forked from ioi/isolate
-
Notifications
You must be signed in to change notification settings - Fork 0
/
isolate.c
1458 lines (1271 loc) · 32.2 KB
/
isolate.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
/*
* A Process Isolator based on Linux Containers
*
* (c) 2012 Martin Mares <mj@ucw.cz>
* (c) 2012 Bernard Blackham <bernard@blackham.com.au>
*/
#define _GNU_SOURCE
#include "autoconf.h"
#include <errno.h>
#include <stdio.h>
#include <fcntl.h>
#include <stdlib.h>
#include <string.h>
#include <stdarg.h>
#include <stdint.h>
#include <unistd.h>
#include <getopt.h>
#include <sched.h>
#include <time.h>
#include <grp.h>
#include <mntent.h>
#include <limits.h>
#include <sys/wait.h>
#include <sys/time.h>
#include <sys/signal.h>
#include <sys/resource.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <sys/quota.h>
#include <sys/vfs.h>
#define NONRET __attribute__((noreturn))
#define UNUSED __attribute__((unused))
#define ARRAY_SIZE(a) (int)(sizeof(a)/sizeof(a[0]))
static int timeout; /* milliseconds */
static int wall_timeout;
static int extra_timeout;
static int pass_environ;
static int verbose;
static int memory_limit;
static int stack_limit;
static int block_quota;
static int inode_quota;
static int max_processes = 1;
static char *redir_stdin, *redir_stdout, *redir_stderr;
static char *set_cwd;
static int cg_enable;
static int cg_memory_limit;
static int cg_timing;
static int box_id;
static char box_dir[1024];
static pid_t box_pid;
static uid_t box_uid;
static gid_t box_gid;
static uid_t orig_uid;
static gid_t orig_gid;
static int partial_line;
static char cleanup_cmd[256];
static struct timeval start_time;
static int ticks_per_sec;
static int total_ms, wall_ms;
static volatile sig_atomic_t timer_tick;
static int error_pipes[2];
static int write_errors_to_fd;
static int read_errors_from_fd;
static void die(char *msg, ...) NONRET;
static void cg_stats(void);
static int get_wall_time_ms(void);
static int get_run_time_ms(struct rusage *rus);
/*** Meta-files ***/
static FILE *metafile;
static void
meta_open(const char *name)
{
if (!strcmp(name, "-"))
{
metafile = stdout;
return;
}
metafile = fopen(name, "w");
if (!metafile)
die("Failed to open metafile '%s'",name);
}
static void
meta_close(void)
{
if (metafile && metafile != stdout)
fclose(metafile);
}
static void __attribute__((format(printf,1,2)))
meta_printf(const char *fmt, ...)
{
if (!metafile)
return;
va_list args;
va_start(args, fmt);
vfprintf(metafile, fmt, args);
va_end(args);
}
static void
final_stats(struct rusage *rus)
{
total_ms = get_run_time_ms(rus);
wall_ms = get_wall_time_ms();
meta_printf("time:%d.%03d\n", total_ms/1000, total_ms%1000);
meta_printf("time-wall:%d.%03d\n", wall_ms/1000, wall_ms%1000);
meta_printf("max-rss:%ld\n", rus->ru_maxrss);
meta_printf("csw-voluntary:%ld\n", rus->ru_nvcsw);
meta_printf("csw-forced:%ld\n", rus->ru_nivcsw);
cg_stats();
}
/*** Messages and exits ***/
static void
xsystem(const char *cmd)
{
int ret = system(cmd);
if (ret < 0)
die("system(\"%s\"): %m", cmd);
if (!WIFEXITED(ret) || WEXITSTATUS(ret))
die("system(\"%s\"): Exited with status %d", cmd, ret);
}
static void NONRET
box_exit(int rc)
{
if (box_pid > 0)
{
kill(-box_pid, SIGKILL);
kill(box_pid, SIGKILL);
meta_printf("killed:1\n");
struct rusage rus;
int p, stat;
do
p = wait4(box_pid, &stat, 0, &rus);
while (p < 0 && errno == EINTR);
if (p < 0)
fprintf(stderr, "UGH: Lost track of the process (%m)\n");
else
final_stats(&rus);
}
if (rc < 2 && cleanup_cmd[0])
xsystem(cleanup_cmd);
meta_close();
exit(rc);
}
static void
flush_line(void)
{
if (partial_line)
fputc('\n', stderr);
partial_line = 0;
}
/* Report an error of the sandbox itself */
static void NONRET __attribute__((format(printf,1,2)))
die(char *msg, ...)
