-
Notifications
You must be signed in to change notification settings - Fork 1.1k
/
start.c
1162 lines (974 loc) · 27.8 KB
/
start.c
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
/*
* lxc: linux Container library
*
* (C) Copyright IBM Corp. 2007, 2008
*
* Authors:
* Daniel Lezcano <daniel.lezcano at free.fr>
*
* This library 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.
*
* This library 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 this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "config.h"
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <dirent.h>
#include <errno.h>
#include <unistd.h>
#include <signal.h>
#include <fcntl.h>
#include <grp.h>
#include <poll.h>
#include <sys/param.h>
#include <sys/file.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/prctl.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <sys/un.h>
#include <sys/syscall.h>
#if HAVE_SYS_CAPABILITY_H
#include <sys/capability.h>
#endif
#if !HAVE_DECL_PR_CAPBSET_DROP
#define PR_CAPBSET_DROP 24
#endif
#include "start.h"
#include "conf.h"
#include "log.h"
#include "cgroup.h"
#include "error.h"
#include "af_unix.h"
#include "mainloop.h"
#include "utils.h"
#include "lxcutmp.h"
#include "monitor.h"
#include "commands.h"
#include "console.h"
#include "sync.h"
#include "namespace.h"
#include "lxcseccomp.h"
#include "caps.h"
#include "lsm/lsm.h"
lxc_log_define(lxc_start, lxc);
const struct ns_info ns_info[LXC_NS_MAX] = {
[LXC_NS_MNT] = {"mnt", CLONE_NEWNS},
[LXC_NS_PID] = {"pid", CLONE_NEWPID},
[LXC_NS_UTS] = {"uts", CLONE_NEWUTS},
[LXC_NS_IPC] = {"ipc", CLONE_NEWIPC},
[LXC_NS_USER] = {"user", CLONE_NEWUSER},
[LXC_NS_NET] = {"net", CLONE_NEWNET}
};
static void print_top_failing_dir(const char *path)
{
size_t len = strlen(path);
char *copy = alloca(len+1), *p, *e, saved;
strcpy(copy, path);
p = copy;
e = copy + len;
while (p < e) {
while (p < e && *p == '/') p++;
while (p < e && *p != '/') p++;
saved = *p;
*p = '\0';
if (access(copy, X_OK)) {
SYSERROR("could not access %s. Please grant it 'x' " \
"access, or add an ACL for the container root.",
copy);
return;
}
*p = saved;
}
}
static void close_ns(int ns_fd[LXC_NS_MAX]) {
int i;
for (i = 0; i < LXC_NS_MAX; i++) {
if (ns_fd[i] > -1) {
close(ns_fd[i]);
ns_fd[i] = -1;
}
}
}
static int preserve_ns(int ns_fd[LXC_NS_MAX], int clone_flags) {
int i, saved_errno;
char path[MAXPATHLEN];
for (i = 0; i < LXC_NS_MAX; i++)
ns_fd[i] = -1;
if (access("/proc/self/ns", X_OK)) {
WARN("Kernel does not support attach; preserve_ns ignored");
return 0;
}
for (i = 0; i < LXC_NS_MAX; i++) {
if ((clone_flags & ns_info[i].clone_flag) == 0)
continue;
snprintf(path, MAXPATHLEN, "/proc/self/ns/%s", ns_info[i].proc_name);
ns_fd[i] = open(path, O_RDONLY | O_CLOEXEC);
if (ns_fd[i] < 0)
goto error;
}
return 0;
error:
saved_errno = errno;
close_ns(ns_fd);
errno = saved_errno;
SYSERROR("failed to open '%s'", path);
return -1;
}
static int attach_ns(const int ns_fd[LXC_NS_MAX]) {
int i;
for (i = 0; i < LXC_NS_MAX; i++) {
if (ns_fd[i] < 0)
continue;
if (setns(ns_fd[i], 0) != 0)
goto error;
}
return 0;
error:
SYSERROR("failed to set namespace '%s'", ns_info[i].proc_name);
return -1;
}
static int match_fd(int fd)
{
return (fd == 0 || fd == 1 || fd == 2);
}
int lxc_check_inherited(struct lxc_conf *conf, int fd_to_ignore)
{
struct dirent dirent, *direntp;
int fd, fddir;
DIR *dir;
restart:
dir = opendir("/proc/self/fd");
if (!dir) {
WARN("failed to open directory: %m");
return -1;
}
fddir = dirfd(dir);
while (!readdir_r(dir, &dirent, &direntp)) {
if (!direntp)
break;
if (!strcmp(direntp->d_name, "."))
