#!/bin/bash

#====================================================================================
#:TITLE                 : check_megasasctl
#:TYPE                  : Nagios Plugin Script
#:AUTHOR                : r0dr1gu3z (jrod@automattic.com)
#:COMPANY               : Automattic, Inc.
#:VERSION               : 0.0.1
#:CREATED               : May 13, 2010
#:DESCRIPTION           : This script checks the local output of megasasctl under /tmp/raidstatus.txt for 
#                         the status of each raid array, drive (online status and Media Error Counts), and BBU status.
#:PARMS                 : N/A
#:OPTIONS               : N/A
#:NOTES                 : Assoc. w/: command[check_megasasctl]=/usr/local/nagios/plugins/check_megasasctl
#:ROLE ASSOC.           : base
#====================================================================================

#====================================================================================
# Variable Defs:
#====================================================================================
SCRIPT=${0##*/}
NORMALRAIDSTATUS="optimal"
NORMALDISKSTATUS="online"
NORMALBBUSTATUS="good"
CURRENTRAIDSTATUS=""
CURRENTDISKSTATUS=""
BADRAIDARRAYS=()
BADDISKARRAY=()
BADBBUARRAY=()
STATUS="OK"
MSG=""
MEDIATHRESHOLD=20
MAXMEDIATHRESHOLD=30
HIGH_MEDIAERRORLIST=()
MED_MEDIAERRORLIST=()
MEGASASCTL="/usr/local/bin/megasasctl"
TEMPRAIDSTATUS="/tmp/raidstatus.txt"

#====================================================================================
# Function Defs:
#====================================================================================

# Function to Check Raid Arrays
function check_raid_array_status() {
	# Get list of RAID Arrays
	RAIDLIST=$( grep -v "bios" $TEMPRAIDSTATUS | grep "RAID" | grep '.' | sed 's/  */\ /g' | cut -d' ' -f1);
	# Next go through raid array list and check each one for its status
	for i in $RAIDLIST; do
		# Log list of degraded raid arrays
		CURRENTRAIDSTATUS=$( grep "${i} " $TEMPRAIDSTATUS | grep "RAID" | sed 's/  */\ /g'| cut -d' ' -f6 );
		# If a RAID Array is found to be be less than optimal then record it. 
		if [ $CURRENTRAIDSTATUS != $NORMALRAIDSTATUS ] ; then 
				BADRAIDARRAYS=( "${BADRAIDARRAYS[@]}" "${i}" )
		fi
	done
	# Check if the number of degraded arrays is greater than 0
	if [ ${#BADRAIDARRAYS[@]} != 0 ] ; then
		# Set Status = Critical
		STATUS="CRITICAL"
		# Report which arrays degraded arrays were found 
		MSG="Degraded RAID ARRAY(s) found: ${BADRAIDARRAYS[@]}"
	else
		MSG="RAID ARRAYS OPTIMAL, "
	fi
}

# Function to Check All Disk Online Status - check all disks that do not have an status of "online"
function check_all_disk_online_status() {
	# Get list of Disks
	DISKLIST=$( grep -v "bios\|RAID\|row" $TEMPRAIDSTATUS | grep '.' | sed 's/  */\ /g'| cut -d' ' -f1 );
	# Next go through raid array list and check each one for their status
	for i in $DISKLIST; do
	# Log list of degraded disks
		CURRENTDISKSTATUS=$( grep "${i} " $TEMPRAIDSTATUS | grep -v "row" | sed 's/  */\ /g'| cut -d' ' -f6 );
		if [ $CURRENTDISKSTATUS != $NORMALDISKSTATUS ] ; then 
			BADDISKARRAY=( "${BADDISKARRAY[@]}" "${i}" )
		fi
	done
	# Set status information 
	if [ ${#BADDISKARRAY[@]} != 0 ] ; then
		# Set Status = Critical
		STATUS="CRITICAL"
		# Report which arrays degraded disks were found 
		MSG="Degraded Disks found: ${BADDISKARRAY[@]}"
	else
		MSG="${MSG} ALL DISKS ONLINE, "
	fi
}