{
va_list args;
va_start(args, msg);
char buf[1024];
int n = vsnprintf(buf, sizeof(buf), msg, args);
if (write_errors_to_fd)
{
// We are inside the box, have to use error pipe for error reporting.
// We hope that the whole error message fits in PIPE_BUF bytes.
write(write_errors_to_fd, buf, n);
exit(2);
}
// Otherwise, we in the box keeper process, so we report errors normally
flush_line();
meta_printf("status:XX\nmessage:%s\n", buf);
fputs(buf, stderr);
fputc('\n', stderr);
box_exit(2);
}
/* Report an error of the program inside the sandbox */
static void NONRET __attribute__((format(printf,1,2)))
err(char *msg, ...)
{
va_list args;
va_start(args, msg);
flush_line();
if (msg[0] && msg[1] && msg[2] == ':' && msg[3] == ' ')
{
meta_printf("status:%c%c\n", msg[0], msg[1]);
msg += 4;
}
char buf[1024];
vsnprintf(buf, sizeof(buf), msg, args);
meta_printf("message:%s\n", buf);
fputs(buf, stderr);
fputc('\n', stderr);
box_exit(1);
}
/* Write a message, but only if in verbose mode */
static void __attribute__((format(printf,1,2)))
msg(char *msg, ...)
{
va_list args;
va_start(args, msg);
if (verbose)
{
int len = strlen(msg);
if (len > 0)
partial_line = (msg[len-1] != '\n');
vfprintf(stderr, msg, args);
fflush(stderr);
}
va_end(args);
}
/*** Utility functions ***/
static void *
xmalloc(size_t size)
{
void *p = malloc(size);
if (!p)
die("Out of memory");
return p;
}
static char *
xstrdup(char *str)
{
char *p = strdup(str);
if (!p)
die("Out of memory");
return p;
}
static int dir_exists(char *path)
{
struct stat st;
return (stat(path, &st) >= 0 && S_ISDIR(st.st_mode));
}
/*** Environment rules ***/
struct env_rule {
char *var; // Variable to match
char *val; // ""=clear, NULL=inherit
int var_len;
struct env_rule *next;
};
static struct env_rule *first_env_rule;
static struct env_rule **last_env_rule = &first_env_rule;
static struct env_rule default_env_rules[] = {
{ "LIBC_FATAL_STDERR_", "1" }
};
static int
set_env_action(char *a0)
{
struct env_rule *r = xmalloc(sizeof(*r) + strlen(a0) + 1);
char *a = (char *)(r+1);
strcpy(a, a0);
char *sep = strchr(a, '=');
if (sep == a)
return 0;
r->var = a;
if (sep)
{
*sep++ = 0;
r->val = sep;
}
else
r->val = NULL;
*last_env_rule = r;
last_env_rule = &r->next;
r->next = NULL;
return 1;
}
static int
match_env_var(char *env_entry, struct env_rule *r)
{
if (strncmp(env_entry, r->var, r->var_len))
return 0;
return (env_entry[r->var_len] == '=');
}
static void
apply_env_rule(char **env, int *env_sizep, struct env_rule *r)
{
// First remove the variable if already set
int pos = 0;
while (pos < *env_sizep && !match_env_var(env[pos], r))
pos++;
if (pos < *env_sizep)
{
(*env_sizep)--;
env[pos] = env[*env_sizep];
env[*env_sizep] = NULL;
}
// What is the new value?
