namespace: unify limit behavior on non-directory paths

Despite the name, `Read{Write,Only}Directories=` already allows for
regular file paths to be masked. This commit adds the same behavior
to `InaccessibleDirectories=` and makes it explicit in the doc.
This patch introduces `/run/systemd/inaccessible/{reg,dir,chr,blk,fifo,sock}`
{dile,device}nodes and mounts on the appropriate one the paths specified
in `InacessibleDirectories=`.

Based on Luca's patch from https://github.com/systemd/systemd/pull/3327

man/systemd.exec.xml

@@ -855,24 +855,26 @@
         <listitem><para>Sets up a new file system namespace for
         executed processes. These options may be used to limit access
         a process might have to the main file system hierarchy. Each
-        setting takes a space-separated list of directory paths relative to
+        setting takes a space-separated list of paths relative to
         the host's root directory (i.e. the system running the service manager).
-        Directories listed in
+        Note that if entries contain symlinks, they are resolved from the host's root directory as well.
+        Entries (files or directories) listed in 
         <varname>ReadWriteDirectories=</varname> are accessible from
         within the namespace with the same access rights as from
-        outside. Directories listed in
+        outside. Entries listed in
         <varname>ReadOnlyDirectories=</varname> are accessible for
         reading only, writing will be refused even if the usual file
-        access controls would permit this. Directories listed in
+        access controls would permit this. Entries listed in
         <varname>InaccessibleDirectories=</varname> will be made
         inaccessible for processes inside the namespace, and may not
         countain any other mountpoints, including those specified by
         <varname>ReadWriteDirectories=</varname> or
         <varname>ReadOnlyDirectories=</varname>.
         Note that restricting access with these options does not extend
-        to submounts of a directory that are created later on. These
+        to submounts of a directory that are created later on.
+        Non-directory paths can be specified as well. These
         options may be specified more than once, in which case all
-        directories listed will have limited access from within the
+        paths listed will have limited access from within the
         namespace. If the empty string is assigned to this option, the
         specific list is reset, and all prior assignments have no
         effect.</para>

src/basic/mount-util.c

@@ -532,3 +532,21 @@ int repeat_unmount(const char *path, int flags) {
                 done = true;
         }
 }
+
+const char* mode_to_inaccessible_node(mode_t mode) {
+        switch(mode & S_IFMT) {
+                case S_IFREG:
+                        return "/run/systemd/inaccessible/reg";
+                case S_IFDIR:
+                        return "/run/systemd/inaccessible/dir";
+                case S_IFCHR:
+                        return "/run/systemd/inaccessible/chr";
+                case S_IFBLK:
+                        return "/run/systemd/inaccessible/blk";
+                case S_IFIFO:
+                        return "/run/systemd/inaccessible/fifo";
+                case S_IFSOCK:
+                        return "/run/systemd/inaccessible/sock";
+        }
+        return NULL;
+}

src/basic/mount-util.h

@@ -49,4 +49,6 @@ union file_handle_union {
         char padding[sizeof(struct file_handle) + MAX_HANDLE_SZ];
 };
 
+const char* mode_to_inaccessible_node(mode_t mode);
+
 #define FILE_HANDLE_INIT { .handle.handle_bytes = MAX_HANDLE_SZ }

src/core/dbus-execute.c

@@ -1346,12 +1346,12 @@ int bus_exec_context_set_transient_property(
                 if (mode != UNIT_CHECK) {
                         _cleanup_free_ char *joined = NULL;
 
-                        if (streq(name, "ReadWriteDirectories"))
-                                dirs = &c->read_write_dirs;
-                        else if (streq(name, "ReadOnlyDirectories"))
-                                dirs = &c->read_only_dirs;
-                        else /* "InaccessibleDirectories" */
-                                dirs = &c->inaccessible_dirs;
+                        if (STR_IN_SET(name, "ReadWriteDirectories", "ReadWritePaths"))
+                                dirs = &c->read_write_paths;
+                        else if (STR_IN_SET(name, "ReadOnlyDirectories", "ReadOnlyPaths"))
+                                dirs = &c->read_only_paths;
+                        else /* "InaccessiblePaths" */
+                                dirs = &c->inaccessible_paths;
 
                         if (strv_length(l) == 0) {
                                 *dirs = strv_free(*dirs);

src/core/mount-setup.c

@@ -28,6 +28,7 @@
 #include "cgroup-util.h"
 #include "dev-setup.h"
 #include "efivars.h"
+#include "fs-util.h"
 #include "label.h"
 #include "log.h"
 #include "macro.h"
@@ -403,9 +404,16 @@ int mount_setup(bool loaded_policy) {
          * really needs to stay for good, otherwise software that
          * copied sd-daemon.c into their sources will misdetect
          * systemd. */
-        mkdir_label("/run/systemd", 0755);
-        mkdir_label("/run/systemd/system", 0755);
-        mkdir_label("/run/systemd/inaccessible", 0000);
+        (void) mkdir_label("/run/systemd", 0755);
+        (void) mkdir_label("/run/systemd/system", 0755);
+        (void) mkdir_label("/run/systemd/inaccessible", 0000);
+        /* Set up inaccessible items */
+        (void) mknod("/run/systemd/inaccessible/reg", S_IFREG | 0000, 0);
+        (void) mkdir_label("/run/systemd/inaccessible/dir", 0000);
+        (void) mknod("/run/systemd/inaccessible/chr", S_IFCHR | 0000, makedev(0, 0));
+        (void) mknod("/run/systemd/inaccessible/blk", S_IFBLK | 0000, makedev(0, 0));
+        (void) mkfifo("/run/systemd/inaccessible/fifo", 0000);
+        (void) mknod("/run/systemd/inaccessible/sock", S_IFSOCK | 0000, 0);
 
         return 0;
 }

src/core/namespace.c

@@ -278,6 +278,7 @@ static int apply_mount(
 
         const char *what;
         int r;
+        struct stat target;
 
         assert(m);
 
@@ -287,12 +288,22 @@ static int apply_mount(
 
                 /* First, get rid of everything that is below if there
                  * is anything... Then, overmount it with an
-                 * inaccessible directory. */
+                 * inaccessible path. */
                 umount_recursive(m->path, 0);
 
-                what = "/run/systemd/inaccessible";
-                break;
+                r = lstat(m->path, &target);
+                if (r != 0) {
+                        if (m->ignore && errno == ENOENT)
+                                return 0;
+                        return -errno;
+                }
 
+                what = mode_to_inaccessible_node(target.st_mode);
+                if (what == NULL) {
+                        log_debug("File type not supported. Note that symlinks are not allowed");
+                        return -ELOOP;
+                }
+                break;
         case READONLY:
         case READWRITE:
                 /* Nothing to mount here, we just later toggle the
@@ -317,12 +328,16 @@ static int apply_mount(
         assert(what);
 
         r = mount(what, m->path, NULL, MS_BIND|MS_REC, NULL);
-        if (r >= 0)
+        if (r >= 0) {
                 log_debug("Successfully mounted %s to %s", what, m->path);
-        else if (m->ignore && errno == ENOENT)
-                return 0;
-
-        return r;
+                return r;
+        }
+        else {
+                if (m->ignore && errno == ENOENT)
+                        return 0;
+                log_debug("Failed mounting %s to %s: %s", what, m->path, strerror(errno));
+                return -errno;
+        }
 }
 
 static int make_read_only(BindMount *m) {