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
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
|
/*
* Copyright (C) 2014-2021 Firejail Authors
*
* This file is part of firejail project
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include "firejail.h"
#include "../include/seccomp.h"
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/wait.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/resource.h>
#include <sys/types.h>
#include <dirent.h>
#include <errno.h>
#include <fcntl.h>
#include <syscall.h>
#include <sched.h>
#ifndef CLONE_NEWUSER
#define CLONE_NEWUSER 0x10000000
#endif
#include <sys/prctl.h>
#ifndef PR_SET_NO_NEW_PRIVS
#define PR_SET_NO_NEW_PRIVS 38
#endif
#ifndef PR_GET_NO_NEW_PRIVS
#define PR_GET_NO_NEW_PRIVS 39
#endif
#ifdef HAVE_APPARMOR
#include <sys/apparmor.h>
#endif
#ifdef HAVE_GCOV
#include <gcov.h>
#endif
static int force_nonewprivs = 0;
static int monitored_pid = 0;
static void sandbox_handler(int sig){
usleep(10000); // don't race to print a message
fmessage("\nChild received signal %d, shutting down the sandbox...\n", sig);
// broadcast sigterm to all processes in the group
kill(-1, SIGTERM);
sleep(1);
if (monitored_pid) {
int monsec = 9;
char *monfile;
if (asprintf(&monfile, "/proc/%d/cmdline", monitored_pid) == -1)
errExit("asprintf");
while (monsec) {
FILE *fp = fopen(monfile, "re");
if (!fp)
break;
char c;
size_t count = fread(&c, 1, 1, fp);
fclose(fp);
if (count == 0)
break;
if (arg_debug)
printf("Waiting on PID %d to finish\n", monitored_pid);
sleep(1);
monsec--;
}
free(monfile);
}
// broadcast a SIGKILL
kill(-1, SIGKILL);
flush_stdin();
exit(sig);
}
static void install_handler(void) {
struct sigaction sga;
// block SIGTERM while handling SIGINT
sigemptyset(&sga.sa_mask);
sigaddset(&sga.sa_mask, SIGTERM);
sga.sa_handler = sandbox_handler;
sga.sa_flags = 0;
sigaction(SIGINT, &sga, NULL);
// block SIGINT while handling SIGTERM
sigemptyset(&sga.sa_mask);
sigaddset(&sga.sa_mask, SIGINT);
sga.sa_handler = sandbox_handler;
sga.sa_flags = 0;
sigaction(SIGTERM, &sga, NULL);
}
static void set_caps(void) {
if (arg_caps_drop_all)
caps_drop_all();
else if (arg_caps_drop)
caps_drop_list(arg_caps_list);
else if (arg_caps_keep)
caps_keep_list(arg_caps_list);
else if (arg_caps_default_filter)
caps_default_filter();
// drop discretionary access control capabilities for root sandboxes
// if caps.keep, the user has to set it manually in the list
if (!arg_caps_keep)
caps_drop_dac_override();
}
#ifdef HAVE_APPARMOR
void set_apparmor(void) {
EUID_ASSERT();
if (checkcfg(CFG_APPARMOR) && arg_apparmor) {
if (aa_change_onexec("firejail-default")) {
fwarning("Cannot confine the application using AppArmor.\n"
"Maybe firejail-default AppArmor profile is not loaded into the kernel.\n"
"As root, run \"aa-enforce firejail-default\" to load it.\n");
}
else if (arg_debug)
printf("AppArmor enabled\n");
}
}
#endif
static void seccomp_debug(void) {
if (arg_debug == 0)
return;
EUID_USER();
printf("Seccomp directory:\n");
ls(RUN_SECCOMP_DIR);
struct stat s;
if (stat(RUN_SECCOMP_LIST, &s) == 0) {
printf("Active seccomp files:\n");
cat(RUN_SECCOMP_LIST);
}
else
printf("No active seccomp files\n");
EUID_ROOT();
}
static void save_nogroups(void) {
if (arg_nogroups == 0)
return;
FILE *fp = fopen(RUN_GROUPS_CFG, "wxe");
if (fp) {
fprintf(fp, "\n");
SET_PERMS_STREAM(fp, 0, 0, 0644); // assume mode 0644
fclose(fp);
}
else {
fprintf(stderr, "Error: cannot save nogroups state\n");
exit(1);
}
}
static void save_nonewprivs(void) {
if (arg_nonewprivs == 0)
return;
FILE *fp = fopen(RUN_NONEWPRIVS_CFG, "wxe");
if (fp) {
fprintf(fp, "\n");
SET_PERMS_STREAM(fp, 0, 0, 0644); // assume mode 0644
fclose(fp);
}
else {
fprintf(stderr, "Error: cannot save nonewprivs state\n");