continue;
if (!strcmp(direntp->d_name, ".."))
continue;
fd = atoi(direntp->d_name);
if (fd == fddir || fd == lxc_log_fd || fd == fd_to_ignore)
continue;
if (match_fd(fd))
continue;
if (conf->close_all_fds) {
close(fd);
closedir(dir);
INFO("closed inherited fd %d", fd);
goto restart;
}
WARN("inherited fd %d", fd);
}
closedir(dir); /* cannot fail */
return 0;
}
static int setup_signal_fd(sigset_t *oldmask)
{
sigset_t mask;
int fd;
/* Block everything except serious error signals */
if (sigfillset(&mask) ||
sigdelset(&mask, SIGILL) ||
sigdelset(&mask, SIGSEGV) ||
sigdelset(&mask, SIGBUS) ||
sigdelset(&mask, SIGWINCH) ||
sigprocmask(SIG_BLOCK, &mask, oldmask)) {
SYSERROR("failed to set signal mask");
return -1;
}
fd = signalfd(-1, &mask, 0);
if (fd < 0) {
SYSERROR("failed to create the signal fd");
return -1;
}
if (fcntl(fd, F_SETFD, FD_CLOEXEC)) {
SYSERROR("failed to set sigfd to close-on-exec");
close(fd);
return -1;
}
DEBUG("sigchild handler set");
return fd;
}
static int signal_handler(int fd, uint32_t events, void *data,
struct lxc_epoll_descr *descr)
{
struct signalfd_siginfo siginfo;
siginfo_t info;
int ret;
pid_t *pid = data;
bool init_died = false;
ret = read(fd, &siginfo, sizeof(siginfo));
if (ret < 0) {
ERROR("failed to read signal info");
return -1;
}
if (ret != sizeof(siginfo)) {
ERROR("unexpected siginfo size");
return -1;
}
// check whether init is running
info.si_pid = 0;
ret = waitid(P_PID, *pid, &info, WEXITED | WNOWAIT | WNOHANG);
if (ret == 0 && info.si_pid == *pid) {
init_died = true;
}
if (siginfo.ssi_signo != SIGCHLD) {
kill(*pid, siginfo.ssi_signo);
INFO("forwarded signal %d to pid %d", siginfo.ssi_signo, *pid);
return init_died ? 1 : 0;
}
if (siginfo.ssi_code == CLD_STOPPED ||
siginfo.ssi_code == CLD_CONTINUED) {
INFO("container init process was stopped/continued");
return init_died ? 1 : 0;
}
/* more robustness, protect ourself from a SIGCHLD sent
* by a process different from the container init
*/
if (siginfo.ssi_pid != *pid) {
WARN("invalid pid for SIGCHLD");
return init_died ? 1 : 0;
}
DEBUG("container init process exited");
return 1;
}
static int lxc_set_state(const char *name, struct lxc_handler *handler, lxc_state_t state)
{
handler->state = state;
lxc_monitor_send_state(name, state, handler->lxcpath);
return 0;
}
static int lxc_poll(const char *name, struct lxc_handler *handler)
{
int sigfd = handler->sigfd;
int pid = handler->pid;
struct lxc_epoll_descr descr;
if (lxc_mainloop_open(&descr)) {
ERROR("failed to create mainloop");
goto out_sigfd;
}
if (lxc_mainloop_add_handler(&descr, sigfd, signal_handler, &pid)) {
ERROR("failed to add handler for the signal");
goto out_mainloop_open;
}
if (lxc_console_mainloop_add(&descr, handler)) {
ERROR("failed to add console handler to mainloop");
goto out_mainloop_open;
}
if (lxc_cmd_mainloop_add(name, &descr, handler)) {
ERROR("failed to add command handler to mainloop");
goto out_mainloop_open;
}
if (handler->conf->need_utmp_watch) {
#if HAVE_SYS_CAPABILITY_H
if (lxc_utmp_mainloop_add(&descr, handler)) {
ERROR("failed to add utmp handler to mainloop");
goto out_mainloop_open;
}
#else
DEBUG("not starting utmp handler as cap_sys_boot cannot be dropped without capabilities support");
#endif
}
return lxc_mainloop(&descr, -1);
out_mainloop_open:
