summaryrefslogtreecommitdiff
path: root/src/core/namespace.c
blob: 3234fab4bc7a85e4014c0078d4deee9c072522b5 (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
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
/***
  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 <sched.h>
#include <stdio.h>
#include <string.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <unistd.h>
#include <linux/fs.h>

#include "alloc-util.h"
#include "dev-setup.h"
#include "fd-util.h"
#include "fs-util.h"
#include "loopback-setup.h"
#include "missing.h"
#include "mkdir.h"
#include "mount-util.h"
#include "namespace.h"
#include "path-util.h"
#include "selinux-util.h"
#include "socket-util.h"
#include "string-table.h"
#include "string-util.h"
#include "strv.h"
#include "umask-util.h"
#include "user-util.h"
#include "util.h"

#define DEV_MOUNT_OPTIONS (MS_NOSUID|MS_STRICTATIME|MS_NOEXEC)

typedef enum MountMode {
        /* This is ordered by priority! */
        INACCESSIBLE,
        READONLY,
        PRIVATE_TMP,
        PRIVATE_VAR_TMP,
        PRIVATE_DEV,
        READWRITE,
} MountMode;

typedef struct BindMount {
        const char *path; /* stack memory, doesn't need to be freed explicitly */
        char *chased; /* malloc()ed memory, needs to be freed */
        MountMode mode;
        bool ignore; /* Ignore if path does not exist */
} BindMount;

typedef struct TargetMount {
        const char *path;
        MountMode mode;
        bool ignore; /* Ignore if path does not exist */
} TargetMount;

/* ProtectKernelTunables= option and the related filesystem APIs */
static const TargetMount protect_kernel_tunables_table[] = {
        { "/proc/sys",                  READONLY,       false },
        { "/proc/sysrq-trigger",        READONLY,       true  },
        { "/proc/latency_stats",        READONLY,       true  },
        { "/proc/mtrr",                 READONLY,       true  },
        { "/proc/apm",                  READONLY,       true  },
        { "/proc/acpi",                 READONLY,       true  },
        { "/proc/timer_stats",          READONLY,       true  },
        { "/proc/asound",               READONLY,       true  },
        { "/proc/bus",                  READONLY,       true  },
        { "/proc/fs",                   READONLY,       true  },
        { "/proc/irq",                  READONLY,       true  },
        { "/sys",                       READONLY,       false },
        { "/sys/kernel/debug",          READONLY,       true  },
        { "/sys/kernel/tracing",        READONLY,       true  },
        { "/sys/fs/cgroup",             READWRITE,      false }, /* READONLY is set by ProtectControlGroups= option */
};

static int append_mounts(BindMount **p, char **strv, MountMode mode) {
        char **i;

        assert(p);

        STRV_FOREACH(i, strv) {
                bool ignore = false;

                if (IN_SET(mode, INACCESSIBLE, READONLY, READWRITE) && startswith(*i, "-")) {
                        (*i)++;
                        ignore = true;
                }

                if (!path_is_absolute(*i))
                        return -EINVAL;

                (*p)->path = *i;
                (*p)->mode = mode;
                (*p)->ignore = ignore;
                (*p)++;
        }

        return 0;
}

static void append_protect_kernel_tunables(BindMount **p, const char *root_directory) {
        unsigned int i;

        assert(p);

        for (i = 0; i < ELEMENTSOF(protect_kernel_tunables_table); i++) {
                const TargetMount *t = &protect_kernel_tunables_table[i];
                (*p)->path = prefix_roota(root_directory, t->path);
                (*p)->mode = t->mode;
                (*p)->ignore = t->ignore;
                (*p)++;
        }
}

static int mount_path_compare(const void *a, const void *b) {
        const BindMount *p = a, *q = b;
        int d;

        /* If the paths are not equal, then order prefixes first */
        d = path_compare(p->path, q->path);
        if (d != 0)
                return d;

