#!/bin/sh

#################################
# interface event script for ctdb
# this adds/removes IPs from your 
# public interface

[ -n "$CTDB_BASE" ] || \
    export CTDB_BASE=$(cd -P $(dirname "$0") ; dirname "$PWD")

. $CTDB_BASE/functions
loadconfig

[ -z "$CTDB_PUBLIC_ADDRESSES" ] && {
	CTDB_PUBLIC_ADDRESSES=$CTDB_BASE/public_addresses
}

[ ! -f "$CTDB_PUBLIC_ADDRESSES" ] && {
	if [ "$1" = "init" ]; then
		echo "No public addresses file found. Nothing to do for 10.interfaces"
	fi
	exit 0
}

mark_up ()
{
    up_interfaces_found=true
    ctdb setifacelink $1 up >/dev/null 2>&1
}

mark_down ()
{
    fail=true
    ctdb setifacelink $1 down >/dev/null 2>&1
}

# This sets $all_interfaces as a side-effect.
get_all_interfaces ()
{
    # Get all the interfaces listed in the public_addresses file
    all_interfaces=$(sed -e "s/^[^\t ]*[\t ]*//" -e "s/,/ /g" -e "s/[\t ]*$//" $CTDB_PUBLIC_ADDRESSES)

    # Add some special interfaces if they're defined
    [ "$CTDB_PUBLIC_INTERFACE" ] && all_interfaces="$CTDB_PUBLIC_INTERFACE $all_interfaces"
    [ "$CTDB_NATGW_PUBLIC_IFACE" ] && all_interfaces="$CTDB_NATGW_PUBLIC_IFACE $all_interfaces"

    # Get the interfaces for which CTDB has public IPs configured.
    # That is, for all but the 1st line, get the 1st field.
    ctdb_ifaces=$(ctdb -X ifaces | sed -e '1d' -e 's@^|@@' -e 's@|.*@@')

    # Add $ctdb_interfaces and uniquify
    all_interfaces=$(echo $all_interfaces $ctdb_ifaces | tr ' ' '\n' | sort -u)
}

get_real_iface ()
{
    # Output of "ip link show <iface>"
    _iface_info="$1"

    # Extract the full interface description to see if it is a VLAN
    _t=$(echo "$_iface_info" |
		awk 'NR == 1 { iface = $2; sub(":$", "", iface) ; \
			       print iface }')
    case "$_t" in
	*@*)
	    # VLAN: use the underlying interface, after the '@'
	    echo "${_t##*@}"
	    ;;
	*)
	    # Not a regular VLAN.  For backward compatibility, assume
	    # there is some other sort of VLAN that doesn't have the
	    # '@' in the output and only use what is before a '.'.  If
	    # there is no '.' then this will be the whole interface
	    # name.
	    echo "${_t%%.*}"
    esac
}

monitor_interfaces()
{
	get_all_interfaces

	fail=false
	up_interfaces_found=false

	# Note that this loop must not exit early.  It must process
	# all interfaces so that the correct state for each interface
	# is set in CTDB using mark_up/mark_down.  If there is a
	# problem with an interface then set fail=true and continue.
	for iface in $all_interfaces ; do

	    _iface_info=$(ip link show $iface 2>&1) || {
		echo "ERROR: Interface $iface does not exist but it is used by public addresses."
		mark_down $iface
		continue
	    }

	    # These interfaces are sometimes bond devices
	    # When we use VLANs for bond interfaces, there will only
	    # be an entry in /proc for the underlying real interface
	    realiface=$(get_real_iface "$_iface_info")
	    bi=$(get_proc "net/bonding/$realiface" 2>/dev/null) && {
		echo "$bi" | grep -q 'Currently Active Slave: None' && {
			echo "ERROR: No active slaves for bond device $realiface"
			mark_down $iface
			continue
		}
		echo "$bi" | grep -q '^MII Status: up' || {
			echo "ERROR: public network interface $realiface is down"
			mark_down $iface
			continue
		}
		echo "$bi" | grep -q '^Bonding Mode: IEEE 802.3ad Dynamic link aggregation' && {
			# This works around a bug in the driver where the
			# overall bond status can be up but none of the actual
			# physical interfaces have a link.
			echo "$bi" | grep 'MII Status:' | tail -n +2 | grep -q '^MII Status: up' || {
				echo "ERROR: No active slaves for 802.ad bond device $realiface"
				mark_down $iface
				continue
			}
		}
		mark_up $iface
		continue
	    }

	    case $iface in
	    lo*)
		# loopback is always working
		mark_up $iface
		;;
	    ib*)
		# we dont know how to test ib links
		mark_up $iface
		;;
	    *)
		ethtool $iface | grep -q 'Link detected: yes' || {
		    # On some systems, this is not successful when a
		    # cable is plugged but the interface has not been
		    # brought up previously. Bring the interface up
		    # and try again...
		    ip link set $iface up
		    ethtool $iface | grep -q 'Link detected: yes' || {
			echo "ERROR: No link on the public network interface $iface"
			mark_down $iface
			continue
		    }
		}
		mark_up $iface
		;;
	    esac

	done

	$fail || return 0

	$up_interfaces_found && \
	    [ "$CTDB_PARTIALLY_ONLINE_INTERFACES" = "yes" ] && \
	    return 0