lxc_mainloop_close(&descr);
out_sigfd:
close(sigfd);
return -1;
}
struct lxc_handler *lxc_init(const char *name, struct lxc_conf *conf, const char *lxcpath)
{
struct lxc_handler *handler;
handler = malloc(sizeof(*handler));
if (!handler)
return NULL;
memset(handler, 0, sizeof(*handler));
handler->conf = conf;
handler->lxcpath = lxcpath;
handler->pinfd = -1;
lsm_init();
handler->name = strdup(name);
if (!handler->name) {
ERROR("failed to allocate memory");
goto out_free;
}
if (lxc_cmd_init(name, handler, lxcpath))
goto out_free_name;
if (lxc_read_seccomp_config(conf) != 0) {
ERROR("failed loading seccomp policy");
goto out_close_maincmd_fd;
}
/* Begin by setting the state to STARTING */
if (lxc_set_state(name, handler, STARTING)) {
ERROR("failed to set state '%s'", lxc_state2str(STARTING));
goto out_close_maincmd_fd;
}
/* Start of environment variable setup for hooks */
if (setenv("LXC_NAME", name, 1)) {
SYSERROR("failed to set environment variable for container name");
}
if (setenv("LXC_CONFIG_FILE", conf->rcfile, 1)) {
SYSERROR("failed to set environment variable for config path");
}
if (setenv("LXC_ROOTFS_MOUNT", conf->rootfs.mount, 1)) {
SYSERROR("failed to set environment variable for rootfs mount");
}
if (setenv("LXC_ROOTFS_PATH", conf->rootfs.path, 1)) {
SYSERROR("failed to set environment variable for rootfs mount");
}
if (conf->console.path && setenv("LXC_CONSOLE", conf->console.path, 1)) {
SYSERROR("failed to set environment variable for console path");
}
if (conf->console.log_path && setenv("LXC_CONSOLE_LOGPATH", conf->console.log_path, 1)) {
SYSERROR("failed to set environment variable for console log");
}
/* End of environment variable setup for hooks */
if (run_lxc_hooks(name, "pre-start", conf, handler->lxcpath, NULL)) {
ERROR("failed to run pre-start hooks for container '%s'.", name);
goto out_aborting;
}
if (lxc_create_tty(name, conf)) {
ERROR("failed to create the ttys");
goto out_aborting;
}
/* the signal fd has to be created before forking otherwise
* if the child process exits before we setup the signal fd,
* the event will be lost and the command will be stuck */
handler->sigfd = setup_signal_fd(&handler->oldmask);
if (handler->sigfd < 0) {
ERROR("failed to set sigchild fd handler");
goto out_delete_tty;
}
/* do this after setting up signals since it might unblock SIGWINCH */
if (lxc_console_create(conf)) {
ERROR("failed to create console");
goto out_restore_sigmask;
}
if (ttys_shift_ids(conf) < 0) {
ERROR("Failed to shift tty into container");
goto out_restore_sigmask;
}
INFO("'%s' is initialized", name);
return handler;
out_restore_sigmask:
sigprocmask(SIG_SETMASK, &handler->oldmask, NULL);
out_delete_tty:
lxc_delete_tty(&conf->tty_info);
out_aborting:
lxc_set_state(name, handler, ABORTING);
out_close_maincmd_fd:
close(conf->maincmd_fd);
conf->maincmd_fd = -1;
out_free_name:
free(handler->name);
handler->name = NULL;
out_free:
free(handler);
return NULL;
}
static void lxc_fini(const char *name, struct lxc_handler *handler)
{
/* The STOPPING state is there for future cleanup code
* which can take awhile
*/
lxc_set_state(name, handler, STOPPING);
lxc_set_state(name, handler, STOPPED);
if (run_lxc_hooks(name, "post-stop", handler->conf, handler->lxcpath, NULL))
ERROR("failed to run post-stop hooks for container '%s'.", name);