        /* If the paths are equal, check the mode */
        if (p->mode < q->mode)
                return -1;

        if (p->mode > q->mode)
                return 1;

        return 0;
}

static void drop_duplicates(BindMount *m, unsigned *n) {
        BindMount *f, *t, *previous;

        assert(m);
        assert(n);

        /* Drops duplicate entries. Expects that the array is properly ordered already. */

        for (f = m, t = m, previous = NULL; f < m+*n; f++) {

                /* The first one wins (which is the one with the more restrictive mode), see mount_path_compare()
                 * above. */
                if (previous && path_equal(f->path, previous->path)) {
                        log_debug("%s is duplicate.", f->path);
                        continue;
                }

                *t = *f;
                previous = t;
                t++;
        }

        *n = t - m;
}

static void drop_inaccessible(BindMount *m, unsigned *n) {
        BindMount *f, *t;
        const char *clear = NULL;

        assert(m);
        assert(n);

        /* Drops all entries obstructed by another entry further up the tree. Expects that the array is properly
         * ordered already. */

        for (f = m, t = m; f < m+*n; f++) {

                /* If we found a path set for INACCESSIBLE earlier, and this entry has it as prefix we should drop
                 * it, as inaccessible paths really should drop the entire subtree. */
                if (clear && path_startswith(f->path, clear)) {
                        log_debug("%s is masked by %s.", f->path, clear);
                        continue;
                }

                clear = f->mode == INACCESSIBLE ? f->path : NULL;

                *t = *f;
                t++;
        }

        *n = t - m;
}

static void drop_nop(BindMount *m, unsigned *n) {
        BindMount *f, *t;

        assert(m);
        assert(n);

        /* Drops all entries which have an immediate parent that has the same type, as they are redundant. Assumes the
         * list is ordered by prefixes. */

        for (f = m, t = m; f < m+*n; f++) {

                /* Only suppress such subtrees for READONLY and READWRITE entries */
                if (IN_SET(f->mode, READONLY, READWRITE)) {
                        BindMount *p;
                        bool found = false;

                        /* Now let's find the first parent of the entry we are looking at. */
                        for (p = t-1; p >= m; p--) {
                                if (path_startswith(f->path, p->path)) {
                                        found = true;
                                        break;
                                }
                        }

                        /* We found it, let's see if it's the same mode, if so, we can drop this entry */
                        if (found && p->mode == f->mode) {
                                log_debug("%s is redundant by %s", f->path, p->path);
                                continue;
                        }
                }

                *t = *f;
                t++;
        }

        *n = t - m;
}

static void drop_outside_root(const char *root_directory, BindMount *m, unsigned *n) {
        BindMount *f, *t;

        assert(m);
        assert(n);

        if (!root_directory)
                return;

        /* Drops all mounts that are outside of the root directory. */

        for (f = m, t = m; f < m+*n; f++) {

                if (!path_startswith(f->path, root_directory)) {
                        log_debug("%s is outside of root directory.", f->path);
                        continue;
                }

                *t = *f;
                t++;
        }

        *n = t - m;
}

static int mount_dev(BindMount *m) {
        static const char devnodes[] =
                "/dev/null\0"
                "/dev/zero\0"
                "/dev/full\0"
                "/dev/random\0"
                "/dev/urandom\0"
                "/dev/tty\0";

        char temporary_mount[] = "/tmp/namespace-dev-XXXXXX";
        const char *d, *dev = NULL, *devpts = NULL, *devshm = NULL, *devhugepages = NULL, *devmqueue = NULL, *devlog = NULL, *devptmx = NULL;
        _cleanup_umask_ mode_t u;
        int r;

        assert(m);

        u = umask(0000);

        if (!mkdtemp(temporary_mount))
                return -errno;

        dev = strjoina(temporary_mount, "/dev");
        (void) mkdir(dev, 0755);
        if (mount("tmpfs", dev, "tmpfs", DEV_MOUNT_OPTIONS, "mode=755") < 0) {
                r = -errno;
                goto fail;
        }

