summaryrefslogtreecommitdiff
path: root/src/core/mount-setup.c
blob: 7295efbf3148b4082c764c80aaaa270ee246c003 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
/***
  This file is part of systemd.

  Copyright 2010 Lennart Poettering

  systemd is free software; you can redistribute it and/or modify it
  under the terms of the GNU Lesser General Public License as published by
  the Free Software Foundation; either version 2.1 of the License, or
  (at your option) any later version.

  systemd is distributed in the hope that it will be useful, but
  WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  Lesser General Public License for more details.

  You should have received a copy of the GNU Lesser General Public License
  along with systemd; If not, see <http://www.gnu.org/licenses/>.
***/

#include <errno.h>
#include <ftw.h>
#include <stdlib.h>
#include <sys/mount.h>
#include <unistd.h>

#include "alloc-util.h"
#include "bus-util.h"
#include "cgroup-util.h"
#include "dev-setup.h"
#include "efivars.h"
#include "fs-util.h"
#include "label.h"
#include "log.h"
#include "macro.h"
#include "missing.h"
#include "mkdir.h"
#include "mount-setup.h"
#include "mount-util.h"
#include "path-util.h"
#include "set.h"
#include "smack-util.h"
#include "strv.h"
#include "user-util.h"
#include "util.h"
#include "virt.h"

typedef enum MountMode {
        MNT_NONE  =        0,
        MNT_FATAL =        1 <<  0,
        MNT_IN_CONTAINER = 1 <<  1,
} MountMode;

typedef struct MountPoint {
        const char *what;
        const char *where;
        const char *type;
        const char *options;
        unsigned long flags;
        bool (*condition_fn)(void);
        MountMode mode;
} MountPoint;

/* The first three entries we might need before SELinux is up. The
 * fourth (securityfs) is needed by IMA to load a custom policy. The
 * other ones we can delay until SELinux and IMA are loaded. When
 * SMACK is enabled we need smackfs, too, so it's a fifth one. */
#ifdef HAVE_SMACK
#define N_EARLY_MOUNT 5
#else
#define N_EARLY_MOUNT 4
#endif

static const MountPoint mount_table[] = {
        { "sysfs",       "/sys",                      "sysfs",      NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          NULL,          MNT_FATAL|MNT_IN_CONTAINER },
        { "proc",        "/proc",                     "proc",       NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          NULL,          MNT_FATAL|MNT_IN_CONTAINER },
        { "devtmpfs",    "/dev",                      "devtmpfs",   "mode=755",                MS_NOSUID|MS_STRICTATIME,
          NULL,          MNT_FATAL|MNT_IN_CONTAINER },
        { "securityfs",  "/sys/kernel/security",      "securityfs", NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          NULL,          MNT_NONE                   },
#ifdef HAVE_SMACK
        { "smackfs",     "/sys/fs/smackfs",           "smackfs",    "smackfsdef=*",            MS_NOSUID|MS_NOEXEC|MS_NODEV,
          mac_smack_use, MNT_FATAL                  },
        { "tmpfs",       "/dev/shm",                  "tmpfs",      "mode=1777,smackfsroot=*", MS_NOSUID|MS_NODEV|MS_STRICTATIME,
          mac_smack_use, MNT_FATAL                  },
#endif
        { "tmpfs",       "/dev/shm",                  "tmpfs",      "mode=1777",               MS_NOSUID|MS_NODEV|MS_STRICTATIME,
          NULL,          MNT_FATAL|MNT_IN_CONTAINER },
        { "devpts",      "/dev/pts",                  "devpts",     "mode=620,gid=" STRINGIFY(TTY_GID), MS_NOSUID|MS_NOEXEC,
          NULL,          MNT_IN_CONTAINER           },
#ifdef HAVE_SMACK
        { "tmpfs",       "/run",                      "tmpfs",      "mode=755,smackfsroot=*",  MS_NOSUID|MS_NODEV|MS_STRICTATIME,
          mac_smack_use, MNT_FATAL                  },
#endif
        { "tmpfs",       "/run",                      "tmpfs",      "mode=755",                MS_NOSUID|MS_NODEV|MS_STRICTATIME,
          NULL,          MNT_FATAL|MNT_IN_CONTAINER },
        { "cgroup",      "/sys/fs/cgroup",            "cgroup2",    NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          cg_is_unified_wanted, MNT_IN_CONTAINER },
        { "tmpfs",       "/sys/fs/cgroup",            "tmpfs",      "mode=755",                MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_STRICTATIME,
          cg_is_legacy_wanted, MNT_FATAL|MNT_IN_CONTAINER },
        { "cgroup",      "/sys/fs/cgroup/unified",    "cgroup2",    NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          cg_is_hybrid_wanted, MNT_IN_CONTAINER },
        { "cgroup",      "/sys/fs/cgroup/systemd",    "cgroup",     "none,name=systemd,xattr", MS_NOSUID|MS_NOEXEC|MS_NODEV,
          cg_is_legacy_wanted, MNT_IN_CONTAINER     },
        { "cgroup",      "/sys/fs/cgroup/systemd",    "cgroup",     "none,name=systemd",       MS_NOSUID|MS_NOEXEC|MS_NODEV,
          cg_is_legacy_wanted, MNT_FATAL|MNT_IN_CONTAINER },
        { "pstore",      "/sys/fs/pstore",            "pstore",     NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          NULL,          MNT_NONE                   },
#ifdef ENABLE_EFI
        { "efivarfs",    "/sys/firmware/efi/efivars", "efivarfs",   NULL,                      MS_NOSUID|MS_NOEXEC|MS_NODEV,
          is_efi_boot,   MNT_NONE                   },
#endif
};