/* reset mask set by setup_signal_fd */
if (sigprocmask(SIG_SETMASK, &handler->oldmask, NULL))
WARN("failed to restore sigprocmask");
lxc_console_delete(&handler->conf->console);
lxc_delete_tty(&handler->conf->tty_info);
close(handler->conf->maincmd_fd);
handler->conf->maincmd_fd = -1;
free(handler->name);
cgroup_destroy(handler);
free(handler);
}
static void lxc_abort(const char *name, struct lxc_handler *handler)
{
int ret, status;
lxc_set_state(name, handler, ABORTING);
if (handler->pid > 0)
kill(handler->pid, SIGKILL);
while ((ret = waitpid(-1, &status, 0)) > 0) ;
}
#include <sys/reboot.h>
#include <linux/reboot.h>
/*
* reboot(LINUX_REBOOT_CMD_CAD_ON) will return -EINVAL
* in a child pid namespace if container reboot support exists.
* Otherwise, it will either succeed or return -EPERM.
*/
static int container_reboot_supported(void *arg)
{
int *cmd = arg;
int ret;
ret = reboot(*cmd);
if (ret == -1 && errno == EINVAL)
return 1;
return 0;
}
static int must_drop_cap_sys_boot(struct lxc_conf *conf)
{
FILE *f;
int ret, cmd, v, flags;
long stack_size = 4096;
void *stack = alloca(stack_size);
int status;
pid_t pid;
f = fopen("/proc/sys/kernel/ctrl-alt-del", "r");
if (!f) {
DEBUG("failed to open /proc/sys/kernel/ctrl-alt-del");
return 1;
}
ret = fscanf(f, "%d", &v);
fclose(f);
if (ret != 1) {
DEBUG("Failed to read /proc/sys/kernel/ctrl-alt-del");
return 1;
}
cmd = v ? LINUX_REBOOT_CMD_CAD_ON : LINUX_REBOOT_CMD_CAD_OFF;
flags = CLONE_NEWPID | SIGCHLD;
if (!lxc_list_empty(&conf->id_map))
flags |= CLONE_NEWUSER;
#ifdef __ia64__
pid = __clone2(container_reboot_supported, stack, stack_size, flags, &cmd);
#else
stack += stack_size;
pid = clone(container_reboot_supported, stack, flags, &cmd);
#endif
if (pid < 0) {
SYSERROR("failed to clone");
return -1;
}
if (wait(&status) < 0) {
SYSERROR("unexpected wait error: %m");
return -1;
}
if (WEXITSTATUS(status) != 1)
return 1;
return 0;
}
/*
* netpipe is used in the unprivileged case to transfer the ifindexes
* from parent to child
*/
static int netpipe = -1;
static inline int count_veths(struct lxc_list *network)
{
struct lxc_list *iterator;
struct lxc_netdev *netdev;
int count = 0;
lxc_list_for_each(iterator, network) {
netdev = iterator->elem;
if (netdev->type != LXC_NET_VETH)
continue;
count++;
}
return count;
}
static int read_unpriv_netifindex(struct lxc_list *network)
{
struct lxc_list *iterator;
struct lxc_netdev *netdev;
if (netpipe == -1)
return 0;
lxc_list_for_each(iterator, network) {
netdev = iterator->elem;
if (netdev->type != LXC_NET_VETH)
continue;
if (!(netdev->name = malloc(IFNAMSIZ))) {
ERROR("Out of memory");
close(netpipe);
return -1;
}
if (read(netpipe, netdev->name, IFNAMSIZ) != IFNAMSIZ) {
close(netpipe);
return -1;
}
}
close(netpipe);
return 0;
}
static int do_start(void *data)
{
struct lxc_handler *handler = data;
const char *lsm_label = NULL;
if (sigprocmask(SIG_SETMASK, &handler->oldmask, NULL)) {
SYSERROR("failed to set sigprocmask");
return -1;
}
/* This prctl must be before the synchro, so if the parent
* dies before we set the parent death signal, we will detect
* its death with the synchro right after, otherwise we have
* a window where the parent can exit before we set the pdeath
* signal leading to a unsupervized container.