        devpts = strjoina(temporary_mount, "/dev/pts");
        (void) mkdir(devpts, 0755);
        if (mount("/dev/pts", devpts, NULL, MS_BIND, NULL) < 0) {
                r = -errno;
                goto fail;
        }

        devptmx = strjoina(temporary_mount, "/dev/ptmx");
        if (symlink("pts/ptmx", devptmx) < 0) {
                r = -errno;
                goto fail;
        }

        devshm = strjoina(temporary_mount, "/dev/shm");
        (void) mkdir(devshm, 01777);
        r = mount("/dev/shm", devshm, NULL, MS_BIND, NULL);
        if (r < 0) {
                r = -errno;
                goto fail;
        }

        devmqueue = strjoina(temporary_mount, "/dev/mqueue");
        (void) mkdir(devmqueue, 0755);
        (void) mount("/dev/mqueue", devmqueue, NULL, MS_BIND, NULL);

        devhugepages = strjoina(temporary_mount, "/dev/hugepages");
        (void) mkdir(devhugepages, 0755);
        (void) mount("/dev/hugepages", devhugepages, NULL, MS_BIND, NULL);

        devlog = strjoina(temporary_mount, "/dev/log");
        (void) symlink("/run/systemd/journal/dev-log", devlog);

        NULSTR_FOREACH(d, devnodes) {
                _cleanup_free_ char *dn = NULL;
                struct stat st;

                r = stat(d, &st);
                if (r < 0) {

                        if (errno == ENOENT)
                                continue;

                        r = -errno;
                        goto fail;
                }

                if (!S_ISBLK(st.st_mode) &&
                    !S_ISCHR(st.st_mode)) {
                        r = -EINVAL;
                        goto fail;
                }

                if (st.st_rdev == 0)
                        continue;

                dn = strappend(temporary_mount, d);
                if (!dn) {
                        r = -ENOMEM;
                        goto fail;
                }

                mac_selinux_create_file_prepare(d, st.st_mode);
                r = mknod(dn, st.st_mode, st.st_rdev);
                mac_selinux_create_file_clear();

                if (r < 0) {
                        r = -errno;
                        goto fail;
                }
        }

        dev_setup(temporary_mount, UID_INVALID, GID_INVALID);

        /* Create the /dev directory if missing. It is more likely to be
         * missing when the service is started with RootDirectory. This is
         * consistent with mount units creating the mount points when missing.
         */
        (void) mkdir_p_label(m->path, 0755);

        /* Unmount everything in old /dev */
        umount_recursive(m->path, 0);
        if (mount(dev, m->path, NULL, MS_MOVE, NULL) < 0) {
                r = -errno;
                goto fail;
        }

        rmdir(dev);
        rmdir(temporary_mount);

        return 0;

fail:
        if (devpts)
                umount(devpts);

        if (devshm)
                umount(devshm);

        if (devhugepages)
                umount(devhugepages);

        if (devmqueue)
                umount(devmqueue);

        umount(dev);
        rmdir(dev);
        rmdir(temporary_mount);

        return r;
}

static int apply_mount(
                BindMount *m,
                const char *tmp_dir,
                const char *var_tmp_dir) {

        const char *what;
        int r;

        assert(m);

        log_debug("Applying namespace mount on %s", m->path);

        switch (m->mode) {

        case INACCESSIBLE: {
                struct stat target;

                /* First, get rid of everything that is below if there
                 * is anything... Then, overmount it with an
                 * inaccessible path. */
                (void) umount_recursive(m->path, 0);

                if (lstat(m->path, &target) < 0)
                        return log_debug_errno(errno, "Failed to lstat() %s to determine what to mount over it: %m", m->path);

                what = mode_to_inaccessible_node(target.st_mode);
                if (!what) {
                        log_debug("File type not supported for inaccessible mounts. Note that symlinks are not allowed");
                        return -ELOOP;
                }
                break;
        }

        case READONLY:
        case READWRITE:

                r = path_is_mount_point(m->path, 0);
                if (r < 0)
                        return log_debug_errno(r, "Failed to determine whether %s is already a mount point: %m", m->path);
                if (r > 0) /* Nothing to do here, it is already a mount. We just later toggle the MS_RDONLY bit for the mount point if needed. */
                        return 0;

