Merge 7a1643f3bb into 5261c521e3

Follow-up for d49d95df0a.
Replaces 9a032ec55a.
Fixes #35075.

TODO

@@ -129,6 +129,10 @@ Deprecations and removals:
 
 Features:
 
+* format-table: introduce new cell type for strings with ansi sequences in
+  them. display them in regular output mode (via strip_tab_ansi()), but
+  suppress them in json mode.
+
 * machined: when registering a machine, also take a relative cgroup path,
   relative to the machine's unit. This is useful when registering unpriv
   machines, as they might sit down the cgroup tree, below a cgroup delegation
@@ -217,12 +221,8 @@ Features:
   services where mount propagation from the root fs is off, an still have
   confext/sysext propagated in.
 
-* support F_DUDFD_QUERY for comparing fds in same_fd (requires kernel 6.10)
-
 * generic interface for varlink for setting log level and stuff that all our daemons can implement
 
-* use pty ioctl to get peer wherever possible (TIOCGPTPEER)
-
 * maybe teach repart.d/ dropins a new setting MakeMountNodes= or so, which is
   just like MakeDirectories=, but uses an access mode of 0000 and sets the +i
   chattr bit. This is useful as protection against early uses of /var/ or /tmp/
@@ -253,8 +253,6 @@ Features:
 * initrd: when transitioning from initrd to host, validate that
   /lib/modules/`uname -r` exists, refuse otherwise
 
-* tmpfiles: add "owning" flag for lines that limits effect of --purge
-
 * signed bpf loading: to address need for signature verification for bpf
   programs when they are loaded, and given the bpf folks don't think this is
   realistic in kernel space, maybe add small daemon that facilitates this
@@ -458,9 +456,6 @@ Features:
 * introduce mntid_t, and make it 64bit, as apparently the kernel switched to
   64bit mount ids
 
-* use udev rule networkd ownership property to take ownership of network
-  interfaces nspawn creates
-
 * mountfsd/nsresourced
   - userdb: maybe allow callers to map one uid to their own uid
   - bpflsm: allow writes if resulting UID on disk would be userns' owner UID
@@ -647,6 +642,7 @@ Features:
   - openpt_allocate_in_namespace()
   - unit_attach_pid_to_cgroup_via_bus()
   - cg_attach() – requires new kernel feature
+  - journald's process cache
 
 * ddi must be listed as block device fstype
 
@@ -1470,9 +1466,6 @@ Features:
 
 * in sd-id128: also parse UUIDs in RFC4122 URN syntax (i.e. chop off urn:uuid: prefix)
 
-* DynamicUser= + StateDirectory= → use uid mapping mounts, too, in order to
-  make dirs appear under right UID.
-
 * systemd-sysext: optionally, run it in initrd already, before transitioning
   into host, to open up possibility for services shipped like that.
 
@@ -1644,14 +1637,6 @@ Features:
 
 * maybe add kernel cmdline params: to force random seed crediting
 
-* introduce a new per-process uuid, similar to the boot id, the machine id, the
-  invocation id, that is derived from process creds, specifically a hashed
-  combination of AT_RANDOM + getpid() + the starttime from
-  /proc/self/status. Then add these ids implicitly when logging. Deriving this
-  uuid from these three things has the benefit that it can be derived easily
-  from /proc/$PID/ in a stable, and unique way that changes on both fork() and
-  exec().
-
 * let's not GC a unit while its ratelimits are still pending
 
 * when killing due to service watchdog timeout maybe detect whether target

src/basic/fs-util.c

@@ -1131,6 +1131,8 @@ int xopenat_full(int dir_fd, const char *path, int open_flags, XOpenFlags xopen_
          *   • If O_CREAT is used with XO_LABEL, any created file will be immediately relabelled.
          *
          *   • If the path is specified NULL or empty, behaves like fd_reopen().
+         *
+         *   • If XO_NOCOW is specified will turn on the NOCOW btrfs flag on the file, if available.
          */
 
         if (isempty(path)) {

src/machine/machine-varlink.c

@@ -12,6 +12,7 @@
 #include "json-util.h"
 #include "machine-varlink.h"
 #include "machine.h"
+#include "mount-util.h"
 #include "path-util.h"
 #include "pidref.h"
 #include "process-util.h"
@@ -570,3 +571,100 @@ int vl_method_open(sd_varlink *link, sd_json_variant *parameters, sd_varlink_met
 