# Function to check media error counts (M.E.C's) across all disks
function check_all_media_error_status() {
	# Get list of Disks
	DISKLIST=$( grep -v "row\|bios\|RAID" $TEMPRAIDSTATUS | grep "errs:" | sed 's/  */\ /g' |  grep '.' | cut -d' ' -f1 );
	# Next go through all disks and group by M.E.C 
	for i in $DISKLIST; do
		DISKMEDIAERRORCOUNT=$( grep "$i " $TEMPRAIDSTATUS | grep -v "row" | sed 's/  */\ /g' | cut -d: -f3 | cut -d' ' -f1 )
		# Check to see if M.E.C's are within thresholds
		# if the number M.E.C is greater then the max threshold record it
		if [ $DISKMEDIAERRORCOUNT -gt $MAXMEDIATHRESHOLD ] ; then
			HIGH_MEDIAERRORLIST=( "${HIGH_MEDIAERRORLIST[@]}" "$i:${DISKMEDIAERRORCOUNT}" )
		# else check if to see if M.E.C's are within the allowed thresholds and record it. 
		elif [ $DISKMEDIAERRORCOUNT -gt $MEDIATHRESHOLD ] && [ $DISKMEDIAERRORCOUNT -lt $MAXMEDIATHRESHOLD ] ; then
			MED_MEDIAERRORLIST=( "${MED_MEDIAERRORLIST[@]}" "$i:${DISKMEDIAERRORCOUNT}" )
		fi
	done
	# Set status information 
	if [ ${#HIGH_MEDIAERRORLIST[@]} != 0 ] ; then
		# Set Status = CRITICAL
		STATUS="CRITICAL"
		# Report which disks were found w/ high MEC 
		MSG="High M.E.C. on: ${HIGH_MEDIAERRORLIST[@]}"
	elif [ ${#MED_MEDIAERRORLIST[@]} != 0 ] ; then
		# Set Status = WARNING
		STATUS="WARNING"
		# Report which disks were found w/ med MEC 
		MSG="M.E.C. Warning on the following disks: ${MED_MEDIAERRORLIST[@]}"
	else
		MSG="${MSG} ALL DISKS OPTIMAL, "
	fi
}

# Function to check the BBU status across all adapters
function check_adapt_bbu_status() {
	# Get Adapter Count
	ADAPTCOUNT=$( grep -c "bios" $TEMPRAIDSTATUS )
	# Go through each Adapter and log bad BBUs
	for ((i=0; i<$ADAPTCOUNT; i++)) ; do
		# Grab the BBU status for each adaptoer 
		BBUSTATUS=$( grep "a${i} " $TEMPRAIDSTATUS | sed 's/  */\ /g' | cut -d: -f7 | cut -d/ -f1 );
		if [ "$BBUSTATUS" != "$NORMALBBUSTATUS" ] ; then 
			BADBBUARRAY=( "${BADBBUARRAY[@]}" "a$i" )
		fi
	done
	# Check check to list of BBU's to see if the number is greater than 0
	if [ ${#BADBBUARRAY[@]} != 0 ] ; then
		# Set Status = Warning
		STATUS="WARNING"
		# report which bad BBU's were found 
		MSG="Bad BBU(s) found: ${BADBBUARRAY[@]}"
	else
		MSG="${MSG} ALL BBU(s) OK"
	fi
}

# Function to check status as we go
function check_status() {
	if [ $STATUS = "CRITICAL" ] ; then 
		echo "$STATUS: $MSG"
		exit 2
	elif [ $STATUS = "WARNING" ] ; then 
		echo "$STATUS: $MSG"
		exit 1
	fi
}

#====================================================================================
# MAIN:
#====================================================================================

# First, capture megasasctl tmp file
${MEGASASCTL} -v > $TEMPRAIDSTATUS
if [ ! $? -eq 0 ] && [ ! -e ${TEMPRAIDSTATUS} ] ; then 
	echo "UNKONWN: Error running ${SCRIPT}. Please check";
	exit 2;
fi

# Check the the status of the RAID Array
check_raid_array_status
check_status

# Check the status of each disk on every array, 
check_all_disk_online_status
check_status


# Check for Media Errors
check_all_media_error_status
check_status


# Check BBU status across all adapters
check_adapt_bbu_status
check_status

# Cleanup: remove tmp raidstatus.txt file
rm -rf $TEMPRAIDSTATUS

# Check final script status
if [ $STATUS = "OK" ] ; then 
	echo "$STATUS: $MSG"
	exit 0 
fi