                /* This isn't a mount point yet, let's make it one. */
                what = m->path;
                break;

        case PRIVATE_TMP:
                what = tmp_dir;
                break;

        case PRIVATE_VAR_TMP:
                what = var_tmp_dir;
                break;

        case PRIVATE_DEV:
                return mount_dev(m);

        default:
                assert_not_reached("Unknown mode");
        }

        assert(what);

        if (mount(what, m->path, NULL, MS_BIND|MS_REC, NULL) < 0)
                return log_debug_errno(errno, "Failed to mount %s to %s: %m", what, m->path);

        log_debug("Successfully mounted %s to %s", what, m->path);
        return 0;
}

static int make_read_only(BindMount *m, char **blacklist) {
        int r = 0;

        assert(m);

        if (IN_SET(m->mode, INACCESSIBLE, READONLY))
                r = bind_remount_recursive(m->path, true, blacklist);
        else if (m->mode == PRIVATE_DEV) { /* Can be readonly but the submounts can't*/
                if (mount(NULL, m->path, NULL, MS_REMOUNT|DEV_MOUNT_OPTIONS|MS_RDONLY, NULL) < 0)
                        r = -errno;
        } else
                return 0;

        /* Not that we only turn on the MS_RDONLY flag here, we never turn it off. Something that was marked read-only
         * already stays this way. This improves compatibility with container managers, where we won't attempt to undo
         * read-only mounts already applied. */

        return r;
}

static int chase_all_symlinks(const char *root_directory, BindMount *m, unsigned *n) {
        BindMount *f, *t;
        int r;

        assert(m);
        assert(n);

        /* Since mount() will always follow symlinks and we need to take the different root directory into account we
         * chase the symlinks on our own first. This call wil do so for all entries and remove all entries where we
         * can't resolve the path, and which have been marked for such removal. */

        for (f = m, t = m; f < m+*n; f++) {

                r = chase_symlinks(f->path, root_directory, &f->chased);
                if (r == -ENOENT && f->ignore) /* Doesn't exist? Then remove it! */
                        continue;
                if (r < 0)
                        return log_debug_errno(r, "Failed to chase symlinks for %s: %m", f->path);

                if (path_equal(f->path, f->chased))
                        f->chased = mfree(f->chased);
                else {
                        log_debug("Chased %s → %s", f->path, f->chased);
                        f->path = f->chased;
                }

                *t = *f;
                t++;
        }

        *n = t - m;
        return 0;
}

static unsigned namespace_calculate_mounts(
                char** read_write_paths,
                char** read_only_paths,
                char** inaccessible_paths,
                const char* tmp_dir,
                const char* var_tmp_dir,
                bool private_dev,
                bool protect_sysctl,
                bool protect_cgroups,
                ProtectHome protect_home,
                ProtectSystem protect_system) {

        return !!tmp_dir + !!var_tmp_dir +
                strv_length(read_write_paths) +
                strv_length(read_only_paths) +
                strv_length(inaccessible_paths) +
                private_dev +
                (protect_sysctl ? ELEMENTSOF(protect_kernel_tunables_table) : 0) +
                (protect_cgroups ? 1 : 0) +
                (protect_home != PROTECT_HOME_NO || protect_system == PROTECT_SYSTEM_STRICT ? 3 : 0) +
                (protect_system == PROTECT_SYSTEM_STRICT ?
                 (2 + !private_dev + !protect_sysctl) :
                 ((protect_system != PROTECT_SYSTEM_NO ? 3 : 0) +
                  (protect_system == PROTECT_SYSTEM_FULL ? 1 : 0)));
}

int setup_namespace(
                const char* root_directory,
                char** read_write_paths,
                char** read_only_paths,
                char** inaccessible_paths,
                const char* tmp_dir,
                const char* var_tmp_dir,
                bool private_dev,
                bool protect_sysctl,
                bool protect_cgroups,
                ProtectHome protect_home,
                ProtectSystem protect_system,
                unsigned long mount_flags) {