exit(1);
}
}
static void save_umask(void) {
FILE *fp = fopen(RUN_UMASK_FILE, "wxe");
if (fp) {
fprintf(fp, "%o\n", orig_umask);
SET_PERMS_STREAM(fp, 0, 0, 0644); // assume mode 0644
fclose(fp);
}
else {
fprintf(stderr, "Error: cannot save umask\n");
exit(1);
}
}
static char *create_join_file(void) {
int fd = open(RUN_JOIN_FILE, O_RDWR|O_CREAT|O_EXCL|O_CLOEXEC, S_IRUSR | S_IWRITE | S_IRGRP | S_IROTH);
if (fd == -1)
errExit("open");
if (ftruncate(fd, 1) == -1)
errExit("ftruncate");
char *rv = mmap(NULL, 1, PROT_WRITE, MAP_SHARED, fd, 0);
if (rv == MAP_FAILED)
errExit("mmap");
close(fd);
return rv;
}
static void sandbox_if_up(Bridge *br) {
assert(br);
if (!br->configured)
return;
char *dev = br->devsandbox;
net_if_up(dev);
if (br->arg_ip_none == 1); // do nothing
else if (br->arg_ip_none == 0 && br->macvlan == 0) {
if (br->ipsandbox == br->ip) {
fprintf(stderr, "Error: %d.%d.%d.%d is interface %s address, exiting...\n", PRINT_IP(br->ipsandbox), br->dev);
exit(1);
}
// just assign the address
assert(br->ipsandbox);
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(br->ipsandbox), dev);
net_config_interface(dev, br->ipsandbox, br->mask, br->mtu);
arp_announce(dev, br);
}
else if (br->arg_ip_none == 0 && br->macvlan == 1) {
// reassign the macvlan address
if (br->ipsandbox == 0)
// ip address assigned by arp-scan for a macvlan device
br->ipsandbox = arp_assign(dev, br); //br->ip, br->mask);
else {
if (br->ipsandbox == br->ip) {
fprintf(stderr, "Error: %d.%d.%d.%d is interface %s address, exiting...\n", PRINT_IP(br->ipsandbox), br->dev);
exit(1);
}
if (br->ipsandbox == cfg.defaultgw) {
fprintf(stderr, "Error: %d.%d.%d.%d is the default gateway, exiting...\n", PRINT_IP(br->ipsandbox));
exit(1);
}
uint32_t rv = arp_check(dev, br->ipsandbox);
if (rv) {
fprintf(stderr, "Error: the address %d.%d.%d.%d is already in use, exiting...\n", PRINT_IP(br->ipsandbox));
exit(1);
}
}
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(br->ipsandbox), dev);
net_config_interface(dev, br->ipsandbox, br->mask, br->mtu);
arp_announce(dev, br);
}
if (br->ip6sandbox)
net_if_ip6(dev, br->ip6sandbox);
}
static void chk_chroot(void) {
// if we are starting firejail inside some other container technology, we don't care about this
if (env_get("container"))
return;
// check if this is a regular chroot
struct stat s;
if (stat("/", &s) == 0) {
if (s.st_ino != 2)
return;
}
fprintf(stderr, "Error: cannot mount filesystem as slave\n");
exit(1);
}
static int monitor_application(pid_t app_pid) {
EUID_ASSERT();
monitored_pid = app_pid;
// block signals and install handler
sigset_t oldmask, newmask;
sigemptyset(&oldmask);
sigemptyset(&newmask);
sigaddset(&newmask, SIGTERM);
sigaddset(&newmask, SIGINT);
sigprocmask(SIG_BLOCK, &newmask, &oldmask);
install_handler();
// handle --timeout
int options = 0;;
unsigned timeout = 0;
if (cfg.timeout) {
options = WNOHANG;
timeout = cfg.timeout;
sleep(1);
}
int status = 0;
int app_status = 0;
while (monitored_pid) {
usleep(20000);
char *msg;
if (asprintf(&msg, "monitoring pid %d\n", monitored_pid) == -1)
errExit("asprintf");
logmsg(msg);
if (arg_debug)
printf("%s\n", msg);
free(msg);
pid_t rv;
do {
// handle signals asynchronously
sigprocmask(SIG_SETMASK, &oldmask, NULL);
rv = waitpid(-1, &status, options);
// block signals again
sigprocmask(SIG_BLOCK, &newmask, NULL);
if (rv == -1) { // we can get here if we have processes joining the sandbox (ECHILD)
sleep(1);
break;
}
else if (rv == app_pid)
app_status = status;
// handle --timeout
if (options) {
if (--timeout == 0) {
// SIGTERM might fail if the process ignores it (SIG_IGN)
// we give it 100ms to close properly and after that we SIGKILL it
kill(-1, SIGTERM);
usleep(100000);
kill(-1, SIGKILL);
