add retry option to supervise-daemon

The --retry option for supervise-daemon defines how the supervisor will
attempt to stop the child process it is monitoring. It is defined when
the supervisor is started since stopping the supervisor just sends a
signal to the active supervisor.

This fixes #160.

man/supervise-daemon.8

@@ -36,6 +36,8 @@
 .Ar pidfile
 .Fl P , -respawn-period
 .Ar seconds
+.Fl R , -retry
+.Ar arg
 .Fl r , -chroot
 .Ar chrootpath
 .Fl u , -user
@@ -115,6 +117,9 @@ Modifies the scheduling priority of the daemon.
 .It Fl P , -respawn-period Ar seconds
 Sets the length of a respawn period. The default is 10 seconds. See the
 description of --respawn-max for more information.
+.It Fl R , -retry Ar timeout | Ar signal Ns / Ns Ar timeout
+The retry specification can be either a timeout in seconds or multiple
+signal/timeout pairs (like SIGTERM/5).
 .It Fl r , -chroot Ar path
 chroot to this directory before starting the daemon. All other paths, such
 as the path to the daemon, chdir and pidfile, should be relative to the chroot.

sh/supervise-daemon.sh

@@ -23,6 +23,7 @@ supervise_start()
 	# command_args="this \"is a\" test"
 	# to work properly.
 	eval supervise-daemon --start \
+		${retry:+--retry} $retry \
 		${chroot:+--chroot} $chroot \
 		${pidfile:+--pidfile} $pidfile \
 		${respawn_delay:+--respawn-delay} $respawn_delay \

src/rc/Makefile

@@ -159,7 +159,7 @@ rc-update: rc-update.o _usage.o rc-misc.o
 start-stop-daemon: start-stop-daemon.o _usage.o rc-misc.o rc-schedules.o
 	${CC} ${LOCAL_CFLAGS} ${LOCAL_LDFLAGS} ${CFLAGS} ${LDFLAGS} -o $@ $^ ${LDADD}
 
-supervise-daemon: supervise-daemon.o _usage.o rc-misc.o
+supervise-daemon: supervise-daemon.o _usage.o rc-misc.o rc-schedules.o
 	${CC} ${LOCAL_CFLAGS} ${LOCAL_LDFLAGS} ${CFLAGS} ${LDFLAGS} -o $@ $^ ${LDADD}
 
 service_get_value service_set_value get_options save_options: do_value.o rc-misc.o

src/rc/supervise-daemon.c

@@ -61,12 +61,13 @@ static struct pam_conv conv = { NULL, NULL};
 #include "queue.h"
 #include "rc.h"
 #include "rc-misc.h"
+#include "rc-schedules.h"
 #include "_usage.h"
 #include "helpers.h"
 
 const char *applet = NULL;
 const char *extraopts = NULL;
-const char *getoptstring = "D:d:e:g:I:Kk:m:N:p:r:Su:1:2:" \
+const char *getoptstring = "D:d:e:g:I:Kk:m:N:p:R:r:Su:1:2:" \
 	getoptstring_COMMON;
 const struct option longopts[] = {
 	{ "respawn-delay",        1, NULL, 'D'},
@@ -80,6 +81,7 @@ const struct option longopts[] = {
 	{ "nicelevel",    1, NULL, 'N'},
 	{ "pidfile",      1, NULL, 'p'},
 	{ "respawn-period",        1, NULL, 'P'},
+	{ "retry",       1, NULL, 'R'},
 	{ "chroot",       1, NULL, 'r'},
 	{ "start",        0, NULL, 'S'},
 	{ "user",         1, NULL, 'u'},
@@ -99,6 +101,7 @@ const char * const longopts_help[] = {
 	"Set a nicelevel when starting",
 	"Match pid found in this file",
 	"Set respawn time period",
+	"Retry schedule to use when stopping",
 	"Chroot to this directory",
 	"Start daemon",
 	"Change the process user",
@@ -410,6 +413,9 @@ int main(int argc, char **argv)
 	bool stop = false;
 	char *exec = NULL;
 	char *pidfile = NULL;
+	char *retry = NULL;
+	int nkilled;
+	int sig = SIGTERM;
 	char *home = NULL;
 	int tid = 0;
 	pid_t child_pid, pid;
@@ -534,6 +540,9 @@ int main(int argc, char **argv)
 			pidfile = optarg;
 			break;
 
+		case 'R':  /* --retry <schedule>|timeout */
+			retry = optarg;
+			break;
 		case 'r':  /* --chroot /new/root */
 			ch_root = optarg;
 			break;
@@ -605,6 +614,10 @@ int main(int argc, char **argv)
 				"than %d to avoid infinite respawning", applet, 
 				respawn_delay * respawn_max);
 		}
+		if (retry)
+			parse_schedule(applet, retry, sig);
+		else
+			parse_schedule(applet, NULL, sig);
 	}
 
 	/* Expand ~ */
@@ -655,9 +668,13 @@ int main(int argc, char **argv)
 		else
 			i = kill(pid, SIGTERM);
 		if (i != 0)
-			/* We failed to stop something */
+			/* We failed to send the signal */
 			exit(EXIT_FAILURE);
 
+		/* wait for the supervisor to go down */
+		while (kill(pid, 0) == 0)
+			sleep(1);
+
 		/* Even if we have not actually killed anything, we should
 		 * remove information about it as it may have unexpectedly
 		 * crashed out. We should also return success as the end
@@ -737,7 +754,10 @@ int main(int argc, char **argv)
 			wait(&i);
 			if (exiting) {
 				syslog(LOG_INFO, "stopping %s, pid %d", exec, child_pid);
-				kill(child_pid, SIGTERM);
+				nkilled = run_stop_schedule(applet, exec, NULL, child_pid,
+						0, false, false);
+				if (nkilled > 0)
+					syslog(LOG_INFO, "killed %d processes", nkilled);
 			} else {
 				sleep(respawn_delay);
 				if (respawn_max > 0 && respawn_period > 0) {