshold handling when -w is given (6.14.1) +# Sep 20, 2023: Claudio Kuenzler - Fix debug output for raw check list, fix --hide-serial in debug output (6.14.1) use strict; use Getopt::Long; use File::Basename qw(basename); my $basename = basename($0); -my $revision = '6.13.0'; +my $revision = '6.14.1'; # Standard Nagios return codes my %ERRORS=('OK'=>0,'WARNING'=>1,'CRITICAL'=>2,'UNKNOWN'=>3,'DEPENDENT'=>4); @@ -71,7 +76,7 @@ my @sys_path = qw(/usr/bin /bin /usr/sbin /sbin /usr/local/bin /usr/local/sbin); $ENV{'BASH_ENV'}=''; $ENV{'ENV'}=''; -use vars qw($opt_b $opt_d $opt_g $opt_debug $opt_h $opt_i $opt_e $opt_E $opt_r $opt_s $opt_v $opt_w $opt_q $opt_l $opt_skip_sa $opt_skip_temp); +use vars qw($opt_b $opt_d $opt_g $opt_debug $opt_h $opt_i $opt_e $opt_E $opt_r $opt_s $opt_v $opt_w $opt_q $opt_l $opt_skip_sa $opt_skip_temp $opt_skip_load_cycles $opt_hide_sn); Getopt::Long::Configure('bundling'); GetOptions( "debug" => \$opt_debug, @@ -90,6 +95,8 @@ GetOptions( "l" => \$opt_l, "ssd-lifetime" => \$opt_l, "skip-self-assessment" => \$opt_skip_sa, "skip-temp-check" => \$opt_skip_temp, + "skip-load-cycles" => \$opt_skip_load_cycles, + "hide-sn" => \$opt_hide_sn, ); if ($opt_v) { @@ -223,7 +230,6 @@ my @raw_check_list_nvme = split /,/, $raw_check_list_nvme; # warning threshold list (for raw checks) my $warn_list = $opt_w // ''; -$warn_list = $opt_w // 'Percent_Lifetime_Remain=90' if $opt_l; my @warn_list = split /,/, $warn_list; my %warn_list; my $warn_key; @@ -232,6 +238,9 @@ foreach my $warn_element (@warn_list) { ($warn_key, $warn_value) = split /=/, $warn_element; $warn_list{ $warn_key } = $warn_value; } +if ($opt_l && ! exists $warn_list{'Percent_Lifetime_Remain'}) { + $warn_list{'Percent_Lifetime_Remain'} = 90; +} # For backward compatibility, add -b parameter to warning thresholds if ($opt_b) { @@ -242,6 +251,7 @@ my @drives_status_okay; my @drives_status_not_okay; my @drives_status_warning; my @drives_status_critical; +my @drives_status_unknown; my $drive_details; foreach $device ( split("\\|",$device) ){ @@ -274,6 +284,7 @@ foreach $device ( split("\\|",$device) ){ warn "###########################################################\n\n\n" if $opt_debug; my $full_command = "$smart_command -d $interface -Hi $device"; + $full_command = "$smart_command -d $interface -Hi $device -q noserial" if $opt_hide_sn; warn "(debug) executing:\n$full_command\n\n" if $opt_debug; my @output = `$full_command`; @@ -355,8 +366,13 @@ foreach $device ( split("\\|",$device) ){ } if($line =~ /$line_serial_ata(.+)/){ warn "(debug) parsing line:\n$line\n\n" if $opt_debug; - $serial = $1; - $serial =~ s/^\s+|\s+$//g; + if($opt_hide_sn) { + $serial = ""; + warn "(debug) Hiding serial number\n\n" if $opt_debug; + } else { + $serial = $1; + $serial =~ s/^\s+|\s+$//g; + } warn "(debug) found serial number $serial\n\n" if $opt_debug; } if($line =~ /$line_serial_scsi(.+)/){ @@ -459,7 +475,7 @@ foreach $device ( split("\\|",$device) ){ @output = `$full_command`; warn "(debug) output:\n@output\n\n" if $opt_debug; my @perfdata = qw//; - warn "(debug) Raw Check List ATA: $raw_check_list\n" if $opt_debug; + warn "(debug) Raw Check List ATA: @raw_check_list\n" if $opt_debug; warn "(debug) Raw Check List NVMe: $raw_check_list_nvme\n" if $opt_debug; warn "(debug) Exclude List for Checks: ", join(",", @exclude_checks), "\n" if $opt_debug; warn "(debug) Exclude List for Perfdata: ", join(",", @exclude_perfdata), "\n" if $opt_debug; @@ -498,6 +514,19 @@ foreach $device ( split("\\|",$device) ){ warn "(debug) SMART Attribute $attribute_name was set to be ignored\n\n" if $opt_debug; next; } else { + # alert for high load cycles, generally up to 600K cycles are considered safe on HDDs + unless($opt_skip_load_cycles) { + if ($attribute_number == 193) { + if ($raw_value > 600000) { + warn "(debug) $attribute_name is above value considered safe (600K)\n\n" if $opt_debug; + push(@error_messages, "$attribute_name is above 600K load cycles ($raw_value) causing possible performance and durability impact"); + escalate_status('CRITICAL'); + } elsif ($raw_value < 600000 && $raw_value > 550000) { + warn "(debug) $attribute_name is nearing 600K load cycles\n\n" if $opt_debug; + push(@warning_messages, "$attribute_name is soon reaching 600K load cycles ($raw_value) causing possible performance and durability impact soon"); + } + } + } # manual checks on raw values for certain attributes deemed significant if (grep {$_ eq $attribute_name} @raw_check_list) { if ($raw_value > 0) { @@ -733,6 +762,8 @@ foreach $device ( split("\\|",$device) ){ push @drives_status_warning, $status_string; } elsif ($exit_status_local eq 'CRITICAL') { push @drives_status_critical, $status_string; + } elsif ($exit_status_local eq 'UNKNOWN') { + push @drives_status_unknown, $status_string; } } else { @@ -771,6 +802,10 @@ if (scalar(@drives_status_warning) > 0) { push @drives_status_not_okay, @drives_status_warning; } +if (scalar(@drives_status_unknown) > 0) { + push @drives_status_not_okay, @drives_status_unknown; +} + if (@drives_status_not_okay) { push @msg_list, grep { $_ } @drives_status_not_okay; } @@ -804,7 +839,7 @@ sub print_revision { sub print_help { print_revision($basename,$revision); - print "\nUsage: $basename {-d=|-g=} -i=(auto|ata|scsi|3ware,N|areca,N|hpt,L/M/N|aacraid,H,L,ID|cciss,N|megaraid,N) [-r list] [-w list] [-b N] [-e list] [-E list] [--debug]\n\n"; + print "\nUsage: $basename {-d=|-g=} -i=(auto|ata|scsi|3ware,N|areca,N|hpt,L/M/N|aacraid,H,L,ID|cciss,N|megaraid,N) [-r list] [-w list] [-b N] [-e list] [-E list] [-s] [-l] [--debug]\n\n"; print "At least one of the below. -d supersedes -g\n"; print " -d/--device: a physical block device to be SMART monitored, eg /dev/sda. Pseudo-device /dev/bus/N is allowed.\n"; print " -g/--global: a glob pattern name of physical devices to be SMART monitored\n"; @@ -828,6 +863,8 @@ sub print_help { print " -l/--ssd-lifetime: Check attribute 'Percent_Lifetime_Remain' available on some SSD drives\n"; print " --skip-self-assessment: Skip SMART self-assessment health status check\n"; print " --skip-temp-check: Skip temperature comparison current vs. drive max temperature\n"; + print " --skip-load-cycles: Do not alert on high load/unload cycle count (600K considered safe on hard drives)\n"; + print " --hide-sn: Do not show drive serial number in output\n"; print " -h/--help: this help\n"; print " -q/--quiet: When faults detected, only show faulted drive(s) (only affects output when used with -g parameter)\n"; print " --debug: show debugging information\n"; diff --git a/check_smart/control b/check_smart/control index ef7686d..5458959 100644 --- a/check_smart/control +++ b/check_smart/control @@ -1,6 +1,6 @@ Homepage: https://github.com/Napsty/check_smart Uploaders: Jan Wagner Watch: https://raw.githubusercontent.com/Napsty/check_smart/master/check_smart.pl \$revision\ =\ '([0-9.]+)' -Version: 6.13.0 +Version: 6.14.1 Description: plugin to check SMART status of ATA/SCSI/NVMe drives, returning any usable metrics as perfdata. Recommends: perl-modules, monitoring-plugins-common | nagios-plugins-common