flush_stdin();
_exit(1);
}
else
sleep(1);
}
}
while(rv != monitored_pid);
if (arg_debug)
printf("Sandbox monitor: waitpid %d retval %d status %d\n", monitored_pid, rv, status);
DIR *dir;
if (!(dir = opendir("/proc"))) {
// sleep 2 seconds and try again
sleep(2);
if (!(dir = opendir("/proc"))) {
fprintf(stderr, "Error: cannot open /proc directory\n");
exit(1);
}
}
struct dirent *entry;
monitored_pid = 0;
while ((entry = readdir(dir)) != NULL) {
unsigned pid;
if (sscanf(entry->d_name, "%u", &pid) != 1)
continue;
if (pid == 1)
continue;
if ((pid_t) pid == dhclient4_pid || (pid_t) pid == dhclient6_pid)
continue;
// todo: make this generic
// Dillo browser leaves a dpid process running, we need to shut it down
int found = 0;
if (strcmp(cfg.command_name, "dillo") == 0) {
char *pidname = pid_proc_comm(pid);
if (pidname && strcmp(pidname, "dpid") == 0)
found = 1;
free(pidname);
}
if (found)
break;
monitored_pid = pid;
break;
}
closedir(dir);
if (monitored_pid != 0 && arg_debug)
printf("Sandbox monitor: monitoring %d\n", monitored_pid);
}
// return the appropriate exit status.
return arg_deterministic_exit_code ? app_status : status;
}
static void print_time(void) {
float delta = timetrace_end();
fmessage("Child process initialized in %.02f ms\n", delta);
}
// check execute permissions for the program
// this is done typically by the shell
// we are here because of --shell=none
// we duplicate execvp functionality (man execvp):
// [...] if the specified
// filename does not contain a slash (/) character. The file is sought
// in the colon-separated list of directory pathnames specified in the
// PATH environment variable.
static int ok_to_run(const char *program) {
if (strstr(program, "/")) {
if (access(program, X_OK) == 0) // it will also dereference symlinks
return 1;
}
else { // search $PATH
const char *path1 = env_get("PATH");
if (path1) {
if (arg_debug)
printf("Searching $PATH for %s\n", program);
char *path2 = strdup(path1);
if (!path2)
errExit("strdup");
// use path2 to count the entries
char *ptr = strtok(path2, ":");
while (ptr) {
char *fname;
if (asprintf(&fname, "%s/%s", ptr, program) == -1)
errExit("asprintf");
if (arg_debug)
printf("trying #%s#\n", fname);
struct stat s;
int rv = stat(fname, &s);
if (rv == 0) {
if (access(fname, X_OK) == 0) {
free(path2);
free(fname);
return 1;
}
else
fprintf(stderr, "Error: execute permission denied for %s\n", fname);
free(fname);
break;
}
free(fname);
ptr = strtok(NULL, ":");
}
free(path2);
}
}
return 0;
}
void start_application(int no_sandbox, int fd, char *set_sandbox_status) {
// set environment
if (no_sandbox == 0)
env_defaults();
env_apply_all();
// restore original umask
umask(orig_umask);
if (arg_debug) {
printf("Starting application\n");
printf("LD_PRELOAD=%s\n", getenv("LD_PRELOAD"));
}
//****************************************
// start the program without using a shell
//****************************************
if (arg_shell_none) {
if (arg_debug) {
int i;
for (i = cfg.original_program_index; i < cfg.original_argc; i++) {
if (cfg.original_argv[i] == NULL)
break;
printf("execvp argument %d: %s\n", i - cfg.original_program_index, cfg.original_argv[i]);
}
}
if (cfg.original_program_index == 0) {
fprintf(stderr, "Error: --shell=none configured, but no program specified\n");
exit(1);
}
if (!arg_command && !arg_quiet)
print_time();
if (ok_to_run(cfg.original_argv[cfg.original_program_index]) == 0) {
fprintf(stderr, "Error: no suitable %s executable found\n", cfg.original_argv[cfg.original_program_index]);
exit(1);
}
#ifdef HAVE_GCOV
__gcov_dump();
#endif
seccomp_install_filters();
if (set_sandbox_status)
*set_sandbox_status = SANDBOX_DONE;
execvp(cfg.original_argv[cfg.original_program_index], &cfg.original_argv[cfg.original_program_index]);
