#! /bin/sh

# Author: Petter Reinholdtsen <pere@hungry.com>
# License: GNU General Public License v2 or later

# mpt-statusd - Check mpt-status values in the background.
#
# chkconfig:   345 60 50
# description: Check mpt-status values in the background \
# using daemonize and the mpt-status utility.

### BEGIN INIT INFO
# Provides:          mpt-statusd
# Required-Start:    $remote_fs $syslog
# Required-Stop:     $remote_fs $syslog
# Default-Start:     2 3 4 5
# Default-Stop:      0 1 6
# Short-Description: Check mpt-status values in the background.
### END INIT INFO

PATH=/sbin:/bin:/usr/sbin:/usr/bin
DESC="mpt-status monitor"
NAME=mpt-statusd
PIDFILE=/var/run/$NAME.pid
STATUSFILE=/var/run/$NAME.status
SCRIPTNAME=/etc/init.d/$NAME


# Do not touch you can configure this in /etc/default/mpt-statusd
MAILTO=root   # Where to report problems
PERIOD=600    # Seconds between each check    (default 10 minutes)
REMIND=7200   # Seconds between each reminder (default 2 hours)
RUN_DAEMON=yes
ID=0

[ -e /etc/default/mpt-statusd ] && . /etc/default/mpt-statusd

# Gracefully exit if the package has been removed.
test -x /usr/sbin/mpt-status || exit 0

# Source function library.
. /etc/rc.d/init.d/functions
. /lib/lsb/init-functions

if [ $RUN_DAEMON = "no" ] ; then
	log_failure_msg $"mpt-statusd is disabled in /etc/default/mpt-statusd, not starting."
	exit 0
fi

if ! [ -e "/proc/mpt/version" ] ; then
  	log_failure_msg "The mptctl module is missing."
	exit 0
fi

check_mpt() {
    echo $$ > $PIDFILE.new && mv $PIDFILE.new $PIDFILE
    while true ; do
        # Check ever $PERIOD seconds, send email on every status
        # change and repeat ever $REMIND seconds if the raid is still
        # bad.
        if (mpt-status -i $ID) |grep -q 'state OPTIMAL' ; then
            BADRAID=false
        else
            BADRAID=true
            logger -t mpt-statusd "detected non-optimal RAID status"
        fi
        STATUSCHANGE=false
        if [ true = "$BADRAID" ] ; then
            # RAID not OK
            (mpt-status -i $ID) > $STATUSFILE.new
            if [ ! -f $STATUSFILE ] ; then # RAID just became broken
                STATUSCHANGE=true
                mv $STATUSFILE.new $STATUSFILE
            elif cmp -s $STATUSFILE $STATUSFILE.new ; then
                # No change.  Should we send reminder?
                LASTTIME="`stat -c '%Z' $STATUSFILE`"
                NOW="`date +%s`"
                SINCELAST="`expr $NOW - $LASTTIME`"
                if [ $REMIND -le "$SINCELAST" ]; then
                    # Time to send reminder
                    STATUSCHANGE=true
                    mv $STATUSFILE.new $STATUSFILE
                else
                    rm $STATUSFILE.new
                fi
            else
                STATUSCHANGE=true
                mv $STATUSFILE.new $STATUSFILE
            fi
        else
            # RAID OK
            if [ -f $STATUSFILE ] ; then
                rm $STATUSFILE
                STATUSCHANGE=true
            fi
        fi
                
        if [ true = "$STATUSCHANGE" ]; then
            hostname="`uname -n`"
            (
                cat <<EOF 
This is a RAID status update from mpt-statusd.  The mpt-status
program reports that one of the RAIDs changed state:

EOF
                if [ -f $STATUSFILE ] ; then
                    cat $STATUSFILE
                else
                    (mpt-status -i $ID)
                fi
                echo
                echo "Report from $0 on $hostname"
            ) | mail -s "info: mpt raid status change on $hostname" $MAILTO
        fi
        sleep $PERIOD
    done
}

check_daemon() {
        # Let's check if there is a daemon which is really running and not timing out
        DAEMON_RUN=`ps aux | grep "/etc/init.d/mpt-statusd check_mpt" | grep -v grep | grep -v daemon`
        if [ -n "$DAEMON_RUN" ] ; then
                return 1;
        else
                return 0;
        fi
}

#
#       Function that starts the daemon/service.
#
d_start() {
    [ -f $PIDFILE ] && PID="`cat $PIDFILE`"
    if [ "$PID" ] ; then
        log_warning_msg "Daemon already running. Refusing to start another"
        return 0
    elif check_daemon ; then
	# Use daemonize to turn it into a daemon and start it with daemon().
        daemon --pidfile $PIDFILE /usr/sbin/daemonize $SCRIPTNAME check_mpt
	return 0
    else
        log_warning_msg "Daemon is already running. Refusing to start another"
        return 0
    fi
}

#
#       Function that stops the daemon/service.
#
d_stop() {
        if [ -f $PIDFILE ] ; then
             	killproc -p $PIDFILE $SCRIPTNAME > /dev/null 2>&1 
		rm -f $PIDFILE
        else
                log_warning_msg "Daemon is already stopped."
                return 0
        fi
}

# This is a workaround function which does not directly exit and
# therefore can be used by a restart
d_stop_by_restart() {
        if [ -f $PIDFILE ] ; then
                killproc -p $PIDFILE $SCRIPTNAME > /dev/null 2>&1 
                rm -f $PIDFILE
        else
                log_warning_msg "Daemon is already stopped."
        fi
}

case "$1" in
  start)
        echo -n ""
        echo -n $"Starting $DESC: $NAME"
        d_start
        ;;
  stop)
        echo -n $"Stopping $DESC: $NAME"
        d_stop
        ;;
  check_mpt)
        check_mpt
        ;;
  restart|force-reload)
        echo -n "Restarting $DESC: $NAME"
        d_stop_by_restart
        sleep 1
        d_start
        ;;
  *)
        echo "Usage: $SCRIPTNAME {start|stop|restart|force-reload}" >&2
        exit 1
        ;;
esac

exit 0