/* These are API file systems that might be mounted by other software,
 * we just list them here so that we know that we should ignore them */

static const char ignore_paths[] =
        /* SELinux file systems */
        "/sys/fs/selinux\0"
        /* Container bind mounts */
        "/proc/sys\0"
        "/dev/console\0"
        "/proc/kmsg\0";

bool mount_point_is_api(const char *path) {
        unsigned i;

        /* Checks if this mount point is considered "API", and hence
         * should be ignored */

        for (i = 0; i < ELEMENTSOF(mount_table); i ++)
                if (path_equal(path, mount_table[i].where))
                        return true;

        return path_startswith(path, "/sys/fs/cgroup/");
}

bool mount_point_ignore(const char *path) {
        const char *i;

        NULSTR_FOREACH(i, ignore_paths)
                if (path_equal(path, i))
                        return true;

        return false;
}

static int mount_one(const MountPoint *p, bool relabel) {
        int r;

        assert(p);

        if (p->condition_fn && !p->condition_fn())
                return 0;

        /* Relabel first, just in case */
        if (relabel)
                (void) label_fix(p->where, true, true);

        r = path_is_mount_point(p->where, NULL, AT_SYMLINK_FOLLOW);
        if (r < 0 && r != -ENOENT) {
                log_full_errno((p->mode & MNT_FATAL) ? LOG_ERR : LOG_DEBUG, r, "Failed to determine whether %s is a mount point: %m", p->where);
                return (p->mode & MNT_FATAL) ? r : 0;
        }
        if (r > 0)
                return 0;

        /* Skip securityfs in a container */
        if (!(p->mode & MNT_IN_CONTAINER) && detect_container() > 0)
                return 0;

        /* The access mode here doesn't really matter too much, since
         * the mounted file system will take precedence anyway. */
        if (relabel)
                (void) mkdir_p_label(p->where, 0755);
        else
                (void) mkdir_p(p->where, 0755);

        log_debug("Mounting %s to %s of type %s with options %s.",
                  p->what,
                  p->where,
                  p->type,
                  strna(p->options));

        if (mount(p->what,
                  p->where,
                  p->type,
                  p->flags,
                  p->options) < 0) {
                log_full_errno((p->mode & MNT_FATAL) ? LOG_ERR : LOG_DEBUG, errno, "Failed to mount %s at %s: %m", p->type, p->where);
                return (p->mode & MNT_FATAL) ? -errno : 0;
        }

        /* Relabel again, since we now mounted something fresh here */
        if (relabel)
                (void) label_fix(p->where, false, false);

        return 1;
}

static int mount_points_setup(unsigned n, bool loaded_policy) {
        unsigned i;
        int r = 0;

        for (i = 0; i < n; i ++) {
                int j;

                j = mount_one(mount_table + i, loaded_policy);
                if (j != 0 && r >= 0)
                        r = j;
        }

        return r;
}

int mount_setup_early(void) {
        assert_cc(N_EARLY_MOUNT <= ELEMENTSOF(mount_table));

        /* Do a minimal mount of /proc and friends to enable the most
         * basic stuff, such as SELinux */
        return mount_points_setup(N_EARLY_MOUNT, false);
}

int mount_cgroup_controllers(char ***join_controllers) {
        _cleanup_set_free_free_ Set *controllers = NULL;
        int r;

        if (!cg_is_legacy_wanted())
                return 0;

        /* Mount all available cgroup controllers that are built into the kernel. */

        controllers = set_new(&string_hash_ops);
        if (!controllers)
                return log_oom();

        r = cg_kernel_controllers(controllers);
        if (r < 0)
                return log_error_errno(r, "Failed to enumerate cgroup controllers: %m");

        for (;;) {
                _cleanup_free_ char *options = NULL, *controller = NULL, *where = NULL;
                MountPoint p = {
                        .what = "cgroup",
                        .type = "cgroup",
                        .flags = MS_NOSUID|MS_NOEXEC|MS_NODEV,
                        .mode = MNT_IN_CONTAINER,
                };
                char ***k = NULL;

                controller = set_steal_first(controllers);
                if (!controller)
                        break;

                if (join_controllers)
                        for (k = join_controllers; *k; k++)
                                if (strv_find(*k, controller))
                                        break;

                if (k && *k) {
                        char **i, **j;

                        for (i = *k, j = *k; *i; i++) {

                                if (!streq(*i, controller)) {
                                        _cleanup_free_ char *t;