}
//****************************************
// start the program using a shell
//****************************************
else {
assert(cfg.shell);
char *arg[5];
int index = 0;
arg[index++] = cfg.shell;
if (cfg.command_line) {
if (arg_debug)
printf("Running %s command through %s\n", cfg.command_line, cfg.shell);
arg[index++] = "-c";
if (arg_doubledash)
arg[index++] = "--";
arg[index++] = cfg.command_line;
}
else if (login_shell) {
if (arg_debug)
printf("Starting %s login shell\n", cfg.shell);
arg[index++] = "-l";
}
else if (arg_debug)
printf("Starting %s shell\n", cfg.shell);
assert(index < 5);
arg[index] = NULL;
if (arg_debug) {
char *msg;
if (asprintf(&msg, "sandbox %d, execvp into %s",
sandbox_pid, cfg.command_line ? cfg.command_line : cfg.shell) == -1)
errExit("asprintf");
logmsg(msg);
free(msg);
int i;
for (i = 0; i < 5; i++) {
if (arg[i] == NULL)
break;
printf("execvp argument %d: %s\n", i, arg[i]);
}
}
if (!arg_command && !arg_quiet)
print_time();
#ifdef HAVE_GCOV
__gcov_dump();
#endif
seccomp_install_filters();
if (set_sandbox_status)
*set_sandbox_status = SANDBOX_DONE;
execvp(arg[0], arg);
// join sandbox without shell in the mount namespace
if (fd > -1)
fexecve(fd, arg, environ);
}
perror("Cannot start application");
exit(1);
}
static void enforce_filters(void) {
fmessage("\n** Warning: dropping all Linux capabilities and setting NO_NEW_PRIVS prctl **\n\n");
// enforce NO_NEW_PRIVS
arg_nonewprivs = 1;
force_nonewprivs = 1;
// disable all capabilities
arg_caps_drop_all = 1;
// drop all supplementary groups; /etc/group file inside chroot
// is controlled by a regular usr
arg_nogroups = 1;
}
int sandbox(void* sandbox_arg) {
// Get rid of unused parameter warning
(void)sandbox_arg;
pid_t child_pid = getpid();
if (arg_debug)
printf("Initializing child process\n");
// close each end of the unused pipes
close(parent_to_child_fds[1]);
close(child_to_parent_fds[0]);
// wait for parent to do base setup
wait_for_other(parent_to_child_fds[0]);
if (arg_debug && child_pid == 1)
printf("PID namespace installed\n");
//****************************
// set hostname
//****************************
if (cfg.hostname) {
if (sethostname(cfg.hostname, strlen(cfg.hostname)) < 0)
errExit("sethostname");
}
//****************************
// mount namespace
//****************************
// mount events are not forwarded between the host the sandbox
if (mount(NULL, "/", NULL, MS_SLAVE | MS_REC, NULL) < 0) {
chk_chroot();
}
// ... and mount a tmpfs on top of /run/firejail/mnt directory
preproc_mount_mnt_dir();
// bind-mount firejail binaries and helper programs
if (mount(LIBDIR "/firejail", RUN_FIREJAIL_LIB_DIR, NULL, MS_BIND, NULL) < 0 ||
mount(NULL, RUN_FIREJAIL_LIB_DIR, NULL, MS_RDONLY|MS_NOSUID|MS_NODEV|MS_BIND|MS_REMOUNT, NULL) < 0)
errExit("mounting " RUN_FIREJAIL_LIB_DIR);
// keep a copy of dhclient executable before the filesystem is modified
dhcp_store_exec();
//****************************
// log sandbox data
//****************************
if (cfg.name)
fs_logger2("sandbox name:", cfg.name);
fs_logger2int("sandbox pid:", (int) sandbox_pid);
if (cfg.chrootdir)
fs_logger("sandbox filesystem: chroot");
else if (arg_overlay)
fs_logger("sandbox filesystem: overlay");
else
fs_logger("sandbox filesystem: local");
fs_logger("install mount namespace");
//****************************
// netfilter
//****************************
if (arg_netfilter && any_bridge_configured()) { // assuming by default the client filter
netfilter(arg_netfilter_file);
}
if (arg_netfilter6 && any_bridge_configured()) { // assuming by default the client filter
netfilter6(arg_netfilter6_file);
}
//****************************
// networking
//****************************
