#! /bin/bash
#
# smart_spindown rev. 1
#
# Copyright (C) 2003 by Bart Samwel
#
# You may do with this file (and parts thereof) whatever you want, as long
# as my copyright notice is retained.
#
# Extended by joerk at gentoo-wiki.com
# modified for a mostly idle backup disk by B Persson
#
#
# A command line parameter --conf=<pathname> is required. The specified file
# will be sourced as shell code. It must set the variable DISK to the pathname
# of the disk to be monitored. It is recommended to use one of the links under
# /dev/disk.
#
# How it works: This program monitors the activity on a disk. If there
# is no activity for a while, the disk is spun down. The time without
# activity that is required is dynamic, using a backoff factor. When
# the recent spun-down periods are relatively short, this means that the
# machine might be busy with something, so the script tries to wait for
# longer periods without activity before spinning down again. When spun-down
# periods are long, the backoff factor is decreased, and the disk is spun
# down after shorter periods without activity.


#
# Configuration
#

# Output levels. Level 2 is verbose, level 1 is normal output.
# Enable all levels you would like to see.
OUTLEVEL1=true
OUTLEVEL2=false

# Decide which output to use. Useful if run in daemon mode
# echo or logger
# output1 is logged to normal, output2 to debug syslog
# Is PID of this shell logged?
#OUTPUT1=echo
#OUTPUT2=echo
OUTPUTTAG="$(basename -- $0)[$$]"
OUTPUT1="logger -t $OUTPUTTAG -p user.notice --"
OUTPUT2="logger -t $OUTPUTTAG -p user.debug  --"

# Multiplication factor for the backoff after a spinup, in percentages.
# Default is 300 = factor 3.
BACKOFF_INCREASE_PCT=300

# Multiplication factor for the backoff at every poll that shows that
# the disk is spun down. This determines how fast the backoff value
# decreases.
BACKOFF_DECREASE_PCT=96

# The base inactivity wait time (in seconds). This is multiplied by
# the backoff factor to determine the real inactivity wait time.
WAITTIME=150

# The maximum inactivity wait time (in seconds).
# This also limits the backoff factor: the backoff factor cannot increase
# above a value that makes the inactivity wait time larger than MAXWAIT.
# Something, probably in Linux, seems to read a few disk sectors every ten
# minutes while the disk is spinning, which prevents this program from ever
# spinning down the disk once the inactivity wait time grows longer than ten
# minutes. To avoid this, MAXWAIT is set to nine minutes.
MAXWAIT=540

# Time (in seconds) between polls to see if the disk is active again.
# Default is 10 seconds.
POLLTIME=10

# Enable this if you don't use laptop_mode. This will make the script
# sync before spinning down the disk.
NO_LAPTOP_MODE=true


#
# Parse the command line:
#

if [[ $# -ne 1 ]] ; then
	echo 'Exactly one parameter is required, namely --conf=<pathname>.' >&2
	exit 1
fi

case "$1" in
	--conf=*)
		CONF="${1#*=}"
	;;
	*)
		echo 'unrecognized parameter' >&2
		exit 1
	;;
esac


#
# Read the configuration:
#

source "$CONF"
if [[ $? -ne 0 ]] ; then
	echo 'Reading the configuration file failed.' >&2
	exit 1
fi

# Get the device name of the disk to monitor:
DEVNAME=`realpath "$DISK"`
if [[ $? -ne 0 ]] ; then
	echo 'bad disk pathname' >&2
	exit 1
fi
if [[ ! -b "$DEVNAME" ]] ; then
	echo 'The disk pathname is not a block device.' >&2
	exit 1
fi

# Stats file: the file used to monitor the disk's activity.
STATSFILE=/sys/block/`basename "$DEVNAME"`/stat


#
# Let's go!
#

# Number of poll times that the disk was found to be spun down.
POLLSSPUNDOWN=0

# Number of spindowns performed
SPINDOWNS=0

# Number of times (*100) the WAITTIME of inactivity required before spindown
BACKOFF_FACTOR=100

# Stats: Total time the disk has been up.
UPTIME=0

# Total duration of last spun-down period.
LASTDOWNTIME=0

# Total duration of the last spun-up period.
LASTUPTIME=0

# Duration of the last poll. Always equal to POLLTIME except the first
# time around.
LASTPOLLTIME=0

# Make sure the stuff we use is in the cache. I've seen it happen
# that the script spun the disk down, and then "sleep" wasn't in
# the cache and the disk spun right up again. :)
true
false
sleep 1