        BindMount *m, *mounts = NULL;
        bool make_slave = false;
        unsigned n;
        int r = 0;

        if (mount_flags == 0)
                mount_flags = MS_SHARED;

        n = namespace_calculate_mounts(read_write_paths,
                                       read_only_paths,
                                       inaccessible_paths,
                                       tmp_dir, var_tmp_dir,
                                       private_dev, protect_sysctl,
                                       protect_cgroups, protect_home,
                                       protect_system);

        /* Set mount slave mode */
        if (root_directory || n > 0)
                make_slave = true;

        if (n > 0) {
                m = mounts = (BindMount *) alloca0(n * sizeof(BindMount));
                r = append_mounts(&m, read_write_paths, READWRITE);
                if (r < 0)
                        return r;

                r = append_mounts(&m, read_only_paths, READONLY);
                if (r < 0)
                        return r;

                r = append_mounts(&m, inaccessible_paths, INACCESSIBLE);
                if (r < 0)
                        return r;

                if (tmp_dir) {
                        m->path = prefix_roota(root_directory, "/tmp");
                        m->mode = PRIVATE_TMP;
                        m++;
                }

                if (var_tmp_dir) {
                        m->path = prefix_roota(root_directory, "/var/tmp");
                        m->mode = PRIVATE_VAR_TMP;
                        m++;
                }

                if (private_dev) {
                        m->path = prefix_roota(root_directory, "/dev");
                        m->mode = PRIVATE_DEV;
                        m++;
                }

                if (protect_sysctl)
                        append_protect_kernel_tunables(&m, root_directory);

                if (protect_cgroups) {
                        m->path = prefix_roota(root_directory, "/sys/fs/cgroup");
                        m->mode = READONLY;
                        m++;
                }

                if (protect_home != PROTECT_HOME_NO || protect_system == PROTECT_SYSTEM_STRICT) {
                        const char *home_dir, *run_user_dir, *root_dir;

                        /* If protection of $HOME and $XDG_RUNTIME_DIR is requested, then go for it. If we are in
                         * strict system protection mode, then also add entries for these directories, but mark them
                         * writable. This is because we want ProtectHome= and ProtectSystem= to be fully orthogonal. */

                        home_dir = prefix_roota(root_directory, "/home");
                        home_dir = strjoina("-", home_dir);
                        run_user_dir = prefix_roota(root_directory, "/run/user");
                        run_user_dir = strjoina("-", run_user_dir);
                        root_dir = prefix_roota(root_directory, "/root");
                        root_dir = strjoina("-", root_dir);

                        r = append_mounts(&m, STRV_MAKE(home_dir, run_user_dir, root_dir),
                                protect_home == PROTECT_HOME_READ_ONLY ? READONLY :
                                protect_home == PROTECT_HOME_YES ? INACCESSIBLE : READWRITE);
                        if (r < 0)
                                return r;
                }

                if (protect_system == PROTECT_SYSTEM_STRICT) {
                        /* In strict mode, we mount everything read-only, except for /proc, /dev, /sys which are the
                         * kernel API VFS, which are left writable, but PrivateDevices= + ProtectKernelTunables=
                         * protect those, and these options should be fully orthogonal. (And of course /home and
                         * friends are also left writable, as ProtectHome= shall manage those, orthogonally, see
                         * above). */

                        m->path = prefix_roota(root_directory, "/");
                        m->mode = READONLY;
                        m++;

                        m->path = prefix_roota(root_directory, "/proc");
                        m->mode = READWRITE;
                        m++;

                        if (!private_dev) {
                                m->path = prefix_roota(root_directory, "/dev");
                                m->mode = READWRITE;
                                m++;
                        }
                        if (!protect_sysctl) {
                                m->path = prefix_roota(root_directory, "/sys");
                                m->mode = READWRITE;
                                m++;
                        }