int gw_cfg_failed = 0; // default gw configuration flag
if (arg_nonetwork) {
net_if_up("lo");
if (arg_debug)
printf("Network namespace enabled, only loopback interface available\n");
}
else if (arg_netns) {
netns(arg_netns);
if (arg_debug)
printf("Network namespace '%s' activated\n", arg_netns);
}
else if (any_bridge_configured() || any_interface_configured()) {
// configure lo and eth0...eth3
net_if_up("lo");
if (mac_not_zero(cfg.bridge0.macsandbox))
net_config_mac(cfg.bridge0.devsandbox, cfg.bridge0.macsandbox);
sandbox_if_up(&cfg.bridge0);
if (mac_not_zero(cfg.bridge1.macsandbox))
net_config_mac(cfg.bridge1.devsandbox, cfg.bridge1.macsandbox);
sandbox_if_up(&cfg.bridge1);
if (mac_not_zero(cfg.bridge2.macsandbox))
net_config_mac(cfg.bridge2.devsandbox, cfg.bridge2.macsandbox);
sandbox_if_up(&cfg.bridge2);
if (mac_not_zero(cfg.bridge3.macsandbox))
net_config_mac(cfg.bridge3.devsandbox, cfg.bridge3.macsandbox);
sandbox_if_up(&cfg.bridge3);
// moving an interface in a namespace using --interface will reset the interface configuration;
// we need to put the configuration back
if (cfg.interface0.configured && cfg.interface0.ip) {
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(cfg.interface0.ip), cfg.interface0.dev);
net_config_interface(cfg.interface0.dev, cfg.interface0.ip, cfg.interface0.mask, cfg.interface0.mtu);
}
if (cfg.interface1.configured && cfg.interface1.ip) {
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(cfg.interface1.ip), cfg.interface1.dev);
net_config_interface(cfg.interface1.dev, cfg.interface1.ip, cfg.interface1.mask, cfg.interface1.mtu);
}
if (cfg.interface2.configured && cfg.interface2.ip) {
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(cfg.interface2.ip), cfg.interface2.dev);
net_config_interface(cfg.interface2.dev, cfg.interface2.ip, cfg.interface2.mask, cfg.interface2.mtu);
}
if (cfg.interface3.configured && cfg.interface3.ip) {
if (arg_debug)
printf("Configuring %d.%d.%d.%d address on interface %s\n", PRINT_IP(cfg.interface3.ip), cfg.interface3.dev);
net_config_interface(cfg.interface3.dev, cfg.interface3.ip, cfg.interface3.mask, cfg.interface3.mtu);
}
// add a default route
if (cfg.defaultgw) {
// set the default route
if (net_add_route(0, 0, cfg.defaultgw)) {
fwarning("cannot configure default route\n");
gw_cfg_failed = 1;
}
}
if (arg_debug)
printf("Network namespace enabled\n");
}
// print network configuration
if (!arg_quiet) {
if (any_bridge_configured() || any_interface_configured() || cfg.defaultgw || cfg.dns1) {
fmessage("\n");
if (any_bridge_configured() || any_interface_configured()) {
if (arg_scan)
sbox_run(SBOX_ROOT | SBOX_CAPS_NETWORK | SBOX_SECCOMP, 3, PATH_FNET, "printif", "scan");
else
sbox_run(SBOX_ROOT | SBOX_CAPS_NETWORK | SBOX_SECCOMP, 2, PATH_FNET, "printif");
}
if (cfg.defaultgw != 0) {
if (gw_cfg_failed)
fmessage("Default gateway configuration failed\n");
else
fmessage("Default gateway %d.%d.%d.%d\n", PRINT_IP(cfg.defaultgw));
}
if (cfg.dns1 != NULL)
fmessage("DNS server %s\n", cfg.dns1);
if (cfg.dns2 != NULL)
fmessage("DNS server %s\n", cfg.dns2);
if (cfg.dns3 != NULL)
fmessage("DNS server %s\n", cfg.dns3);
if (cfg.dns4 != NULL)
fmessage("DNS server %s\n", cfg.dns4);
fmessage("\n");
}
}
// load IBUS env variables
if (arg_nonetwork || any_bridge_configured() || any_interface_configured()) {
// do nothing - there are problems with ibus version 1.5.11
}
else {
EUID_USER();
env_ibus_load();
EUID_ROOT();
}
//****************************
// fs pre-processing:
// - build seccomp filters
// - create an empty /etc/ld.so.preload
//****************************
if (cfg.protocol) {
if (arg_debug)
printf("Build protocol filter: %s\n", cfg.protocol);
// build the seccomp filter as a regular user