*/
if (prctl(PR_SET_PDEATHSIG, SIGKILL, 0, 0, 0)) {
SYSERROR("failed to set pdeath signal");
return -1;
}
lxc_sync_fini_parent(handler);
/* don't leak the pinfd to the container */
if (handler->pinfd >= 0) {
close(handler->pinfd);
}
/* Tell the parent task it can begin to configure the
* container and wait for it to finish
*/
if (lxc_sync_barrier_parent(handler, LXC_SYNC_CONFIGURE))
return -1;
if (read_unpriv_netifindex(&handler->conf->network) < 0)
goto out_warn_father;
/*
* if we are in a new user namespace, become root there to have
* privilege over our namespace
*/
if (!lxc_list_empty(&handler->conf->id_map)) {
NOTICE("switching to gid/uid 0 in new user namespace");
if (setgid(0)) {
SYSERROR("setgid");
goto out_warn_father;
}
if (setuid(0)) {
SYSERROR("setuid");
goto out_warn_father;
}
if (setgroups(0, NULL)) {
SYSERROR("setgroups");
goto out_warn_father;
}
}
if (access(handler->lxcpath, X_OK)) {
print_top_failing_dir(handler->lxcpath);
goto out_warn_father;
}
#if HAVE_SYS_CAPABILITY_H
if (handler->conf->need_utmp_watch) {
if (prctl(PR_CAPBSET_DROP, CAP_SYS_BOOT, 0, 0, 0)) {
SYSERROR("failed to remove CAP_SYS_BOOT capability");
goto out_warn_father;
}
DEBUG("Dropped cap_sys_boot");
}
#endif
/* Setup the container, ip, names, utsname, ... */
if (lxc_setup(handler)) {
ERROR("failed to setup the container");
goto out_warn_father;
}
/* ask father to setup cgroups and wait for him to finish */
if (lxc_sync_barrier_parent(handler, LXC_SYNC_CGROUP))
return -1;
/* Set the label to change to when we exec(2) the container's init */
if (!strcmp(lsm_name(), "AppArmor"))
lsm_label = handler->conf->lsm_aa_profile;
else if (!strcmp(lsm_name(), "SELinux"))
lsm_label = handler->conf->lsm_se_context;
if (lsm_process_label_set(lsm_label, 1, 1) < 0)
goto out_warn_father;
/* Some init's such as busybox will set sane tty settings on stdin,
* stdout, stderr which it thinks is the console. We already set them
* the way we wanted on the real terminal, and we want init to do its
* setup on its console ie. the pty allocated in lxc_console_create()
* so make sure that that pty is stdin,stdout,stderr.