                                        t = set_remove(controllers, *i);
                                        if (!t) {
                                                free(*i);
                                                continue;
                                        }
                                }

                                *(j++) = *i;
                        }

                        *j = NULL;

                        options = strv_join(*k, ",");
                        if (!options)
                                return log_oom();
                } else {
                        options = controller;
                        controller = NULL;
                }

                where = strappend("/sys/fs/cgroup/", options);
                if (!where)
                        return log_oom();

                p.where = where;
                p.options = options;

                r = mount_one(&p, true);
                if (r < 0)
                        return r;

                if (r > 0 && k && *k) {
                        char **i;

                        for (i = *k; *i; i++) {
                                _cleanup_free_ char *t = NULL;

                                t = strappend("/sys/fs/cgroup/", *i);
                                if (!t)
                                        return log_oom();

                                r = symlink(options, t);
                                if (r >= 0) {
#ifdef SMACK_RUN_LABEL
                                        _cleanup_free_ char *src;
                                        src = strappend("/sys/fs/cgroup/", options);
                                        if (!src)
                                                return log_oom();
                                        r = mac_smack_copy(t, src);
                                        if (r < 0 && r != -EOPNOTSUPP)
                                                return log_error_errno(r, "Failed to copy smack label from %s to %s: %m", src, t);
#endif
                                } else if (errno != EEXIST)
                                        return log_error_errno(errno, "Failed to create symlink %s: %m", t);
                        }
                }
        }

        /* Now that we mounted everything, let's make the tmpfs the
         * cgroup file systems are mounted into read-only. */
        (void) mount("tmpfs", "/sys/fs/cgroup", "tmpfs", MS_REMOUNT|MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_STRICTATIME|MS_RDONLY, "mode=755");

        return 0;
}

#if defined(HAVE_SELINUX) || defined(HAVE_SMACK)
static int nftw_cb(
                const char *fpath,
                const struct stat *sb,
                int tflag,
                struct FTW *ftwbuf) {

        /* No need to label /dev twice in a row... */
        if (_unlikely_(ftwbuf->level == 0))
                return FTW_CONTINUE;

        label_fix(fpath, false, false);

        /* /run/initramfs is static data and big, no need to
         * dynamically relabel its contents at boot... */
        if (_unlikely_(ftwbuf->level == 1 &&
                      tflag == FTW_D &&
                      streq(fpath, "/run/initramfs")))
                return FTW_SKIP_SUBTREE;

        return FTW_CONTINUE;
};
#endif

int mount_setup(bool loaded_policy) {
        int r = 0;

        r = mount_points_setup(ELEMENTSOF(mount_table), loaded_policy);
        if (r < 0)
                return r;

#if defined(HAVE_SELINUX) || defined(HAVE_SMACK)
        /* Nodes in devtmpfs and /run need to be manually updated for
         * the appropriate labels, after mounting. The other virtual
         * API file systems like /sys and /proc do not need that, they
         * use the same label for all their files. */
        if (loaded_policy) {
                usec_t before_relabel, after_relabel;
                char timespan[FORMAT_TIMESPAN_MAX];

                before_relabel = now(CLOCK_MONOTONIC);

                nftw("/dev", nftw_cb, 64, FTW_MOUNT|FTW_PHYS|FTW_ACTIONRETVAL);
                nftw("/dev/shm", nftw_cb, 64, FTW_MOUNT|FTW_PHYS|FTW_ACTIONRETVAL);
                nftw("/run", nftw_cb, 64, FTW_MOUNT|FTW_PHYS|FTW_ACTIONRETVAL);

                after_relabel = now(CLOCK_MONOTONIC);

                log_info("Relabelled /dev and /run in %s.",
                         format_timespan(timespan, sizeof(timespan), after_relabel - before_relabel, 0));
        }
#endif

        /* Create a few default symlinks, which are normally created
         * by udevd, but some scripts might need them before we start
         * udevd. */
        dev_setup(NULL, UID_INVALID, GID_INVALID);

        /* Mark the root directory as shared in regards to mount propagation. The kernel defaults to "private", but we
         * think it makes more sense to have a default of "shared" so that nspawn and the container tools work out of
         * the box. If specific setups need other settings they can reset the propagation mode to private if
         * needed. Note that we set this only when we are invoked directly by the kernel. If we are invoked by a
         * container manager we assume the container manager knows what it is doing (for example, because it set up
         * some directories with different propagation modes). */
        if (detect_container() <= 0)
                if (mount(NULL, "/", NULL, MS_REC|MS_SHARED, NULL) < 0)
                        log_warning_errno(errno, "Failed to set up the root directory for shared mount propagation: %m");

        /* Create a few directories we always want around, Note that sd_booted() checks for /run/systemd/system, so
         * this mkdir really needs to stay for good, otherwise software that copied sd-daemon.c into their sources will
         * misdetect systemd. */
        (void) mkdir_label("/run/systemd", 0755);
        (void) mkdir_label("/run/systemd/system", 0755);

        /* Set up inaccessible items */
        (void) mkdir_label("/run/systemd/inaccessible", 0000);
        (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;
}