	return 1
}

# Sets: iface, ip, maskbits, family
get_iface_ip_maskbits_family ()
{
    _iface_in="$1"
    ip="$2"
    _maskbits_in="$3"

    set -- $(ip_maskbits_iface "$ip")
    if [ -n "$1" ] ; then
	maskbits="$1"
	iface="$2"
	family="$3"

	if [ "$iface" != "$_iface_in" ] ; then
	    printf \
		'WARNING: Public IP %s hosted on interface %s but VNN says %s\n' \
		"$ip" "$iface" "$_iface_in"
	fi
	if [ "$maskbits" != "$_maskbits_in" ] ; then
	    printf \
		'WARNING: Public IP %s has %s bit netmask but VNN says %s\n' \
		    "$ip" "$maskbits" "$_maskbits_in"
	fi
    else
	die "ERROR: Unable to determine interface for IP ${ip}"
    fi
}

ctdb_check_args "$@"

case "$1" in 
     #############################
     # called when ctdbd starts up
     init)
	# make sure that we only respond to ARP messages from the NIC where
	# a particular ip address is associated.
	get_proc sys/net/ipv4/conf/all/arp_filter >/dev/null 2>&1 && {
	    set_proc sys/net/ipv4/conf/all/arp_filter 1
	}

	_promote="sys/net/ipv4/conf/all/promote_secondaries"
	get_proc "$_promote" >/dev/null 2>&1 || \
	    die "Public IPs only supported if promote_secondaries is available"

	# make sure we drop any ips that might still be held if
	# previous instance of ctdb got killed with -9 or similar
	drop_all_public_ips
	;;

     #############################
     # called after ctdbd has done its initial recovery
     # and we start the services to become healthy
     startup)
	monitor_interfaces
	;;


     ################################################
     # called when ctdbd wants to claim an IP address
     takeip)
	iface=$2
	ip=$3
	maskbits=$4

	add_ip_to_iface $iface $ip $maskbits || {
		exit 1;
	}

	# cope with the script being killed while we have the interface blocked
	case "$ip" in
	    *:*) family="inet6" ;;
	    *)   family="inet"  ;;
	esac
	iptables_wrapper $family -D INPUT -i $iface -d $ip -j DROP 2> /dev/null

	flush_route_cache
	;;


     ##################################################
     # called when ctdbd wants to release an IP address
     releaseip)
	# releasing an IP is a bit more complex than it seems. Once the IP
	# is released, any open tcp connections to that IP on this host will end
	# up being stuck. Some of them (such as NFS connections) will be unkillable
	# so we need to use the killtcp ctdb function to kill them off. We also
	# need to make sure that no new connections get established while we are 
	# doing this! So what we do is this:
	# 1) firewall this IP, so no new external packets arrive for it
	# 2) use netstat -tn to find existing connections, and kill them 
	# 3) remove the IP from the interface
	# 4) remove the firewall rule
	shift
	get_iface_ip_maskbits_family "$@"

	# we do an extra delete to cope with the script being killed
	iptables_wrapper $family -D INPUT -i $iface -d $ip -j DROP 2> /dev/null
	iptables_wrapper $family -I INPUT -i $iface -d $ip -j DROP
	kill_tcp_connections $ip

	delete_ip_from_iface $iface $ip $maskbits || {
	    iptables_wrapper $family \
			     -D INPUT -i $iface -d $ip -j DROP 2> /dev/null
		exit 1
	}

	iptables_wrapper $family -D INPUT -i $iface -d $ip -j DROP 2> /dev/null

	flush_route_cache
	;;

     ##################################################
     # called when ctdbd wants to update an IP address
     updateip)
	# moving an IP is a bit more complex than it seems.
	# First we drop all traffic on the old interface.
	# Then we try to add the ip to the new interface and before
	# we finally remove it from the old interface.
	#
	# 1) firewall this IP, so no new external packets arrive for it
	# 2) remove the IP from the old interface (and new interface, to be sure)
	# 3) add the IP to the new interface
	# 4) remove the firewall rule
	# 5) use ctdb gratiousarp to propagate the new mac address
	# 6) use netstat -tn to find existing connections, and tickle them
	_oiface=$2
	niface=$3
	_ip=$4
	_maskbits=$5

	get_iface_ip_maskbits_family "$_oiface" "$ip" "$maskbits"
	oiface="$iface"

	# we do an extra delete to cope with the script being killed
	iptables_wrapper $family -D INPUT -i $oiface -d $ip -j DROP 2> /dev/null
	iptables_wrapper $family -I INPUT -i $oiface -d $ip -j DROP

	delete_ip_from_iface $oiface $ip $maskbits 2>/dev/null
	delete_ip_from_iface $niface $ip $maskbits 2>/dev/null

	add_ip_to_iface $niface $ip $maskbits || {
	    iptables_wrapper $family \
			     -D INPUT -i $oiface -d $ip -j DROP 2> /dev/null
	    exit 1
	}

	# cope with the script being killed while we have the interface blocked
	iptables_wrapper $family -D INPUT -i $oiface -d $ip -j DROP 2> /dev/null

	flush_route_cache

	# propagate the new mac address
	ctdb gratiousarp $ip $niface

	# tickle all existing connections, so that dropped packets
	# are retransmited and the tcp streams work

	tickle_tcp_connections $ip

	;;

     monitor)
	monitor_interfaces || exit 1
	;;
    *)
	ctdb_standard_event_handler "$@"
	;;
esac

exit 0