| 1 |
--- smeserver-zabbix-agent-0.1/root/etc/e-smith/templates/etc/zabbix/zabbix_agentd.conf/90UserParameters_swRaid.mdstat_one_line 2009-04-24 15:45:42.000000000 +0200 |
| 2 |
+++ smeserver-zabbix-agent-0.1/root/etc/e-smith/templates/etc/zabbix/zabbix_agentd.conf/90UserParameters_swRaid 2009-04-24 18:53:03.000000000 +0200 |
| 3 |
@@ -8,11 +8,9 @@ |
| 4 |
# Show Value: As is |
| 5 |
|
| 6 |
# The value reported is like: |
| 7 |
-# OK - md3 [UU] has 2 of 2 devices active (active=sdc1,sdd1 failed=none spare=none) |
| 8 |
-# CRITICAL - md2 [_U] has 1 of 2 devices active (active=sdb2 failed=sda2 spare=none) |
| 9 |
-# OK - md1 [UU] has 2 of 2 devices active (active=sda1,sdb1 failed=none spare=none) |
| 10 |
+# OK: md3:raid1:2 drives:931GB:Optimal md2:raid1:2 drives:931GB:Optimal md1:raid1:2 drives:101MB:Optimal |
| 11 |
|
| 12 |
# Tips: You can add a simple trigger on this check like: |
| 13 |
-# \{ hostname:raid.sw.status.str( CRITICAL ) \}=1 |
| 14 |
+# \{ hostname:raid.sw.status.str( OK ) \}=0 |
| 15 |
UserParameter=raid.sw.status,/var/lib/zabbix/bin/mdstat-parser.pl |
| 16 |
|
| 17 |
--- smeserver-zabbix-agent-0.1/root/var/lib/zabbix/bin/mdstat-parser.pl.mdstat_one_line 2009-04-24 15:45:42.000000000 +0200 |
| 18 |
+++ smeserver-zabbix-agent-0.1/root/var/lib/zabbix/bin/mdstat-parser.pl 2009-04-24 18:48:28.000000000 +0200 |
| 19 |
@@ -3,6 +3,10 @@ |
| 20 |
# Get status of Linux software RAID for SNMP / Nagios |
| 21 |
# Author: Michal Ludvig <michal@logix.cz> |
| 22 |
# http://www.logix.cz/michal/devel/nagios |
| 23 |
+ |
| 24 |
+# Slightly modified by Daniel B. for integration on SME Server / Zabbix |
| 25 |
+# 24 Apr 2009 |
| 26 |
+ |
| 27 |
# |
| 28 |
# Simple parser for /proc/mdstat that outputs status of all |
| 29 |
# or some RAID devices. Possible results are OK and CRITICAL. |
| 30 |
@@ -51,16 +55,20 @@ |
| 31 |
## This is a global return value - set to the worst result we get overall |
| 32 |
my $retval = 0; |
| 33 |
|
| 34 |
-my (%active_devs, %failed_devs, %spare_devs); |
| 35 |
+my (%active_devs, %failed_devs, %spare_devs, %devs_total, %level, %size, %status); |
| 36 |
+my @raids; |
| 37 |
+my $result = 'OK'; |
| 38 |
|
| 39 |
open FILE, "< $file" or die "Can't open $file : $!"; |
| 40 |
while (<FILE>) { |
| 41 |
next if ! /^(md\d+)+\s*:/; |
| 42 |
next if $device ne "all" and $device ne $1; |
| 43 |
my $dev = $1; |
| 44 |
+ push @raids, $dev; |
| 45 |
|
| 46 |
my @array = split(/ /); |
| 47 |
for $_ (@array) { |
| 48 |
+ $level{$dev} = $1 if /^(raid\d+)$/; |
| 49 |
next if ! /(\w+)\[\d+\](\(.\))*/; |
| 50 |
if ($2 eq "(F)") { |
| 51 |
$failed_devs{$dev} .= "$1,"; |
| 52 |
@@ -80,18 +88,32 @@ |
| 53 |
else { $failed_devs{$dev} =~ s/,$//; } |
| 54 |
|
| 55 |
$_ = <FILE>; |
| 56 |
- /\[(\d+)\/(\d+)\]\s+\[(.*)\]$/; |
| 57 |
- my $devs_total = $1; |
| 58 |
- my $devs_up = $2; |
| 59 |
- my $stat = $3; |
| 60 |
- my $result = "OK"; |
| 61 |
- if ($devs_total > $devs_up or $failed_devs{$dev} ne "none") { |
| 62 |
+ /(\d+)\ blocks\ \[(\d+)\/(\d+)\]\s+\[(.*)\]$/; |
| 63 |
+ $size{$dev} = int($1/1024); |
| 64 |
+ if ($size{$dev} > 1024){ |
| 65 |
+ $size{$dev} = int($size{$dev}/1024)."GB"; |
| 66 |
+ } |
| 67 |
+ else{ |
| 68 |
+ $size{$dev} .= "MB"; |
| 69 |
+ } |
| 70 |
+ $devs_total{$dev} = $2; |
| 71 |
+ my $devs_up = $3; |
| 72 |
+ my $stat = $4; |
| 73 |
+ if ($devs_total{$dev} > $devs_up or $failed_devs{$dev} ne "none") { |
| 74 |
+ $status{$dev} = "Degraded"; |
| 75 |
$result = "CRITICAL"; |
| 76 |
$retval = $ERRORS{"CRITICAL"}; |
| 77 |
} |
| 78 |
+ else{ |
| 79 |
+ $status{$dev} = "Optimal"; |
| 80 |
+ } |
| 81 |
|
| 82 |
- print "$result - $dev [$stat] has $devs_up of $devs_total devices active (active=$active_devs{$dev} failed=$failed_devs{$dev} spare=$spare_devs{$dev})\n"; |
| 83 |
} |
| 84 |
+print "$result: "; |
| 85 |
+foreach my $raid (@raids){ |
| 86 |
+ print "$raid:$level{$raid}:$devs_total{$raid} drives:$size{$raid}:$status{$raid} "; |
| 87 |
+} |
| 88 |
+print "\n"; |
| 89 |
close FILE; |
| 90 |
exit $retval; |
| 91 |
|
Parent Directory
| 
Revision Log
| 
Revision Graph