char *new;
if (r->val)
{
if (!r->val[0])
return;
new = xmalloc(r->var_len + 1 + strlen(r->val) + 1);
sprintf(new, "%s=%s", r->var, r->val);
}
else
{
pos = 0;
while (environ[pos] && !match_env_var(environ[pos], r))
pos++;
if (!(new = environ[pos]))
return;
}
// Add it at the end of the array
env[(*env_sizep)++] = new;
env[*env_sizep] = NULL;
}
static char **
setup_environment(void)
{
// Link built-in rules with user rules
for (int i=ARRAY_SIZE(default_env_rules)-1; i >= 0; i--)
{
default_env_rules[i].next = first_env_rule;
first_env_rule = &default_env_rules[i];
}
// Scan the original environment
char **orig_env = environ;
int orig_size = 0;
while (orig_env[orig_size])
orig_size++;
// For each rule, reserve one more slot and calculate length
int num_rules = 0;
for (struct env_rule *r = first_env_rule; r; r=r->next)
{
num_rules++;
r->var_len = strlen(r->var);
}
// Create a new environment
char **env = xmalloc((orig_size + num_rules + 1) * sizeof(char *));
int size;
if (pass_environ)
{
memcpy(env, environ, orig_size * sizeof(char *));
size = orig_size;
}
else
size = 0;
env[size] = NULL;
// Apply the rules one by one
for (struct env_rule *r = first_env_rule; r; r=r->next)
apply_env_rule(env, &size, r);
// Return the new env and pass some gossip
if (verbose > 1)
{
fprintf(stderr, "Passing environment:\n");
for (int i=0; env[i]; i++)
fprintf(stderr, "\t%s\n", env[i]);
}
return env;
}
/*** Directory rules ***/
struct dir_rule {
char *inside; // A relative path
char *outside; // This can be an absolute path or a relative path starting with "./"
unsigned int flags; // DIR_FLAG_xxx
struct dir_rule *next;
};
enum dir_rule_flags {
DIR_FLAG_RW = 1,
DIR_FLAG_NOEXEC = 2,
DIR_FLAG_FS = 4,
DIR_FLAG_MAYBE = 8,
DIR_FLAG_DEV = 16,
};
static const char * const dir_flag_names[] = { "rw", "noexec", "fs", "maybe", "dev" };
static struct dir_rule *first_dir_rule;
static struct dir_rule **last_dir_rule = &first_dir_rule;
static int add_dir_rule(char *in, char *out, unsigned int flags)
{
// Make sure that "in" is relative
while (in[0] == '/')
in++;
if (!*in)
return 0;
// Check "out"
if (flags & DIR_FLAG_FS)
{
if (!out || out[0] == '/')
return 0;
}
else
{
if (out && out[0] != '/' && strncmp(out, "./", 2))
return 0;
}
// Override an existing rule
struct dir_rule *r;
for (r = first_dir_rule; r; r = r->next)
if (!strcmp(r->inside, in))
break;
// Add a new rule
if (!r)
{
r = xmalloc(sizeof(*r));
r->inside = in;
*last_dir_rule = r;
last_dir_rule = &r->next;
r->next = NULL;
}
r->outside = out;
r->flags = flags;
return 1;
}
static unsigned int parse_dir_option(char *opt)
{
for (unsigned int i = 0; i < ARRAY_SIZE(dir_flag_names); i++)
if (!strcmp(opt, dir_flag_names[i]))
return 1U << i;
die("Unknown directory option %s", opt);
}
static int set_dir_action(char *arg)
{
arg = xstrdup(arg);
char *colon = strchr(arg, ':');
unsigned int flags = 0;
while (colon)
{
*colon++ = 0;
char *next = strchr(colon, ':');
if (next)
*next = 0;
flags |= parse_dir_option(colon);
colon = next;
}
char *eq = strchr(arg, '=');
if (eq)
{
*eq++ = 0;
return add_dir_rule(arg, (*eq ? eq : NULL), flags);
}
else
{
char *out = xmalloc(1 + strlen(arg) + 1);
sprintf(out, "/%s", arg);
return add_dir_rule(arg, out, flags);
}
}
static void init_dir_rules(void)