                } else if (protect_system != PROTECT_SYSTEM_NO) {
                        const char *usr_dir, *boot_dir, *efi_dir, *etc_dir;

                        /* In any other mode we simply mark the relevant three directories ready-only. */

                        usr_dir = prefix_roota(root_directory, "/usr");
                        boot_dir = prefix_roota(root_directory, "/boot");
                        boot_dir = strjoina("-", boot_dir);
                        efi_dir = prefix_roota(root_directory, "/efi");
                        efi_dir = strjoina("-", efi_dir);
                        etc_dir = prefix_roota(root_directory, "/etc");

                        r = append_mounts(&m, protect_system == PROTECT_SYSTEM_FULL
                                          ? STRV_MAKE(usr_dir, boot_dir, efi_dir, etc_dir)
                                          : STRV_MAKE(usr_dir, boot_dir, efi_dir), READONLY);
                        if (r < 0)
                                return r;
                }

                assert(mounts + n == m);

                /* Resolve symlinks manually first, as mount() will always follow them relative to the host's
                 * root. Moreover we want to suppress duplicates based on the resolved paths. This of course is a bit
                 * racy. */
                r = chase_all_symlinks(root_directory, mounts, &n);
                if (r < 0)
                        goto finish;

                qsort(mounts, n, sizeof(BindMount), mount_path_compare);

                drop_duplicates(mounts, &n);
                drop_outside_root(root_directory, mounts, &n);
                drop_inaccessible(mounts, &n);
                drop_nop(mounts, &n);
        }

        if (unshare(CLONE_NEWNS) < 0) {
                r = -errno;
                goto finish;
        }

        if (make_slave) {
                /* Remount / as SLAVE so that nothing now mounted in the namespace
                   shows up in the parent */
                if (mount(NULL, "/", NULL, MS_SLAVE|MS_REC, NULL) < 0) {
                        r = -errno;
                        goto finish;
                }
        }

        if (root_directory) {
                /* Turn directory into bind mount, if it isn't one yet */
                r = path_is_mount_point(root_directory, AT_SYMLINK_FOLLOW);
                if (r < 0)
                        goto finish;
                if (r == 0) {
                        if (mount(root_directory, root_directory, NULL, MS_BIND|MS_REC, NULL) < 0) {
                                r = -errno;
                                goto finish;
                        }
                }
        }

        if (n > 0) {
                char **blacklist;
                unsigned j;

                /* First round, add in all special mounts we need */
                for (m = mounts; m < mounts + n; ++m) {
                        r = apply_mount(m, tmp_dir, var_tmp_dir);
                        if (r < 0)
                                goto finish;
                }

                /* Create a blacklist we can pass to bind_mount_recursive() */
                blacklist = newa(char*, n+1);
                for (j = 0; j < n; j++)
                        blacklist[j] = (char*) mounts[j].path;
                blacklist[j] = NULL;

                /* Second round, flip the ro bits if necessary. */
                for (m = mounts; m < mounts + n; ++m) {
                        r = make_read_only(m, blacklist);
                        if (r < 0)
                                goto finish;
                }
        }

        if (root_directory) {
                /* MS_MOVE does not work on MS_SHARED so the remount MS_SHARED will be done later */
                r = mount_move_root(root_directory);
                if (r < 0)
                        goto finish;
        }

        /* Remount / as the desired mode. Not that this will not
         * reestablish propagation from our side to the host, since
         * what's disconnected is disconnected. */
        if (mount(NULL, "/", NULL, mount_flags | MS_REC, NULL) < 0) {
                r = -errno;
                goto finish;
        }

        r = 0;

finish:
        for (m = mounts; m < mounts + n; m++)
                free(m->chased);

        return r;
}

static int setup_one_tmp_dir(const char *id, const char *prefix, char **path) {
        _cleanup_free_ char *x = NULL;
        char bid[SD_ID128_STRING_MAX];
        sd_id128_t boot_id;
        int r;

        assert(id);
        assert(prefix);
        assert(path);