         return sd_varlink_reply(link, v);
 }
+
+typedef struct MachineMountParameters {
+        const char *name;
+        PidRef pidref;
+        char *src, *dest;
+        bool read_only, mkdir;
+} MachineMountParameters;
+
+static void machine_mount_paramaters_done(MachineMountParameters *p) {
+        assert(p);
+        pidref_done(&p->pidref);
+        free(p->src);
+        free(p->dest);
+}
+
+int vl_method_mount(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata) {
+        static const sd_json_dispatch_field dispatch_table[] = {
+                VARLINK_DISPATCH_MACHINE_LOOKUP_FIELDS(MachineOpenParameters),
+                { "source",              SD_JSON_VARIANT_STRING,  json_dispatch_path,       offsetof(MachineMountParameters, src),       SD_JSON_MANDATORY },
+                { "destination",         SD_JSON_VARIANT_STRING,  json_dispatch_path,       offsetof(MachineMountParameters, dest),      0                 },
+                { "readOnly",            SD_JSON_VARIANT_BOOLEAN, sd_json_dispatch_stdbool, offsetof(MachineMountParameters, read_only), 0                 },
+                { "makeFileOrDirectory", SD_JSON_VARIANT_BOOLEAN, sd_json_dispatch_stdbool, offsetof(MachineMountParameters, mkdir),     0                 },
+                VARLINK_DISPATCH_POLKIT_FIELD,
+                {}
+        };
+
+        Manager *manager = ASSERT_PTR(userdata);
+        _cleanup_(machine_mount_paramaters_done) MachineMountParameters p = { .pidref = PIDREF_NULL };
+        MountInNamespaceFlags mount_flags = 0;
+        uid_t uid_shift;
+        int r;
+
+        assert(link);
+        assert(parameters);
+
+        r = sd_varlink_dispatch(link, parameters, dispatch_table, &p);
+        if (r != 0)
+                return r;
+
+        /* There is no need for extra validation since path_is_absolute() does path_is_valid() and path_is_absolute().*/
+        const char *dest = p.dest ?: p.src;
+
+        Machine *machine;
+        r = lookup_machine_by_name_or_pidref(link, manager, p.name, &p.pidref, &machine);
+        if (r == -ESRCH)
+                return sd_varlink_error(link, "io.systemd.Machine.NoSuchMachine", NULL);
+        if (r != 0)
+                return r;
+
+        if (machine->class != MACHINE_CONTAINER)
+                return sd_varlink_error(link, "io.systemd.Machine.NotSupported", NULL);
+
+        r = varlink_verify_polkit_async(
+                        link,
+                        manager->bus,
+                        "org.freedesktop.machine1.manage-machines",
+                        (const char**) STRV_MAKE("name", machine->name,
+                                                 "verb", "bind", // TODO(ikruglov): mount seems a better verb here, but it's bind in DBus
+                                                 "src", p.src,
+                                                 "dest", dest),
+                        &manager->polkit_registry);
+        if (r <= 0)
+                return r;
+
+        r = machine_get_uid_shift(machine, &uid_shift);
+        if (r < 0)
+                return log_debug_errno(r, "Failed to get machine UID shift: %m");
+        if (uid_shift != 0) {
+                log_debug_errno(EOPNOTSUPP, "Can't bind mount on container with user namespacing applied.");
+                return sd_varlink_error(link, "io.systemd.Machine.NotSupported", NULL);
+        }
+
+        if (p.read_only)
+                mount_flags |= MOUNT_IN_NAMESPACE_READ_ONLY;
+        if (p.mkdir)
+                mount_flags |= MOUNT_IN_NAMESPACE_MAKE_FILE_OR_DIRECTORY;
+
+        const char *propagate_directory = strjoina("/run/systemd/nspawn/propagate/", machine->name);
+        if (!propagate_directory)
+                return -ENOMEM;
+
+        /* TODO(ikruglov): base on implementation of mount_in_namespace() it
+         * feels better to convert this in an Operation() so we can apply
+         * `manager->n_operations >= OPERATIONS_MAX)`.  But it will require
+         * some surgery. */
+        r = bind_mount_in_namespace(
+                        &machine->leader,
+                        propagate_directory,
+                        "/run/host/incoming/",
+                        p.src,
+                        dest,
+                        mount_flags);
+        if (r < 0)
+                return log_debug_errno(r, "Failed to mount %s on %s in machine's namespace: %m", p.src, dest);
+
+        return sd_varlink_reply(link, NULL);
+}

