diff --git a/sys/kern/vfs_mount.c b/sys/kern/vfs_mount.c
index 702efc3..57cb877 100644
--- a/sys/kern/vfs_mount.c
+++ b/sys/kern/vfs_mount.c
@@ -1222,7 +1222,7 @@ dounmount(mp, flags, td)
 			VOP_UNLOCK(coveredvp, 0, td);
 		return (EBUSY);
 	}
-	mp->mnt_kern_flag |= MNTK_UNMOUNT;
+	mp->mnt_kern_flag |= MNTK_UNMOUNT | MNTK_NOINSMNTQ;
 	/* Allow filesystems to detect that a forced unmount is in progress. */
 	if (flags & MNT_FORCE)
 		mp->mnt_kern_flag |= MNTK_UNMOUNTF;
@@ -1230,7 +1230,8 @@ dounmount(mp, flags, td)
 	    ((flags & MNT_FORCE) ? 0 : LK_NOWAIT), MNT_MTX(mp), td);
 	if (error) {
 		MNT_ILOCK(mp);
-		mp->mnt_kern_flag &= ~(MNTK_UNMOUNT | MNTK_UNMOUNTF);
+		mp->mnt_kern_flag &= ~(MNTK_UNMOUNT | MNTK_NOINSMNTQ |
+		    MNTK_UNMOUNTF);
 		if (mp->mnt_kern_flag & MNTK_MWAIT)
 			wakeup(mp);
 		MNT_IUNLOCK(mp);
@@ -1285,9 +1286,13 @@ dounmount(mp, flags, td)
 			}
 			vput(fsrootvp);
 		}
-		if ((mp->mnt_flag & MNT_RDONLY) == 0 && mp->mnt_syncer == NULL)
-			(void) vfs_allocate_syncvnode(mp);
 		MNT_ILOCK(mp);
+		mp->mnt_kern_flag &= ~MNTK_NOINSMNTQ;
+		if ((mp->mnt_flag & MNT_RDONLY) == 0 && mp->mnt_syncer == NULL) {
+			MNT_IUNLOCK(mp);
+			(void) vfs_allocate_syncvnode(mp);
+			MNT_ILOCK(mp);
+		}
 		mp->mnt_kern_flag &= ~(MNTK_UNMOUNT | MNTK_UNMOUNTF);
 		mp->mnt_flag |= async_flag;
 		if ((mp->mnt_flag & MNT_ASYNC) != 0 && mp->mnt_noasync == 0)
diff --git a/sys/kern/vfs_subr.c b/sys/kern/vfs_subr.c
index ad4ca11..76dad07 100644
--- a/sys/kern/vfs_subr.c
+++ b/sys/kern/vfs_subr.c
@@ -1021,7 +1021,7 @@ insmntque1(struct vnode *vp, struct mount *mp,
 		("insmntque: vnode already on per mount vnode list"));
 	VNASSERT(mp != NULL, vp, ("Don't call insmntque(foo, NULL)"));
 	MNT_ILOCK(mp);
-	if ((mp->mnt_kern_flag & MNTK_UNMOUNT) != 0 &&
+	if ((mp->mnt_kern_flag & MNTK_NOINSMNTQ) != 0 &&
 	    mp->mnt_nvnodelistsize == 0) {
 		MNT_IUNLOCK(mp);
 		if (dtr != NULL)
diff --git a/sys/sys/mount.h b/sys/sys/mount.h
index ab6f181..66c0a46 100644
--- a/sys/sys/mount.h
+++ b/sys/sys/mount.h
@@ -303,10 +303,15 @@ void          __mnt_vnode_markerfree(struct vnode **mvp, struct mount *mp);
  * dounmount() is still waiting to lock the mountpoint. This allows
  * the filesystem to cancel operations that might otherwise deadlock
  * with the unmount attempt (used by NFS).
+ *
+ * MNTK_NOINSMNTQ is strict subset of MNTK_UNMOUNT. They are separated
+ * to allow for failed unmount attempt to restore the syncer vnode for
+ * the mount.
  */
 #define MNTK_UNMOUNTF	0x00000001	/* forced unmount in progress */
 #define MNTK_ASYNC	0x00000002	/* filtered async flag */
 #define MNTK_SOFTDEP	0x00000004	/* async disabled by softdep */
+#define MNTK_NOINSMNTQ	0x00000008	/* insmntque is not allowed */
 #define MNTK_UNMOUNT	0x01000000	/* unmount in progress */
 #define	MNTK_MWAIT	0x02000000	/* waiting for unmount to finish */
 #define	MNTK_SUSPEND	0x08000000	/* request write suspension */