[pkg-ntp-maintainers] Bug#679051: ntp: Start of ntp during boot is delayed by 60 seconds at lockfile-touch step

[Andy Smith]

Package: ntp
Version: 1:4.2.6.p5+dfsg-2
Severity: minor

Dear Maintainer,

For a long while now (certainly through lenny and squeeze) I've noticed that on
boot of my Xen-based virtual machines, ntp seems to take an inordinately long
time to get started. I initially wrote this off as some peculiarity of Xen, as
ntp did start after about a minute and all seemed well.

I have just looked into it further by inserting a "set -x" in /etc/init.d/ntp
to see which part it was actually waiting on, and was surprised to find it was
lockfile-touch.

Here is a portion of the output of /etc/init.d/ntp up until the place where it
pauses:

+ /bin/echo -n Starting NTP server: ntpd
Starting NTP server: ntpd+ log_daemon_msg_post Starting NTP server ntpd
+ :
+ [ -z 101:104 ]
+ lock_ntpdate
+ [ -x /usr/bin/lockfile-create ]
+ lockfile-create /var/lock/ntpdate

At this point there is a pause of 60 seconds, and then:

+ LOCKTOUCHPID=1422
+ start-stop-daemon --start --quiet --oknodo --pidfile /var/run/ntpd.pid --startas /usr/sbin/ntpd -- -p /var/run/ntpd.pid -x -g -u 101:104
+ lockfile-touch /var/lock/ntpdate
+ status=0
+ unlock_ntpdate
+ [ -x /usr/bin/lockfile-create ]
+ kill 1422
+ lockfile-remove /var/lock/ntpdate
+ log_end_msg 0

...and so on.

In /etc/init.d/ntp there's this:

lock_ntpdate() {
        if [ -x /usr/bin/lockfile-create ]; then
                lockfile-create $LOCKFILE
                lockfile-touch $LOCKFILE &
                LOCKTOUCHPID="$!"
        fi
}

I don't understand why, but it seems like the lockfile-touch part is waiting
for 60 seconds even though it's invoked with "&".

Subsequent stop/start of the daemon with invoke-rc.d after boot don't have any
such pause.

This VM, though running wheezy, is based on an image that was initially running
sarge and has been dist-upgraded through all releases over the years. I haven't
yet tested whether the same thing happens on a clean install of wheezy in the
same environment.

-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)

Kernel: Linux 3.2.0-2-686-pae (SMP w/1 CPU core)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages ntp depends on:
ii  adduser      3.113+nmu3
ii  dpkg         1.16.3
ii  libc6        2.13-33
ii  libcap2      1:2.22-1
ii  libedit2     2.11-20080614-5
ii  libopts25    1:5.12-0.1
ii  libssl1.0.0  1.0.1c-3
ii  lsb-base     4.1+Debian6
ii  netbase      5.0

Versions of packages ntp recommends:
ii  perl  5.14.2-12

Versions of packages ntp suggests:
pn  ntp-doc  <none>

-- Configuration Files:
/etc/default/ntp changed:
NTPD_OPTS='-x -g'

/etc/init.d/ntp changed:
set -x
PATH=/sbin:/bin:/usr/sbin:/usr/bin
. /lib/lsb/init-functions
DAEMON=/usr/sbin/ntpd
PIDFILE=/var/run/ntpd.pid
test -x $DAEMON || exit 5
if [ -r /etc/default/ntp ]; then
	. /etc/default/ntp
fi
if [ -e /var/lib/ntp/ntp.conf.dhcp ]; then
	NTPD_OPTS="$NTPD_OPTS -c /var/lib/ntp/ntp.conf.dhcp"
fi
LOCKFILE=/var/lock/ntpdate
lock_ntpdate() {
	if [ -x /usr/bin/lockfile-create ]; then
		lockfile-create $LOCKFILE
		lockfile-touch $LOCKFILE &
		LOCKTOUCHPID="$!"
	fi
}
unlock_ntpdate() {
	if [ -x /usr/bin/lockfile-create ] ; then
		kill $LOCKTOUCHPID
		lockfile-remove $LOCKFILE
	fi
}
RUNASUSER=ntp
UGID=$(getent passwd $RUNASUSER | cut -f 3,4 -d:) || true
if test "$(uname -s)" = "Linux"; then
        NTPD_OPTS="$NTPD_OPTS -u $UGID"
fi
case $1 in
	start)
		log_daemon_msg "Starting NTP server" "ntpd"
		if [ -z "$UGID" ]; then
			log_failure_msg "user \"$RUNASUSER\" does not exist"
			exit 1
		fi
		lock_ntpdate
  		start-stop-daemon --start --quiet --oknodo --pidfile $PIDFILE --startas $DAEMON -- -p $PIDFILE $NTPD_OPTS
		status=$?
		unlock_ntpdate
		log_end_msg $status
  		;;
	stop)
		log_daemon_msg "Stopping NTP server" "ntpd"
  		start-stop-daemon --stop --quiet --oknodo --pidfile $PIDFILE
		log_end_msg $?
		rm -f $PIDFILE
  		;;
	restart|force-reload)
		$0 stop && sleep 2 && $0 start
  		;;
	try-restart)
		if $0 status >/dev/null; then
			$0 restart
		else
			exit 0
		fi
		;;
	reload)
		exit 3
		;;
	status)
		status_of_proc $DAEMON "NTP server"
		;;
	*)
		echo "Usage: $0 {start|stop|restart|try-restart|force-reload|status}"
		exit 2
		;;
esac

-- no debconf information