*/
if (lxc_console_set_stdfds(handler) < 0)
goto out_warn_father;
/* If we mounted a temporary proc, then unmount it now */
tmp_proc_unmount(handler->conf);
if (lxc_seccomp_load(handler->conf) != 0)
goto out_warn_father;
if (run_lxc_hooks(handler->name, "start", handler->conf, handler->lxcpath, NULL)) {
ERROR("failed to run start hooks for container '%s'.", handler->name);
goto out_warn_father;
}
/* The clearenv() and putenv() calls have been moved here
* to allow us to use enviroment variables passed to the various
* hooks, such as the start hook above. Not all of the
* variables like CONFIG_PATH or ROOTFS are valid in this
* context but others are. */
if (clearenv()) {
SYSERROR("failed to clear environment");
/* don't error out though */
}
if (putenv("container=lxc")) {
SYSERROR("failed to set environment variable");
goto out_warn_father;
}
close(handler->sigfd);
/* after this call, we are in error because this
* ops should not return as it execs */
handler->ops->start(handler, handler->data);
out_warn_father:
/* we want the parent to know something went wrong, so any
* value other than what it expects is ok. */
lxc_sync_wake_parent(handler, LXC_SYNC_POST_CONFIGURE);
return -1;
}
static int save_phys_nics(struct lxc_conf *conf)
{
struct lxc_list *iterator;
lxc_list_for_each(iterator, &conf->network) {
struct lxc_netdev *netdev = iterator->elem;
if (netdev->type != LXC_NET_PHYS)
continue;
conf->saved_nics = realloc(conf->saved_nics,
(conf->num_savednics+1)*sizeof(struct saved_nic));
if (!conf->saved_nics) {
SYSERROR("failed to allocate memory");
return -1;
}
conf->saved_nics[conf->num_savednics].ifindex = netdev->ifindex;
conf->saved_nics[conf->num_savednics].orig_name = strdup(netdev->link);
if (!conf->saved_nics[conf->num_savednics].orig_name) {
SYSERROR("failed to allocate memory");
return -1;
}
INFO("stored saved_nic #%d idx %d name %s", conf->num_savednics,
conf->saved_nics[conf->num_savednics].ifindex,
conf->saved_nics[conf->num_savednics].orig_name);
conf->num_savednics++;
}
return 0;
}
static int lxc_spawn(struct lxc_handler *handler)
{
int failed_before_rename = 0;
const char *name = handler->name;
bool cgroups_connected = false;
int saved_ns_fd[LXC_NS_MAX];
int preserve_mask = 0, i;
int netpipepair[2], nveths;
netpipe = -1;
for (i = 0; i < LXC_NS_MAX; i++)
if (handler->conf->inherit_ns_fd[i] != -1)
preserve_mask |= ns_info[i].clone_flag;
if (lxc_sync_init(handler))
return -1;
handler->clone_flags = CLONE_NEWPID|CLONE_NEWNS;
if (!lxc_list_empty(&handler->conf->id_map)) {
INFO("Cloning a new user namespace");
handler->clone_flags |= CLONE_NEWUSER;
}
if (handler->conf->inherit_ns_fd[LXC_NS_NET] == -1) {
if (!lxc_requests_empty_network(handler))
handler->clone_flags |= CLONE_NEWNET;
if (!lxc_list_empty(&handler->conf->network)) {
/* Find gateway addresses from the link device, which is
* no longer accessible inside the container. Do this
* before creating network interfaces, since goto
* out_delete_net does not work before lxc_clone. */
if (lxc_find_gateway_addresses(handler)) {
ERROR("failed to find gateway addresses");
lxc_sync_fini(handler);
return -1;
}
/* that should be done before the clone because we will
* fill the netdev index and use them in the child
*/
if (lxc_create_network(handler)) {
ERROR("failed to create the network");
lxc_sync_fini(handler);
return -1;
}
}
if (save_phys_nics(handler->conf)) {
ERROR("failed to save physical nic info");
goto out_abort;
}
} else {
INFO("Inheriting a net namespace");
}
if (handler->conf->inherit_ns_fd[LXC_NS_IPC] == -1) {
handler->clone_flags |= CLONE_NEWIPC;
} else {
INFO("Inheriting an IPC namespace");
}
if (handler->conf->inherit_ns_fd[LXC_NS_UTS] == -1) {
handler->clone_flags |= CLONE_NEWUTS;
} else {
INFO("Inheriting a UTS namespace");
}
if (!cgroup_init(handler)) {
ERROR("failed initializing cgroup support");
goto out_delete_net;
}
cgroups_connected = true;
if (!cgroup_create(handler)) {
ERROR("failed creating cgroups");
goto out_delete_net;
}
/*
* if the rootfs is not a blockdev, prevent the container from
* marking it readonly.