{
set_dir_action("box=./box:rw");
set_dir_action("bin");
set_dir_action("dev:dev");
set_dir_action("lib");
set_dir_action("lib64:maybe");
set_dir_action("proc=proc:fs");
set_dir_action("usr");
}
static void make_dir(char *path)
{
char *sep = (path[0] == '/' ? path+1 : path);
for (;;)
{
sep = strchr(sep, '/');
if (sep)
*sep = 0;
if (!dir_exists(path) && mkdir(path, 0777) < 0)
die("Cannot create directory %s: %m\n", path);
if (!sep)
return;
*sep++ = '/';
}
}
static void apply_dir_rules(void)
{
for (struct dir_rule *r = first_dir_rule; r; r=r->next)
{
char *in = r->inside;
char *out = r->outside;
if (!out)
{
msg("Not binding anything on %s\n", r->inside);
continue;
}
if ((r->flags & DIR_FLAG_MAYBE) && !dir_exists(out))
{
msg("Not binding %s on %s (does not exist)\n", out, r->inside);
continue;
}
char root_in[1024];
snprintf(root_in, sizeof(root_in), "root/%s", in);
make_dir(root_in);
unsigned long mount_flags = 0;
if (!(r->flags & DIR_FLAG_RW))
mount_flags |= MS_RDONLY;
if (r->flags & DIR_FLAG_NOEXEC)
mount_flags |= MS_NOEXEC;
if (!(r->flags & DIR_FLAG_DEV))
mount_flags |= MS_NODEV;
if (r->flags & DIR_FLAG_FS)
{
msg("Mounting %s on %s (flags %lx)\n", out, in, mount_flags);
if (mount("none", root_in, out, mount_flags, "") < 0)
die("Cannot mount %s on %s: %m", out, in);
}
else
{
mount_flags |= MS_BIND | MS_NOSUID;
msg("Binding %s on %s (flags %lx)\n", out, in, mount_flags);
// Most mount flags need remount to work
if (mount(out, root_in, "none", mount_flags, "") < 0 ||
mount(out, root_in, "none", MS_REMOUNT | mount_flags, "") < 0)
die("Cannot mount %s on %s: %m", out, in);
}
}
}
/*** Control groups ***/
static char cg_path[256];
#define CG_BUFSIZE 1024
static int
cg_read(char *attr, char *buf)
{
int maybe = 0;
if (attr[0] == '?')
{
attr++;
maybe = 1;
}
char path[256];
snprintf(path, sizeof(path), "%s/%s", cg_path, attr);
int fd = open(path, O_RDONLY);
if (fd < 0)
{
if (maybe)
return 0;
die("Cannot read %s: %m", path);
}
int n = read(fd, buf, CG_BUFSIZE);
if (n < 0)
die("Cannot read %s: %m", path);
if (n >= CG_BUFSIZE - 1)
die("Attribute %s too long", path);
if (n > 0 && buf[n-1] == '\n')
n--;
buf[n] = 0;
if (verbose > 1)
msg("CG: Read %s = %s\n", attr, buf);
close(fd);
return 1;
}
static void __attribute__((format(printf,2,3)))
cg_write(char *attr, char *fmt, ...)
{
va_list args;
va_start(args, fmt);
char buf[CG_BUFSIZE];
int n = vsnprintf(buf, sizeof(buf), fmt, args);
if (n >= CG_BUFSIZE)
die("cg_writef: Value for attribute %s is too long", attr);
if (verbose > 1)
msg("CG: Write %s = %s", attr, buf);
char path[256];
snprintf(path, sizeof(path), "%s/%s", cg_path, attr);
int fd = open(path, O_WRONLY | O_TRUNC);
if (fd < 0)
die("Cannot write %s: %m", path);
int written = write(fd, buf, n);
if (written < 0)
die("Cannot set %s to %s: %m", path, buf);
if (written != n)
die("Short write to %s (%d out of %d bytes)", path, written, n);
close(fd);
va_end(args);
}
static void
cg_init(void)
{
if (!cg_enable)
return;
char *cg_root = CONFIG_ISOLATE_CGROUP_ROOT;
if (!dir_exists(cg_root))
die("Control group filesystem at %s not mounted", cg_root);
snprintf(cg_path, sizeof(cg_path), "%s/box-%d", cg_root, box_id);
msg("Using control group %s\n", cg_path);
}
static void
cg_prepare(void)
{