# Terminate on request.
trap 'exit 0' HUP INT TERM

# Log the end of script execution.
trap "$OUTPUT1 'Exiting.'" EXIT


$OUTLEVEL1 && ${OUTPUT1} "Monitoring spindown opportunities for $DEVNAME."
if ($OUTLEVEL1) ; then
	hdparm -C $DEVNAME |grep active >/dev/null
	if [ $? -eq 0 ] ; then
		${OUTPUT1} "$DEVNAME is currently spun up." ;
	else
		${OUTPUT1} "$DEVNAME is currently spun down." ;
	fi ;
fi
while [[ true ]]; do
	hdparm -C $DEVNAME |grep active >/dev/null
	if [ $? -eq 0 ] ; then
		THISWAIT=$(($WAITTIME*$BACKOFF_FACTOR/100)) ;
		if [[ $THISWAIT -gt $MAXWAIT ]] ; then
			THISWAIT=$MAXWAIT ;
		fi ;
		# Increase the backoff irrespective of whether we failed
		# or not. The backoff should drop again by the lack of
		# spinups afterwards.
		BACKOFF_FACTOR=$(($BACKOFF_FACTOR*$BACKOFF_INCREASE_PCT/100)) ;
		if [[ $(($BACKOFF_FACTOR*$WAITTIME/100)) -gt $MAXWAIT ]] ; then
			BACKOFF_FACTOR=$(($MAXWAIT*100/$WAITTIME)) ;
		fi ;
		if ( $OUTLEVEL2 ) ; then
			# UPTIME is used only for OUTLEVEL2 messages.
			UPTIME=$(($UPTIME+$LASTPOLLTIME)) ;
		fi ;
		LASTUPTIME=$(($LASTUPTIME+$LASTPOLLTIME)) ;
		if [ $LASTPOLLTIME -ne 0 ] ; then
			$OUTLEVEL1 && ${OUTPUT1} "$DEVNAME spun up after $LASTDOWNTIME seconds." ;
		fi
		NUM_EQUALS=0 ;
		$OUTLEVEL2 && ${OUTPUT2} "Waiting for $THISWAIT seconds of inactivity..." ;
		PREVIOUS_STATS=`cat $STATSFILE` ;
		while [[ $(($NUM_EQUALS*5)) -lt $THISWAIT ]]; do
			sleep 5 ;
			if ( $OUTLEVEL2 ) ; then
				# UPTIME is used only for OUTLEVEL2 messages.
				UPTIME=$(($UPTIME+5)) ;
			fi ;
			LASTUPTIME=$(($LASTUPTIME+5)) ;
			NEXT_STATS=`cat $STATSFILE` ;
			if [ "$PREVIOUS_STATS" != "$NEXT_STATS" ] ; then
				NUM_EQUALS=0 ;
				PREVIOUS_STATS="$NEXT_STATS"
				$OUTLEVEL2 && ${OUTPUT2} "Restarting..." ;
			else
				NUM_EQUALS=$(($NUM_EQUALS+1)) ;
				$OUTLEVEL2 && ${OUTPUT2} "Seconds of quiet: $(($NUM_EQUALS*5))" ;
			fi
		done
		# We've just had $THISWAIT seconds of inactivity. The inactivity indicates
		# that we're ready to go to sleep. Laptop mode will have synced all
		# writes for us after the last read, so we don't have to explicitly
		# sync.
		if ( $NO_LAPTOP_MODE ) ; then
			sync ;
		fi ;
		hdparm -q -y $DEVNAME ;
		if ( $OUTLEVEL2 ) ; then
			# SPINDOWNS is used only for OUTLEVEL2 messages.
			SPINDOWNS=$(($SPINDOWNS+1)) ;
		fi ;
		$OUTLEVEL1 && ${OUTPUT1} "$DEVNAME spun down after $LASTUPTIME seconds (with $THISWAIT seconds of inactivity)." ;
		LASTUPTIME=0 ;
		LASTDOWNTIME=0 ;
	else
		if ( $OUTLEVEL2 ) ; then
			# POLLSSPUNDOWN and SPINDOWNS are used only for OUTLEVEL2 messages.
			POLLSSPUNDOWN=$(($POLLSSPUNDOWN+1)) ;
			if [[ $SPINDOWNS -eq 0 ]] ; then
				SPINDOWNS=1 ;
			fi
		fi ;
		LASTDOWNTIME=$(($LASTDOWNTIME+$LASTPOLLTIME)) ;
		BACKOFF_FACTOR=$(($BACKOFF_FACTOR*$BACKOFF_DECREASE_PCT/100)) ;
		if [ $BACKOFF_FACTOR -lt 100 ] ; then
			BACKOFF_FACTOR=100 ;
		fi
	fi ;
	if ( $OUTLEVEL2 ) ; then
		${OUTPUT2} "spindowns: $SPINDOWNS, time up/down: $UPTIME/$(($POLLSSPUNDOWN*$POLLTIME)), backoff $BACKOFF_FACTOR, down for $LASTDOWNTIME (avg $(($POLLSSPUNDOWN*$POLLTIME/$SPINDOWNS)))." ;
	fi ;
	sleep $POLLTIME ;
	LASTPOLLTIME=$POLLTIME ;
done