int rv = sbox_run(SBOX_USER | SBOX_CAPS_NONE | SBOX_SECCOMP, 5,
PATH_FSECCOMP, "protocol", "build", cfg.protocol, RUN_SECCOMP_PROTOCOL);
if (rv)
exit(rv);
}
#ifdef HAVE_FORCE_NONEWPRIVS
bool always_enforce_filters = true;
#else
bool always_enforce_filters = false;
#endif
// for --appimage, --chroot and --overlay* we force NO_NEW_PRIVS
// and drop all capabilities
if (getuid() != 0 && (arg_appimage || cfg.chrootdir || arg_overlay || always_enforce_filters))
enforce_filters();
// need ld.so.preload if tracing or seccomp with any non-default lists
bool need_preload = arg_trace || arg_tracelog || arg_seccomp_postexec;
// trace pre-install
if (need_preload)
fs_trace_preload();
// store hosts file
if (cfg.hosts_file)
fs_store_hosts_file();
//****************************
// configure filesystem
//****************************
#ifdef HAVE_CHROOT
if (cfg.chrootdir) {
fs_chroot(cfg.chrootdir);
//****************************
// trace pre-install, this time inside chroot
//****************************
if (need_preload)
fs_trace_preload();
}
else
#endif
#ifdef HAVE_OVERLAYFS
if (arg_overlay)
fs_overlayfs();
else
#endif
fs_basic_fs();
//****************************
// appimage
//****************************
appimage_mount();
//****************************
// private mode
//****************************
if (arg_private) {
EUID_USER();
if (cfg.home_private) { // --private=
if (cfg.chrootdir)
fwarning("private=directory feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private=directory feature is disabled in overlay\n");
else
fs_private_homedir();
}
else if (cfg.home_private_keep) { // --private-home=
if (cfg.chrootdir)
fwarning("private-home= feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-home= feature is disabled in overlay\n");
else
fs_private_home_list();
}
else // --private
fs_private();
EUID_ROOT();
}
if (arg_private_dev)
fs_private_dev();
if (arg_private_opt) {
if (cfg.chrootdir)
fwarning("private-opt feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-opt feature is disabled in overlay\n");
else {
fs_private_dir_list("/opt", RUN_OPT_DIR, cfg.opt_private_keep);
}
}
if (arg_private_srv) {
if (cfg.chrootdir)
fwarning("private-srv feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-srv feature is disabled in overlay\n");
else {
fs_private_dir_list("/srv", RUN_SRV_DIR, cfg.srv_private_keep);
}
}
// private-bin is disabled for appimages
if (arg_private_bin && !arg_appimage) {
if (cfg.chrootdir)
fwarning("private-bin feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-bin feature is disabled in overlay\n");
else {
// for --x11=xorg we need to add xauth command
if (arg_x11_xorg) {
EUID_USER();
char *tmp;
if (asprintf(&tmp, "%s,xauth", cfg.bin_private_keep) == -1)
errExit("asprintf");
cfg.bin_private_keep = tmp;
EUID_ROOT();
}
fs_private_bin_list();
}
}
// private-lib is disabled for appimages
if (arg_private_lib && !arg_appimage) {
if (cfg.chrootdir)
fwarning("private-lib feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-lib feature is disabled in overlay\n");
else {
fs_private_lib();
}
}
#ifdef HAVE_USERTMPFS
if (arg_private_cache) {
EUID_USER();
profile_add("tmpfs ${HOME}/.cache");
EUID_ROOT();
}
#endif
if (arg_private_tmp) {
// private-tmp is implemented as a whitelist
EUID_USER();
fs_private_tmp();
EUID_ROOT();
}
//****************************
// Session D-BUS
//****************************
#ifdef HAVE_DBUSPROXY
dbus_apply_policy();
#endif
//****************************
// hosts and hostname
//****************************
if (cfg.hostname)
fs_hostname(cfg.hostname);
if (cfg.hosts_file)
fs_mount_hosts_file();
//****************************
// /etc overrides from the network namespace
//****************************
if (arg_netns)
netns_mounts(arg_netns);
//****************************