src/machine/machine-varlink.h

@@ -25,3 +25,4 @@ int vl_method_unregister_internal(sd_varlink *link, sd_json_variant *parameters,
 int vl_method_terminate_internal(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata);
 int vl_method_kill(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata);
 int vl_method_open(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata);
+int vl_method_mount(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata);

src/machine/machined-varlink.c

@@ -774,6 +774,7 @@ static int manager_varlink_init_machine(Manager *m) {
                         "io.systemd.Machine.Terminate",   vl_method_terminate,
                         "io.systemd.Machine.Kill",        vl_method_kill,
                         "io.systemd.Machine.Open",        vl_method_open,
+                        "io.systemd.Machine.Mount",       vl_method_mount,
                         "io.systemd.MachineImage.List",   vl_method_list_images,
                         "io.systemd.MachineImage.Update", vl_method_update_image,
                         "io.systemd.MachineImage.Clone",  vl_method_clone_image,

src/shared/mount-util.c

@@ -1808,63 +1808,81 @@ char* umount_and_unlink_and_free(char *p) {
         return mfree(p);
 }
 
-static int path_get_mount_info(
+static int path_get_mount_info_at(
+                int dir_fd,
                 const char *path,
                 char **ret_fstype,
                 char **ret_options) {
 
         _cleanup_(mnt_free_tablep) struct libmnt_table *table = NULL;
-        _cleanup_free_ char *fstype = NULL, *options = NULL;
-        struct libmnt_fs *fs;
-        int r;
+        _cleanup_(mnt_free_iterp) struct libmnt_iter *iter = NULL;
+        int r, mnt_id;
 
-        assert(path);
+        assert(dir_fd >= 0 || dir_fd == AT_FDCWD);
 
-        table = mnt_new_table();
-        if (!table)
-                return -ENOMEM;
-
-        r = mnt_table_parse_mtab(table, /* filename = */ NULL);
+        r = path_get_mnt_id_at(dir_fd, path, &mnt_id);
         if (r < 0)
-                return r;
+                return log_debug_errno(r, "Failed to get mount ID: %m");
 
-        fs = mnt_table_find_mountpoint(table, path, MNT_ITER_FORWARD);
-        if (!fs)
-                return -EINVAL;
+        r = libmount_parse("/proc/self/mountinfo", NULL, &table, &iter);
+        if (r < 0)
+                return log_debug_errno(r, "Failed to parse /proc/self/mountinfo: %m");
 
-        if (ret_fstype) {
-                fstype = strdup(strempty(mnt_fs_get_fstype(fs)));
-                if (!fstype)
-                        return -ENOMEM;
+        for (;;) {
+                struct libmnt_fs *fs;
+
+                r = mnt_table_next_fs(table, iter, &fs);
+                if (r == 1)
+                        break; /* EOF */
+                if (r < 0)
+                        return log_debug_errno(r, "Failed to get next entry from /proc/self/mountinfo: %m");
+
+                if (mnt_fs_get_id(fs) != mnt_id)
+                        continue;
+
+                _cleanup_free_ char *fstype = NULL, *options = NULL;
+
+                if (ret_fstype) {
+                        fstype = strdup(strempty(mnt_fs_get_fstype(fs)));
+                        if (!fstype)
+                                return log_oom_debug();
+                }
+
+                if (ret_options) {
+                        options = strdup(strempty(mnt_fs_get_options(fs)));
+                        if (!options)
+                                return log_oom_debug();
+                }
+
+                if (ret_fstype)
+                        *ret_fstype = TAKE_PTR(fstype);
+                if (ret_options)
+                        *ret_options = TAKE_PTR(options);
+
+                return 0;
         }
 
-        if (ret_options) {
-                options = strdup(strempty(mnt_fs_get_options(fs)));
-                if (!options)
-                        return -ENOMEM;
-        }
-
-        if (ret_fstype)
-                *ret_fstype = TAKE_PTR(fstype);
-        if (ret_options)
-                *ret_options = TAKE_PTR(options);
-
-        return 0;
+        return log_debug_errno(SYNTHETIC_ERRNO(ESTALE), "Cannot find mount ID %i from /proc/self/mountinfo.", mnt_id);
 }
 