if (!cg_enable)
return;
struct stat st;
char buf[CG_BUFSIZE];
if (stat(cg_path, &st) >= 0 || errno != ENOENT)
{
msg("Control group %s already exists, trying to empty it.\n", cg_path);
if (rmdir(cg_path) < 0)
die("Failed to reset control group %s: %m", cg_path);
}
if (mkdir(cg_path, 0777) < 0)
die("Failed to create control group %s: %m", cg_path);
// If cpuset module is enabled, copy allowed cpus and memory nodes from parent group
if (cg_read("?../cpuset.cpus", buf))
cg_write("cpuset.cpus", "%s", buf);
if (cg_read("?../cpuset.mems", buf))
cg_write("cpuset.mems", "%s", buf);
}
static void
cg_enter(void)
{
if (!cg_enable)
return;
msg("Entering control group %s\n", cg_path);
struct stat st;
if (stat(cg_path, &st) < 0)
die("Control group %s does not exist: %m", cg_path);
if (cg_memory_limit)
{
cg_write("memory.limit_in_bytes", "%lld\n", (long long) cg_memory_limit << 10);
cg_write("memory.memsw.limit_in_bytes", "%lld\n", (long long) cg_memory_limit << 10);
}
if (cg_timing)
cg_write("cpuacct.usage", "0\n");
cg_write("tasks", "%d\n", (int) getpid());
}
static int
cg_get_run_time_ms(void)
{
if (!cg_enable)
return 0;
char buf[CG_BUFSIZE];
cg_read("cpuacct.usage", buf);
unsigned long long ns = atoll(buf);
return ns / 1000000;
}
static void
cg_stats(void)
{
if (!cg_enable)
return;
char buf[CG_BUFSIZE];
// Memory usage statistics
unsigned long long mem=0, memsw=0;
if (cg_read("?memory.max_usage_in_bytes", buf))
mem = atoll(buf);
if (cg_read("?memory.memsw.max_usage_in_bytes", buf))
{
memsw = atoll(buf);
if (memsw > mem)
mem = memsw;
}
if (mem)
meta_printf("cg-mem:%lld\n", mem >> 10);
}
static void
cg_remove(void)
{
char buf[CG_BUFSIZE];
if (!cg_enable)
return;
cg_read("tasks", buf);
if (buf[0])
die("Some tasks left in control group %s, failed to remove it", cg_path);
if (rmdir(cg_path) < 0)
die("Cannot remove control group %s: %m", cg_path);
}
/*** Disk quotas ***/
static int
path_begins_with(char *path, char *with)
{
while (*with)
if (*path++ != *with++)
return 0;
return (!*with || *with == '/');
}
static char *
find_device(char *path)
{
FILE *f = setmntent("/proc/mounts", "r");
if (!f)
die("Cannot open /proc/mounts: %m");
struct mntent *me;
int best_len = 0;
char *best_dev = NULL;
while (me = getmntent(f))
{
if (!path_begins_with(me->mnt_fsname, "/dev"))
continue;
if (path_begins_with(path, me->mnt_dir))
{
int len = strlen(me->mnt_dir);
if (len > best_len)
{
best_len = len;
free(best_dev);
best_dev = xstrdup(me->mnt_fsname);
}
}
}
endmntent(f);
return best_dev;
}
static void
set_quota(void)
{
if (!block_quota)
return;
char cwd[PATH_MAX];
if (!getcwd(cwd, sizeof(cwd)))
die("getcwd: %m");
char *dev = find_device(cwd);
if (!dev)
die("Cannot identify filesystem which contains %s", cwd);
msg("Quota: Mapped path %s to a filesystem on %s\n", cwd, dev);
// Sanity check
struct stat dev_st, cwd_st;
if (stat(dev, &dev_st) < 0)
die("Cannot identify block device %s: %m", dev);
if (!S_ISBLK(dev_st.st_mode))
die("Expected that %s is a block device", dev);
if (stat(".", &cwd_st) < 0)
die("Cannot stat cwd: %m");
if (cwd_st.st_dev != dev_st.st_rdev)
die("Identified %s as a filesystem on %s, but it is obviously false", cwd, dev);
struct dqblk dq = {
.dqb_bhardlimit = block_quota,
.dqb_bsoftlimit = block_quota,