// update /proc, /sys, /dev, /boot directory
//****************************
fs_proc_sys_dev_boot();
//****************************
// handle /mnt and /media
//****************************
if (checkcfg(CFG_DISABLE_MNT))
fs_mnt(1);
else if (arg_disable_mnt)
fs_mnt(0);
// Install new /etc last, so we can use it as long as possible
if (arg_private_etc) {
if (cfg.chrootdir)
fwarning("private-etc feature is disabled in chroot\n");
else if (arg_overlay)
fwarning("private-etc feature is disabled in overlay\n");
else {
/* Current /etc/passwd and /etc/group files are bind
* mounted filtered versions of originals. Leaving
* them underneath private-etc mount causes problems
* in devices with older kernels, e.g. attempts to
* update the real /etc/passwd file yield EBUSY.
*
* As we do want to retain filtered /etc content:
* 1. duplicate /etc content to RUN_ETC_DIR
* 2. unmount bind mounts from /etc
* 3. mount RUN_ETC_DIR at /etc
*/
timetrace_start();
fs_private_dir_copy("/etc", RUN_ETC_DIR, cfg.etc_private_keep);
if (umount2("/etc/group", MNT_DETACH) == -1)
fprintf(stderr, "/etc/group: unmount: %s\n", strerror(errno));
if (umount2("/etc/passwd", MNT_DETACH) == -1)
fprintf(stderr, "/etc/passwd: unmount: %s\n", strerror(errno));
fs_private_dir_mount("/etc", RUN_ETC_DIR);
fmessage("Private /etc installed in %0.2f ms\n", timetrace_end());
// create /etc/ld.so.preload file again
if (need_preload)
fs_trace_preload();
// openSUSE configuration is split between /etc and /usr/etc
// process private-etc a second time
fs_private_dir_list("/usr/etc", RUN_USR_ETC_DIR, cfg.etc_private_keep);
}
}
//****************************
// apply the profile file
//****************************
// apply all whitelist commands ...
fs_whitelist();
// ... followed by blacklist commands
fs_blacklist(); // mkdir and mkfile are processed all over again
//****************************
// nosound/no3d/notv/novideo and fix for pulseaudio 7.0
//****************************
if (arg_nosound) {
// disable pulseaudio
pulseaudio_disable();
// disable /dev/snd
fs_dev_disable_sound();
}
else if (!arg_keep_config_pulse)
pulseaudio_init();
if (arg_no3d)
fs_dev_disable_3d();
if (arg_notv)
fs_dev_disable_tv();
if (arg_nodvd)
fs_dev_disable_dvd();
if (arg_nou2f)
fs_dev_disable_u2f();
if (arg_novideo)
fs_dev_disable_video();
if (arg_noinput)
fs_dev_disable_input();
//****************************
// set dns
//****************************
fs_resolvconf();
//****************************
// start dhcp client
//****************************
dhcp_start();
//****************************
// set application environment
//****************************
EUID_USER();
int cwd = 0;
if (cfg.cwd) {
if (chdir(cfg.cwd) == 0)
cwd = 1;
else if (arg_private_cwd) {
fprintf(stderr, "Error: unable to enter private working directory: %s: %s\n", cfg.cwd, strerror(errno));
exit(1);
}
}
if (!cwd) {
if (chdir("/") < 0)
errExit("chdir");
if (cfg.homedir) {
struct stat s;
if (stat(cfg.homedir, &s) == 0) {
/* coverity[toctou] */
if (chdir(cfg.homedir) < 0)
errExit("chdir");
}
}
}
if (arg_debug) {
char *cpath = get_current_dir_name();
if (cpath) {
printf("Current directory: %s\n", cpath);
free(cpath);
}
}
EUID_ROOT();
// clean /tmp/.X11-unix sockets
fs_x11();
if (arg_x11_xorg)
x11_xorg();
// save original umask
save_umask();
//****************************
// fs post-processing
//****************************
fs_logger_print();
fs_logger_change_owner();
//****************************
// set security filters
//****************************
// save state of nonewprivs
save_nonewprivs();
// save cpu affinity mask to CPU_CFG file
save_cpu();
// save cgroup in CGROUP_CFG file
save_cgroup();
// set seccomp
// install protocol filter
#ifdef SYS_socket
