multipath: don't set queue_if_no_path without multipathd

If multipathd is not running, when all paths to a device have failed, there's
no way for them to automatically get restored.  If the device is set to queue,
whatever is accessing it will hang forever. This can lead to problems if it
happens at boot-up.  This patch unsets queue_if_no_path for all devices created
when multipathd is not running. When multipathd starts, it will automatically
get reset queue_if_no_path to the proper value.  This new behaviour can be
overridden using the new "-q" option to multipath.

This version of the patch contacts multipathd's client socket to tell if it's
running.

Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>

3 files changed, 46 insertions, 3 deletions

diff --git a/libmultipath/config.h b/libmultipath/config.h
index 595aa79..6615e94 100644
--- a/libmultipath/config.h
+++ b/libmultipath/config.h
@@ -92,6 +92,7 @@ struct config {
 	int attribute_flags;
 	int fast_io_fail;
 	unsigned int dev_loss;
+	int allow_queueing;
 	uid_t uid;
 	gid_t gid;
 	mode_t mode;
diff --git a/libmultipath/configure.c b/libmultipath/configure.c
index 99d4515..c2788b6 100644
--- a/libmultipath/configure.c
+++ b/libmultipath/configure.c
@@ -35,6 +35,7 @@
 #include "alias.h"
 #include "prio.h"
 #include "util.h"
+#include "uxsock.h"
 
 extern int
 setup_map (struct multipath * mpp, char * params, int params_size)
@@ -448,6 +449,34 @@ deadmap (struct multipath * mpp)
 	return 1; /* dead */
 }
 
+int check_daemon(void)
+{
+	int fd;
+	char *reply;
+	size_t len;
+	int ret = 0;
+
+	fd = ux_socket_connect(DEFAULT_SOCKET);
+	if (fd == -1)
+		return 0;
+
+	if (send_packet(fd, "show daemon", 12) != 0)
+		goto out;
+	if (recv_packet(fd, &reply, &len) != 0)
+		goto out;
+
+	if (strstr(reply, "shutdown"))
+		goto out_free;
+
+	ret = 1;
+
+out_free:
+	FREE(reply);
+out:
+	close(fd);
+	return ret;
+}
+
 extern int
 coalesce_paths (struct vectors * vecs, vector newmp, char * refwwid, int force_reload)
 {
@@ -555,7 +584,16 @@ coalesce_paths (struct vectors * vecs, vector newmp, char * refwwid, int force_r
 		if (r == DOMAP_DRY)
 			continue;
 
-		if (mpp->no_path_retry != NO_PATH_RETRY_UNDEF) {
+		if (!conf->daemon && !conf->allow_queueing && !check_daemon()) {
+			if (mpp->no_path_retry != NO_PATH_RETRY_UNDEF &&
+			    mpp->no_path_retry != NO_PATH_RETRY_FAIL)
+				condlog(3, "%s: multipathd not running, unset "
+					"queue_if_no_path feature", mpp->alias);
+			if (!dm_queue_if_no_path(mpp->alias, 0))
+				remove_feature(&mpp->features,
+					       "queue_if_no_path");
+		}
+		else if (mpp->no_path_retry != NO_PATH_RETRY_UNDEF) {
 			if (mpp->no_path_retry == NO_PATH_RETRY_FAIL) {
 				condlog(3, "%s: unset queue_if_no_path feature",
 					mpp->alias);
diff --git a/multipath/main.c b/multipath/main.c
index f455985..654e031 100644
--- a/multipath/main.c
+++ b/multipath/main.c
@@ -79,7 +79,7 @@ usage (char * progname)
 {
 	fprintf (stderr, VERSION_STRING);
 	fprintf (stderr, "Usage:\n");
-	fprintf (stderr, "  %s [-d] [-r] [-v lvl] [-p pol] [-b fil] [dev]\n", progname);
+	fprintf (stderr, "  %s [-d] [-r] [-v lvl] [-p pol] [-b fil] [-q] [dev]\n", progname);
 	fprintf (stderr, "  %s -l|-ll|-f [-v lvl] [-b fil] [dev]\n", progname);
 	fprintf (stderr, "  %s -F [-v lvl]\n", progname);
 	fprintf (stderr, "  %s -t\n", progname);
@@ -92,6 +92,7 @@ usage (char * progname)
 		"  -ll     show multipath topology (maximum info)\n" \
 		"  -f      flush a multipath device map\n" \
 		"  -F      flush all multipath device maps\n" \
+		"  -q      allow queue_if_no_path when multipathd is not running\n"\
 		"  -d      dry run, do not create or update devmaps\n" \
 		"  -t      dump internal hardware table\n" \
 		"  -r      force devmap reload\n" \
@@ -397,7 +398,7 @@ main (int argc, char *argv[])
 		condlog(0, "multipath tools need sysfs mounted");
 		exit(1);
 	}
-	while ((arg = getopt(argc, argv, ":dhl::FfM:v:p:b:Brt")) != EOF ) {
+	while ((arg = getopt(argc, argv, ":dhl::FfM:v:p:b:Brtq")) != EOF ) {
 		switch(arg) {
 		case 1: printf("optarg : %s\n",optarg);
 			break;
@@ -414,6 +415,9 @@ main (int argc, char *argv[])
 		case 'B':
 			conf->bindings_read_only = 1;
 			break;
+		case 'q':
+			conf->allow_queueing = 1;
+			break;
 		case 'd':
 			conf->dry_run = 1;
 			break;