*
* if the container is unprivileged then skip rootfs pinning
*/
if (lxc_list_empty(&handler->conf->id_map)) {
handler->pinfd = pin_rootfs(handler->conf->rootfs.path);
if (handler->pinfd == -1)
INFO("failed to pin the container's rootfs");
}
if (preserve_ns(saved_ns_fd, preserve_mask) < 0)
goto out_delete_net;
if (attach_ns(handler->conf->inherit_ns_fd) < 0)
goto out_delete_net;
if (am_unpriv() && (nveths = count_veths(&handler->conf->network))) {
if (pipe(netpipepair) < 0) {
SYSERROR("Error creating pipe");
goto out_delete_net;
}
/* store netpipe in the global var for do_start's use */
netpipe = netpipepair[0];
}
/* Create a process in a new set of namespaces */
handler->pid = lxc_clone(do_start, handler, handler->clone_flags);
if (handler->pid < 0) {
SYSERROR("failed to fork into a new namespace");
goto out_delete_net;
}
if (attach_ns(saved_ns_fd))
WARN("failed to restore saved namespaces");
lxc_sync_fini_child(handler);
if (lxc_sync_wait_child(handler, LXC_SYNC_CONFIGURE))
failed_before_rename = 1;
if (!cgroup_create_legacy(handler)) {
ERROR("failed to setup the legacy cgroups for %s", name);
goto out_delete_net;
}
if (!cgroup_setup_limits(handler, false)) {
ERROR("failed to setup the cgroup limits for '%s'", name);
goto out_delete_net;
}
if (!cgroup_enter(handler))
goto out_delete_net;
if (!cgroup_chown(handler))
goto out_delete_net;
if (failed_before_rename)
goto out_delete_net;
/* Create the network configuration */
if (handler->clone_flags & CLONE_NEWNET) {
if (lxc_assign_network(&handler->conf->network, handler->pid)) {
ERROR("failed to create the configured network");
goto out_delete_net;
}
}
if (netpipe != -1) {
struct lxc_list *iterator;
struct lxc_netdev *netdev;
close(netpipe);
lxc_list_for_each(iterator, &handler->conf->network) {
netdev = iterator->elem;
if (netdev->type != LXC_NET_VETH)
continue;
if (write(netpipepair[1], netdev->name, IFNAMSIZ) != IFNAMSIZ) {
ERROR("Error writing veth name to container");
goto out_delete_net;
}
}
close(netpipepair[1]);
}
/* map the container uids - the container became an invalid
* userid the moment it was cloned with CLONE_NEWUSER - this
* call doesn't change anything immediately, but allows the
* container to setuid(0) (0 being mapped to something else on
* the host) later to become a valid uid again */
if (lxc_map_ids(&handler->conf->id_map, handler->pid)) {
ERROR("failed to set up id mapping");
goto out_delete_net;
}
/* Tell the child to continue its initialization. we'll get
* LXC_SYNC_CGROUP when it is ready for us to setup cgroups
*/
if (lxc_sync_barrier_child(handler, LXC_SYNC_POST_CONFIGURE))
goto out_delete_net;
if (!cgroup_setup_limits(handler, true)) {
ERROR("failed to setup the devices cgroup for '%s'", name);
goto out_delete_net;
}
cgroup_disconnect();
cgroups_connected = false;
/* Tell the child to complete its initialization and wait for
* it to exec or return an error. (the child will never
* return LXC_SYNC_POST_CGROUP+1. It will either close the
* sync pipe, causing lxc_sync_barrier_child to return
* success, or return a different value, causing us to error
* out).
*/
if (lxc_sync_barrier_child(handler, LXC_SYNC_POST_CGROUP))
return -1;
if (detect_shared_rootfs())
umount2(handler->conf->rootfs.mount, MNT_DETACH);
if (handler->ops->post_start(handler, handler->data))
goto out_abort;
if (lxc_set_state(name, handler, RUNNING)) {
ERROR("failed to set state to %s",
lxc_state2str(RUNNING));
goto out_abort;
}
lxc_sync_fini(handler);
return 0;
out_delete_net:
if (cgroups_connected)
cgroup_disconnect();
if (handler->clone_flags & CLONE_NEWNET)
lxc_delete_network(handler);
out_abort:
lxc_abort(name, handler);