if (cfg.protocol) {
if (arg_debug)
printf("Install protocol filter: %s\n", cfg.protocol);
seccomp_load(RUN_SECCOMP_PROTOCOL); // install filter
protocol_filter_save(); // save filter in RUN_PROTOCOL_CFG
}
else {
int rv = unlink(RUN_SECCOMP_PROTOCOL);
(void) rv;
}
#endif
// if a keep list is available, disregard the drop list
if (arg_seccomp == 1) {
if (cfg.seccomp_list_keep)
seccomp_filter_keep(true);
else
seccomp_filter_drop(true);
}
if (arg_seccomp32 == 1) {
if (cfg.seccomp_list_keep32)
seccomp_filter_keep(false);
else
seccomp_filter_drop(false);
}
if (arg_memory_deny_write_execute) {
if (arg_seccomp_error_action != EPERM) {
seccomp_filter_mdwx(true);
seccomp_filter_mdwx(false);
}
if (arg_debug)
printf("Install memory write&execute filter\n");
seccomp_load(RUN_SECCOMP_MDWX); // install filter
seccomp_load(RUN_SECCOMP_MDWX_32);
}
// make seccomp filters read-only
fs_remount(RUN_SECCOMP_DIR, MOUNT_READONLY, 0);
seccomp_debug();
//****************************
// install trace - still need capabilities
//****************************
if (need_preload)
fs_trace();
//****************************
// continue security filters
//****************************
// set capabilities
set_caps();
//****************************************
// relay status information to join option
//****************************************
char *set_sandbox_status = create_join_file();
//****************************************
// create a new user namespace
// - too early to drop privileges
//****************************************
save_nogroups();
if (arg_noroot) {
int rv = unshare(CLONE_NEWUSER);
if (rv == -1) {
fwarning("cannot create a new user namespace, going forward without it...\n");
arg_noroot = 0;
}
}
// notify parent that new user namespace has been created so a proper
// UID/GID map can be setup
notify_other(child_to_parent_fds[1]);
close(child_to_parent_fds[1]);
// wait for parent to finish setting up a proper UID/GID map
wait_for_other(parent_to_child_fds[0]);
close(parent_to_child_fds[0]);
// somehow, the new user namespace resets capabilities;
// we need to do them again
if (arg_noroot) {
if (arg_debug)
printf("noroot user namespace installed\n");
set_caps();
}
//****************************************
// Set NO_NEW_PRIVS if desired
//****************************************
if (arg_nonewprivs) {
prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
if (prctl(PR_GET_NO_NEW_PRIVS, 0, 0, 0, 0) != 1) {
fwarning("cannot set NO_NEW_PRIVS, it requires a Linux kernel version 3.5 or newer.\n");
if (force_nonewprivs) {
fprintf(stderr, "Error: NO_NEW_PRIVS required for this sandbox, exiting ...\n");
exit(1);
}
}
else if (arg_debug)
printf("NO_NEW_PRIVS set\n");
}
//****************************************
// drop privileges
//****************************************
drop_privs(arg_nogroups);
// kill the sandbox in case the parent died
prctl(PR_SET_PDEATHSIG, SIGKILL, 0, 0, 0);
//****************************************
// set cpu affinity
//****************************************
if (cfg.cpus)
set_cpu_affinity();
//****************************************
// fork the application and monitor it
//****************************************
pid_t app_pid = fork();
if (app_pid == -1)
errExit("fork");
if (app_pid == 0) {
#ifdef HAVE_APPARMOR
// add apparmor confinement after the execve
set_apparmor();
#endif
// set nice and rlimits
if (arg_nice)
set_nice(cfg.nice);
set_rlimits();
start_application(0, -1, set_sandbox_status);
}
munmap(set_sandbox_status, 1);
int status = monitor_application(app_pid); // monitor application
flush_stdin();
if (WIFEXITED(status)) {
// if we had a proper exit, return that exit status
return WEXITSTATUS(status);
} else {
// something else went wrong!
return -1;
}
}
|