Add a new exit-timeout option to watchdogd.

Watchdogd currently disables the watchdog when it exits, such as during
rc.shutdown processing.  That leaves the system vulnerable to getting hung
or deadlocked during the shutdown part of a reboot.  For embedded systems
it's especially important that the hardware watchdog always be active.  It
can also be useful for servers that are administered remotely.

The new -x <seconds> option tells watchdogd to program the watchdog with the
given timeout just before exiting.  The -x value can be longer or shorter
than the -t normal time value, to allow for various exceptional conditions
at shutdown such as allowing extra time for buffer flushing.

The exit value is also used internally in the "failsafe" handling (which
used to just disable the watchdog), on the theory that if you're using this
option, "safe" means having the watchdog always running, not disabled.

The default is still to disable the watchdog on exit if -x is not specified.

Differential Revision:	https://reviews.freebsd.org/D2556 (timed out)

2 files changed, 26 insertions, 8 deletions

diff --git a/usr.sbin/watchdogd/watchdogd.8 b/usr.sbin/watchdogd/watchdogd.8
index e60f42d..a83e775 100644
--- a/usr.sbin/watchdogd/watchdogd.8
+++ b/usr.sbin/watchdogd/watchdogd.8
@@ -27,7 +27,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd November 16, 2014
+.Dd May 11, 2015
 .Dt WATCHDOGD 8
 .Os
 .Sh NAME
@@ -46,6 +46,7 @@
 .Op Fl s Ar sleep
 .Op Fl t Ar timeout
 .Op Fl T Ar script_timeout
+.Op Fl x Ar exit_timeout
 .Sh DESCRIPTION
 The
 .Nm
@@ -103,14 +104,25 @@ defaults to the value specified by the
 .Fl s Ar sleep
 option.
 .Pp
+The
+.Fl x Ar exit_timeout
+argument is the timeout period (in seconds) to leave in effect when the
+program exits.
+Using
+.Fl x
+with a non-zero value protects against lockup during a reboot by
+triggering a hardware reset if the software reboot doesn't complete
+before the given timeout expires.
+.Pp
 Upon receiving the
 .Dv SIGTERM
 or
 .Dv SIGINT
 signals,
 .Nm
-will first instruct the kernel to no longer perform watchdog checks and then
-will terminate.
+will terminate, after first instructing the kernel to either disable the
+timeout or reset it to the value given by
+.Fl x Ar exit_timeout .
 .Pp
 The
 .Nm
diff --git a/usr.sbin/watchdogd/watchdogd.c b/usr.sbin/watchdogd/watchdogd.c
index 767aa8e..55210d6 100644
--- a/usr.sbin/watchdogd/watchdogd.c
+++ b/usr.sbin/watchdogd/watchdogd.c
@@ -77,6 +77,7 @@ static int debugging = 0;
 static int end_program = 0;
 static const char *pidfile = _PATH_VARRUN "watchdogd.pid";
 static u_int timeout = WD_TO_128SEC;
+static u_int exit_timeout = WD_TO_NEVER;
 static u_int pretimeout = 0;
 static u_int timeout_sec;
 static u_int passive = 0;
@@ -461,10 +462,10 @@ watchdog_onoff(int onoff)
 		/* pat one more time for good measure */
 		return watchdog_patpat((timeout|WD_ACTIVE));
 	 } else {
-		return watchdog_patpat(0);
+		return watchdog_patpat(exit_timeout);
 	 }
 failsafe:
-	watchdog_patpat(0);
+	watchdog_patpat(exit_timeout);
 	return (error);
 }
 
@@ -476,8 +477,8 @@ usage(void)
 {
 	if (is_daemon)
 		fprintf(stderr, "usage:\n"
-"  watchdogd [-dnSw] [-e cmd] [-I file] [-s sleep] [-t timeout]\n"
-"            [-T script_timeout]\n"
+"  watchdogd [-dnSw] [-e cmd] [-I pidfile] [-s sleep] [-t timeout]\n"
+"            [-T script_timeout] [-x exit_timeout]\n"
 "            [--debug]\n"
 "            [--pretimeout seconds] [-pretimeout-action action]\n"
 "            [--softtimeout] [-softtimeout-action action]\n"
@@ -697,7 +698,7 @@ parseargs(int argc, char *argv[])
 		is_daemon = 1;
 
 	if (is_daemon)
-		getopt_shortopts = "I:de:ns:t:ST:w?";
+		getopt_shortopts = "I:de:ns:t:ST:wx:?";
 	else
 		getopt_shortopts = "dt:?";
 
@@ -741,6 +742,11 @@ parseargs(int argc, char *argv[])
 		case 'w':
 			do_timedog = 1;
 			break;
+		case 'x':
+			exit_timeout = parse_timeout_to_pow2ns(c, NULL, optarg);
+			if (exit_timeout != 0)
+				exit_timeout |= WD_ACTIVE;
+			break;
 		case 0:
 			lopt = longopts[longindex].name;
 			if (!strcmp(lopt, "pretimeout")) {