#!/bin/sh
#

#  License:      GNU General Public License (GPL)
#  (c) 2017 O. Albrigtsen
#           and Linux-HA contributors
#
# -----------------------------------------------------------------------------
#      O C F    R E S O U R C E    S C R I P T   S P E C I F I C A T I O N
# -----------------------------------------------------------------------------
#
# NAME
#       azure-lb : OCF resource agent script for Azure Load Balancer
#
# Initialization:
: ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat}
. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs

# Defaults
if is_suse_based; then
	OCF_RESKEY_nc_default="/usr/bin/socat"
else
	OCF_RESKEY_nc_default="/usr/bin/nc"
fi

OCF_RESKEY_port_default="61000"

: ${OCF_RESKEY_nc=${OCF_RESKEY_nc_default}}
: ${OCF_RESKEY_port=${OCF_RESKEY_port_default}}

process="$OCF_RESOURCE_INSTANCE"
pidfile="/var/run/$OCF_RESOURCE_INSTANCE.pid"


lb_usage() {
	cat <<END
    usage: $0 (start|stop|validate-all|meta-data|help|usage|monitor)
    $0 manages service that answers Azure Load Balancer health probe requests as a OCF HA resource.
    The 'start' operation starts the instance.
    The 'stop' operation stops the instance.
    The 'monitor' operation reports whether the instance seems to be working
    The 'validate-all' operation reports whether the parameters are valid
END
}

lb_metadata() {
cat <<END
<?xml version="1.0"?>
<!DOCTYPE resource-agent SYSTEM "ra-api-1.dtd">
<resource-agent name="azure-lb" version="1.0">
<version>1.0</version>
<longdesc lang="en">
Resource agent to answer Azure Load Balancer health probe requests
</longdesc>
<shortdesc lang="en">Answers Azure Load Balancer health probe requests</shortdesc>

<parameters>

<parameter name="nc">
<longdesc lang="en">
The full path of the used binary. This can be nc or socat path.
The default is /usr/bin/nc and /usr/bin/socat for SUSE distributions.
</longdesc>
<shortdesc lang="en">Full path of the used binary (nc or socat are allowed)</shortdesc>
<content type="string" default="${OCF_RESKEY_nc_default}"/>
</parameter>

<parameter name="port">
<longdesc lang="en">
Port to listen to.
</longdesc>
<shortdesc lang="en">Listen to port</shortdesc>
<content type="string" default="${OCF_RESKEY_port_default}"/>
</parameter>

</parameters>

<actions>
<action name="start"   timeout="20s" />
<action name="stop"    timeout="20s" />
<action name="monitor" depth="0"  timeout="20s" interval="10s" />
<action name="meta-data"  timeout="5s" />
<action name="validate-all"  timeout="5s" />
</actions>
</resource-agent>
END
exit 0
}

getpid() {
        grep -o '[0-9]*' $1
}

lb_monitor() {
	if test -f "$pidfile"; then
		[ "$__OCF_ACTION" = "stop" ] && level="debug" || level="err"

		if pid=$(getpid "$pidfile") && [ -n "$pid" ]; then
			output=$(kill -s 0 "$pid" 2>&1)
			mon_rc=$?

			[ -n "$output" ] && ocf_log "$level" "$output"
			[ "$mon_rc" -eq 0 ] && return $OCF_SUCCESS
		fi

		# pidfile w/o process means the process died
		return $OCF_ERR_GENERIC
	else
		return $OCF_NOT_RUNNING
	fi
}

lb_start() {
	cmd="$OCF_RESKEY_nc -l -k $OCF_RESKEY_port"
	if [ $( basename $OCF_RESKEY_nc ) = 'socat' ]; then
		#socat has different parameters
		cmd="$OCF_RESKEY_nc -U TCP6-LISTEN:$OCF_RESKEY_port,backlog=10,fork,reuseaddr /dev/null"
	fi
	if ! lb_monitor; then
		ocf_log debug "Starting $process: $cmd"
		# Execute the command as created above
		$cmd >/dev/null 2>&1 &
		echo $! > $pidfile
		if lb_monitor; then
			ocf_log debug "$process: $cmd started successfully, calling monitor"
			lb_monitor
			return $?
		else 
			ocf_log err "$process: $cmd could not be started"
			return $OCF_ERR_GENERIC
		fi
	else
		# If already running, consider start successful
		ocf_log debug "$process: $cmd is already running"
		return $OCF_SUCCESS
	fi
}

lb_stop() {
	stop_rc=$OCF_SUCCESS

        if [ -n "$OCF_RESKEY_CRM_meta_timeout" ]; then
                # Allow 2/3 of the action timeout for the orderly shutdown
                # (The origin unit is ms, hence the conversion)
                stop_timeout=$((OCF_RESKEY_CRM_meta_timeout/1500))
        else
                stop_timeout=10
        fi

	if lb_monitor; then
                pid=`getpid $pidfile`
                kill $pid

                i=0
                while [ $i -lt $stop_timeout ]; do
                        if ! lb_monitor; then
                        	rm -f $pidfile
                                return $OCF_SUCCESS
                        fi
                        sleep 1 
                        i=$((i+1))
                done

                ocf_log warn "Stop with SIGTERM failed/timed out, now sending SIGKILL."
                kill -s 9 $pid
                while :; do
                        if ! lb_monitor; then
                                ocf_log warn "SIGKILL did the job."
                                stop_rc=$OCF_SUCCESS
                                break
                        fi
                        ocf_log info "The job still hasn't stopped yet. Waiting..."
                        sleep 1
                done
	fi
	rm -f $pidfile 
	return $stop_rc
}

lb_validate() {
	check_binary "$OCF_RESKEY_nc"

	if ! ocf_is_decimal "$OCF_RESKEY_port"; then
		ocf_exit_reason "$OCF_RESKEY_port is not a valid port"
		exit $OCF_ERR_CONFIGURED
	fi

	return $OCF_SUCCESS
}

###############################################################################
#
# MAIN
#
###############################################################################

case $__OCF_ACTION in
	meta-data)
		lb_metadata
		exit $OCF_SUCCESS
		;;
	usage|help)
		lb_usage
		exit $OCF_SUCCESS
		;;
esac

if ! ocf_is_root; then
	ocf_log err "You must be root for $__OCF_ACTION operation."
	exit $OCF_ERR_PERM
fi

case $__OCF_ACTION in
	start)
		lb_validate 
		lb_start;;
	stop)
		lb_stop;;
	monitor)
		lb_monitor;;
	validate-all)
		lb_validate;;
	*)	
		echo $USAGE
		exit $OCF_ERR_UNIMPLEMENTED
		;;
esac

exit $?