.dqb_ihardlimit = inode_quota,
.dqb_isoftlimit = inode_quota,
.dqb_valid = QIF_LIMITS,
};
if (quotactl(QCMD(Q_SETQUOTA, USRQUOTA), dev, box_uid, (caddr_t) &dq) < 0)
die("Cannot set disk quota: %m");
msg("Quota: Set block quota %d and inode quota %d\n", block_quota, inode_quota);
free(dev);
}
/*** The keeper process ***/
static void
signal_alarm(int unused UNUSED)
{
/* Time limit checks are synchronous, so we only schedule them there. */
timer_tick = 1;
alarm(1);
}
static void
signal_int(int unused UNUSED)
{
/* Interrupts are fatal, so no synchronization requirements. */
meta_printf("exitsig:%d\n", SIGINT);
err("SG: Interrupted");
}
#define PROC_BUF_SIZE 4096
static void
read_proc_file(char *buf, char *name, int *fdp)
{
int c;
if (!*fdp)
{
sprintf(buf, "/proc/%d/%s", (int) box_pid, name);
*fdp = open(buf, O_RDONLY);
if (*fdp < 0)
die("open(%s): %m", buf);
}
lseek(*fdp, 0, SEEK_SET);
if ((c = read(*fdp, buf, PROC_BUF_SIZE-1)) < 0)
die("read on /proc/$pid/%s: %m", name);
if (c >= PROC_BUF_SIZE-1)
die("/proc/$pid/%s too long", name);
buf[c] = 0;
}
static int
get_wall_time_ms(void)
{
struct timeval now, wall;
gettimeofday(&now, NULL);
timersub(&now, &start_time, &wall);
return wall.tv_sec*1000 + wall.tv_usec/1000;
}
static int
get_run_time_ms(struct rusage *rus)
{
if (cg_timing)
return cg_get_run_time_ms();
if (rus)
{
struct timeval total;
timeradd(&rus->ru_utime, &rus->ru_stime, &total);
return total.tv_sec*1000 + total.tv_usec/1000;
}
char buf[PROC_BUF_SIZE], *x;
int utime, stime;
static int proc_stat_fd;
read_proc_file(buf, "stat", &proc_stat_fd);
x = buf;
while (*x && *x != ' ')
x++;
while (*x == ' ')
x++;
if (*x++ != '(')
die("proc stat syntax error 1");
while (*x && (*x != ')' || x[1] != ' '))
x++;
while (*x == ')' || *x == ' ')
x++;
if (sscanf(x, "%*c %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d %d %d", &utime, &stime) != 2)
die("proc stat syntax error 2");
return (utime + stime) * 1000 / ticks_per_sec;
}
static void
check_timeout(void)
{
if (wall_timeout)
{
int wall_ms = get_wall_time_ms();
if (wall_ms > wall_timeout)
err("TO: Time limit exceeded (wall clock)");
if (verbose > 1)
fprintf(stderr, "[wall time check: %d msec]\n", wall_ms);
}
if (timeout)
{
int ms = get_run_time_ms(NULL);
if (verbose > 1)
fprintf(stderr, "[time check: %d msec]\n", ms);
if (ms > timeout && ms > extra_timeout)
err("TO: Time limit exceeded");
}
}
static void
box_keeper(void)
{
read_errors_from_fd = error_pipes[0];
close(error_pipes[1]);
struct sigaction sa;
bzero(&sa, sizeof(sa));
sa.sa_handler = signal_int;
sigaction(SIGINT, &sa, NULL);
gettimeofday(&start_time, NULL);
ticks_per_sec = sysconf(_SC_CLK_TCK);
if (ticks_per_sec <= 0)
die("Invalid ticks_per_sec!");
if (timeout || wall_timeout)
{
sa.sa_handler = signal_alarm;
sigaction(SIGALRM, &sa, NULL);
alarm(1);
}
for(;;)
{
struct rusage rus;
int stat;
pid_t p;
if (timer_tick)
{
check_timeout();
timer_tick = 0;
}
p = wait4(box_pid, &stat, 0, &rus);
if (p < 0)
{
if (errno == EINTR)
continue;
die("wait4: %m");
}
if (p != box_pid)
die("wait4: unknown pid %d exited!", p);
box_pid = 0;
// Check error pipe if there is an internal error passed from inside the box
char interr[1024];
int n = read(read_errors_from_fd, interr, sizeof(interr) - 1);
if (n > 0)