        /* We include the boot id in the directory so that after a
         * reboot we can easily identify obsolete directories. */

        r = sd_id128_get_boot(&boot_id);
        if (r < 0)
                return r;

        x = strjoin(prefix, "/systemd-private-", sd_id128_to_string(boot_id, bid), "-", id, "-XXXXXX", NULL);
        if (!x)
                return -ENOMEM;

        RUN_WITH_UMASK(0077)
                if (!mkdtemp(x))
                        return -errno;

        RUN_WITH_UMASK(0000) {
                char *y;

                y = strjoina(x, "/tmp");

                if (mkdir(y, 0777 | S_ISVTX) < 0)
                        return -errno;
        }

        *path = x;
        x = NULL;

        return 0;
}

int setup_tmp_dirs(const char *id, char **tmp_dir, char **var_tmp_dir) {
        char *a, *b;
        int r;

        assert(id);
        assert(tmp_dir);
        assert(var_tmp_dir);

        r = setup_one_tmp_dir(id, "/tmp", &a);
        if (r < 0)
                return r;

        r = setup_one_tmp_dir(id, "/var/tmp", &b);
        if (r < 0) {
                char *t;

                t = strjoina(a, "/tmp");
                rmdir(t);
                rmdir(a);

                free(a);
                return r;
        }

        *tmp_dir = a;
        *var_tmp_dir = b;

        return 0;
}

int setup_netns(int netns_storage_socket[2]) {
        _cleanup_close_ int netns = -1;
        int r, q;

        assert(netns_storage_socket);
        assert(netns_storage_socket[0] >= 0);
        assert(netns_storage_socket[1] >= 0);

        /* We use the passed socketpair as a storage buffer for our
         * namespace reference fd. Whatever process runs this first
         * shall create a new namespace, all others should just join
         * it. To serialize that we use a file lock on the socket
         * pair.
         *
         * It's a bit crazy, but hey, works great! */

        if (lockf(netns_storage_socket[0], F_LOCK, 0) < 0)
                return -errno;

        netns = receive_one_fd(netns_storage_socket[0], MSG_DONTWAIT);
        if (netns == -EAGAIN) {
                /* Nothing stored yet, so let's create a new namespace */

                if (unshare(CLONE_NEWNET) < 0) {
                        r = -errno;
                        goto fail;
                }

                loopback_setup();

                netns = open("/proc/self/ns/net", O_RDONLY|O_CLOEXEC|O_NOCTTY);
                if (netns < 0) {
                        r = -errno;
                        goto fail;
                }

                r = 1;

        } else if (netns < 0) {
                r = netns;
                goto fail;

        } else {
                /* Yay, found something, so let's join the namespace */
                if (setns(netns, CLONE_NEWNET) < 0) {
                        r = -errno;
                        goto fail;
                }

                r = 0;
        }

        q = send_one_fd(netns_storage_socket[1], netns, MSG_DONTWAIT);
        if (q < 0) {
                r = q;
                goto fail;
        }

fail:
        (void) lockf(netns_storage_socket[0], F_ULOCK, 0);
        return r;
}

static const char *const protect_home_table[_PROTECT_HOME_MAX] = {
        [PROTECT_HOME_NO] = "no",
        [PROTECT_HOME_YES] = "yes",
        [PROTECT_HOME_READ_ONLY] = "read-only",
};

DEFINE_STRING_TABLE_LOOKUP(protect_home, ProtectHome);

static const char *const protect_system_table[_PROTECT_SYSTEM_MAX] = {
        [PROTECT_SYSTEM_NO] = "no",
        [PROTECT_SYSTEM_YES] = "yes",
        [PROTECT_SYSTEM_FULL] = "full",
        [PROTECT_SYSTEM_STRICT] = "strict",
};

DEFINE_STRING_TABLE_LOOKUP(protect_system, ProtectSystem);