-int path_is_network_fs_harder(const char *path) {
+int path_is_network_fs_harder_at(int dir_fd, const char *path) {
+        _cleanup_close_ int fd = -EBADF;
+        int r;
+
+        assert(dir_fd >= 0 || dir_fd == AT_FDCWD);
+
+        fd = xopenat(dir_fd, path, O_PATH | O_CLOEXEC | O_NOFOLLOW);
+        if (fd < 0)
+                return fd;
+
+        r = fd_is_network_fs(fd);
+        if (r != 0)
+                return r;
+
         _cleanup_free_ char *fstype = NULL, *options = NULL;
-        int r, ret;
-
-        assert(path);
-
-        ret = path_is_network_fs(path);
-        if (ret > 0)
-                return true;
-
-        r = path_get_mount_info(path, &fstype, &options);
+        r = path_get_mount_info_at(fd, /* path = */ NULL, &fstype, &options);
         if (r < 0)
-                return RET_GATHER(ret, r);
+                return r;
 
         if (fstype_is_network(fstype))
                 return true;

src/shared/mount-util.h

@@ -181,4 +181,7 @@ int mount_credentials_fs(const char *path, size_t size, bool ro);
 
 int make_fsmount(int error_log_level, const char *what, const char *type, unsigned long flags, const char *options, int userns_fd);
 
-int path_is_network_fs_harder(const char *path);
+int path_is_network_fs_harder_at(int dir_fd, const char *path);
+static inline int path_is_network_fs_harder(const char *path) {
+        return path_is_network_fs_harder_at(AT_FDCWD, path);
+}

src/test/test-mount-util.c

@@ -538,9 +538,53 @@ TEST(bind_mount_submounts) {
 }
 
 TEST(path_is_network_fs_harder) {
-        ASSERT_OK_ZERO(path_is_network_fs_harder("/dev"));
-        ASSERT_OK_ZERO(path_is_network_fs_harder("/sys"));
-        ASSERT_OK_ZERO(path_is_network_fs_harder("/run"));
+        _cleanup_close_ int dir_fd = -EBADF;
+        int r;
+
+        ASSERT_OK(dir_fd = open("/", O_PATH | O_CLOEXEC));
+        FOREACH_STRING(s,
+                       "/", "/dev/", "/proc/", "/run/", "/sys/", "/tmp/", "/usr/", "/var/tmp/",
+                       "", ".", "../../../", "/this/path/should/not/exist/for/test-mount-util/") {
+
+                r = path_is_network_fs_harder(s);
+                log_debug("path_is_network_fs_harder(%s) → %i: %s", s, r, r < 0 ? STRERROR(r) : yes_no(r));
+
+                const char *q = path_startswith(s, "/") ?: s;
+                r = path_is_network_fs_harder_at(dir_fd, q);
+                log_debug("path_is_network_fs_harder_at(root, %s) → %i: %s", q, r, r < 0 ? STRERROR(r) : yes_no(r));
+        }
+
+        if (geteuid() != 0 || have_effective_cap(CAP_SYS_ADMIN) <= 0) {
+                (void) log_tests_skipped("not running privileged");
+                return;
+        }
+
+        _cleanup_(rm_rf_physical_and_freep) char *t = NULL;
+        assert_se(mkdtemp_malloc("/tmp/test-mount-util.path_is_network_fs_harder.XXXXXXX", &t) >= 0);
+
+        r = safe_fork("(make_mount-point)",
+                      FORK_RESET_SIGNALS |
+                      FORK_CLOSE_ALL_FDS |
+                      FORK_DEATHSIG_SIGTERM |
+                      FORK_WAIT |
+                      FORK_REOPEN_LOG |
+                      FORK_LOG |
+                      FORK_NEW_MOUNTNS |
+                      FORK_MOUNTNS_SLAVE,
+                      NULL);
+        ASSERT_OK(r);
+
+        if (r == 0) {
+                ASSERT_OK(mount_nofollow_verbose(LOG_INFO, "tmpfs", t, "tmpfs", 0, NULL));
+                ASSERT_OK_ZERO(path_is_network_fs_harder(t));
+                ASSERT_OK_ERRNO(umount(t));
+
+                ASSERT_OK(mount_nofollow_verbose(LOG_INFO, "tmpfs", t, "tmpfs", 0, "x-systemd-growfs,x-systemd-automount"));
+                ASSERT_OK_ZERO(path_is_network_fs_harder(t));
+                ASSERT_OK_ERRNO(umount(t));
+
+                _exit(EXIT_SUCCESS);
+        }
 }
 
