contrib-munin/plugins/other/raid

#!/usr/bin/perl -w
# 
# (c) 2007 Nathan Rutman nathan@clusterfs.com
#
# Plugin to monitor RAID status 
#
# Results are % of healthy drives in a raid device
# and % rebuilt of devices that are resyncing. 
#
#%# family=contrib
#%# capabilities=autoconf

if ($ARGV[0] and $ARGV[0] eq "autoconf")
{
    if (-r "/proc/mdstat" and `grep md /proc/mdstat`)
    {
	print "yes\n";
	exit 0;
    }
    else
    {
	print "no RAID devices\n";
	exit 1;
    }
}

if ( $ARGV[0] and $ARGV[0] eq "config" )
{
    print "graph_title RAID status\n";
    print "graph_category disk\n";
    print "graph_info This graph monitors RAID disk health.  Values are percentage of healthy drives in each raid group.  Degraded devices are marked Critical.\n";
    print "graph_args --base 1000 -l 0\n";
    print "graph_vlabel % healthy/rebuilt\n";
    print "graph_scale  no\n";
}

{
    local( $/, *MDSTAT ) ;
    open (MDSTAT, "/proc/mdstat") or exit 1;
    #open (MDSTAT, "/etc/munin/plugins/sample.failed") or exit 1;
    my $text = <MDSTAT>;
    close MDSTAT;

    while ($text =~ /(md\d)\s+:\s+active\s+(\w+)\s+(.*)\n.*\[(\d+)\/(\d+)]\s+\[(\w+)]/ )
    {
	my($dev,$type,$members,$nmem,$nact,$status) = ($1,$2,$3,$4,$5,$6);
	#print "item: $dev $type ($members) status=$status \n";
	if ( $ARGV[0] and $ARGV[0] eq "config" ) 
	{
	    print "$dev.label $dev\n";
	    print "$dev.info $type $members\n";
	    # 100: means less than 100
	    print "$dev.critical 100:\n";	
	    print $dev, "_rebuild.label $dev rebuilt\n";
	    print $dev, "_rebuild.info $type\n";
	    print $dev, "_rebuild.critical 100:\n";	
	} 
	else
	{
	    my $pct = 100 * $nact / $nmem;
	    my $rpct = 100;
	    if ( $pct < 100 )
	    {
			my $output = `/sbin/mdadm -D /dev/$dev | grep Rebuild`;
			if( $output =~ /([0-9]+)% complete/ )
			{
				$rpct = $1;
			}
			else
			{
				$rpct = 0;
			}
	    }
	    print "$dev.value $pct\n";
	    print $dev, "_rebuild.value $rpct\n";
	}
	$text = $';
    } 
}

exit 0;