 DEFINE_TEST_MAIN(LOG_DEBUG);

test/meson.build

@@ -142,11 +142,13 @@ endif
 ############################################################
 
 if install_tests
-        foreach script : ['integration-test-setup.sh', 'run-unit-tests.py']
-                install_data(script,
-                             install_mode : 'rwxr-xr-x',
-                             install_dir : testsdir)
-        endforeach
+        install_data('run-unit-tests.py',
+                     install_mode : 'rwxr-xr-x',
+                     install_dir : testsdir)
+
+        install_data('integration-test-setup.sh',
+                     install_mode : 'rwxr-xr-x',
+                     install_dir : testdata_dir)
 endif
 
 ############################################################

test/test.service.in

@@ -7,9 +7,9 @@ Before=getty-pre.target
 
 [Service]
 ExecStartPre=rm -f /failed /testok
-ExecStartPre=/usr/lib/systemd/tests/integration-test-setup.sh setup
+ExecStartPre=/usr/lib/systemd/tests/testdata/integration-test-setup.sh setup
 ExecStart=@command@
-ExecStopPost=/usr/lib/systemd/tests/integration-test-setup.sh finalize
+ExecStopPost=/usr/lib/systemd/tests/testdata/integration-test-setup.sh finalize
 Type=oneshot
 MemoryAccounting=@memory-accounting@
 StateDirectory=%N

test/units/TEST-07-PID1.private-pids.sh

@@ -132,10 +132,12 @@ testcase_unpriv() {
         return 0
     fi
 
-    # The kernel has a restriction for unprivileged user namespaces where they cannot mount a less restrictive
-    # instance of /proc/. So if /proc/ is masked (e.g. /proc/kmsg is over-mounted with tmpfs as systemd-nspawn does),
-    # then mounting a new /proc/ will fail and we will still see the host's /proc/. Thus, to allow tests to run in
-    # a VM or nspawn, we mount a new proc on a temporary directory with no masking to bypass this kernel restriction.
+    # IMPORTANT: For /proc/ to be remounted in pid namespace within an unprivileged user namespace, there needs to
+    # be at least 1 unmasked procfs mount in ANY directory. Otherwise, if /proc/ is masked (e.g. /proc/scsi is
+    # over-mounted with tmpfs), then mounting a new /proc/ will fail.
+    #
+    # Thus, to guarantee PrivatePIDs=yes tests for unprivileged users pass, we mount a new procfs on a temporary
+    # directory with no masking. This will guarantee an unprivileged user can mount a new /proc/ successfully.
     mkdir -p /tmp/TEST-07-PID1-private-pids-proc
     mount -t proc proc /tmp/TEST-07-PID1-private-pids-proc
 
@@ -146,7 +148,16 @@ testcase_unpriv() {
     umount /tmp/TEST-07-PID1-private-pids-proc
     rm -rf /tmp/TEST-07-PID1-private-pids-proc
 
-    # Now verify the behavior with masking - units should fail as PrivatePIDs=yes has no graceful fallback.
+    # Now we will mask /proc/ by mounting tmpfs over /proc/scsi. This will guarantee that mounting /proc/ will fail
+    # for unprivileged users when using PrivatePIDs=yes. Now units should fail as PrivatePIDs=yes has no graceful
+    # fallback.
+    #
+    # Note some kernels do not have /proc/scsi so we verify the directory exists prior to running the test.
+    if [ ! -d /proc/scsi ]; then
+        echo "/proc/scsi does not exist, skipping unprivileged PrivatePIDs=yes test with masked /proc/"
+        return 0
+    fi
+
     if [[ "$HAS_EXISTING_SCSI_MOUNT" == "no" ]]; then
         mount -t tmpfs tmpfs /proc/scsi
     fi