-
Notifications
You must be signed in to change notification settings - Fork 3
/
selfdock.c
413 lines (370 loc) · 10.1 KB
/
selfdock.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
/*
* Copyright 2015-2020 Andreas Nordal
*
* This Source Code Form is subject to the terms of the
* Mozilla Public License, v. 2.0.
* If a copy of the MPL was not distributed with this file,
* you can obtain one at https://mozilla.org/MPL/2.0/.
*/
#define _GNU_SOURCE
#include <narg.h>
#include <libintl.h>
#include <sched.h> //clone
#include <errno.h>
#include <fcntl.h> //open, AT_FDCWD
#include <sys/stat.h> //mkdir, utimensat
#include <sys/wait.h>
#include <sys/mount.h>
#include <unistd.h>
#include <stdlib.h> //getenv
#include <stdio.h>
#include <string.h>
#include <sys/mman.h> //mmap
#define STRINGIFY(x) #x
#define TOSTRING(x) STRINGIFY(x)
#define ARRAY_SIZE(x) sizeof(x)/sizeof(x[0])
#ifndef PATH_MAX
# define PATH_MAX 1024
#endif
#ifndef ROOTOVERLAY
# error ROOTOVERLAY not defined
#endif
// Try to conform or give way to existing exit status conventions
#define EXIT_NAME_IN_USE 123 //self-defined
#define EXIT_CANNOT 124 //self-defined
#define EXIT_UNTESTABLE 125 //inapplicable convention (git-bisect)
#define EXIT_CMDNOTEXEC 126 //applicable convention
#define EXIT_CMDNOTFOUND 127 //applicable convention
static volatile int sigrid;
static void take_signal(int sig)
{
sigrid = sig;
}
static int start_handling_signals()
{
struct sigaction sa;
sa.sa_handler = take_signal;
sigemptyset(&sa.sa_mask);
sa.sa_flags = 0;
static const int handleable_signals[] = {
SIGHUP,
SIGINT,
SIGUSR1,
SIGUSR2,
SIGPIPE,
SIGTERM
};
for (unsigned i=0; i != ARRAY_SIZE(handleable_signals); ++i) {
int sig = handleable_signals[i];
if (sigaction(sig, &sa, NULL)) {
return sig;
}
}
return 0;
}
static _Bool is_suid(const char *path)
{
struct stat info;
return 0 == stat(path, &info) && info.st_mode & S_ISUID;
}
// is what execvp calls a pathname && exists && is (a symlink to) a directory
static _Bool isdir_pathname(const char *path)
{
struct stat info;
return strchr(path, '/') && 0 == lstat(path, &info) && S_ISDIR(info.st_mode);
}
// Updates atime if writable → only suitable when not supposed to.
static _Bool check_erofs(const char* path)
{
static const struct timespec atime[2] = {
{ 0, UTIME_NOW },
{ 0, UTIME_OMIT },
};
return utimensat(AT_FDCWD, path, atime, 0) != 0 && errno == EROFS;
}
static int mount_bind_rw(const char* src, const char* dst)
{
if (mount(src, dst, NULL, MS_BIND, NULL)) {
fprintf(stderr, "bindmount «%s» → «%s»: %s\n", src, dst, strerror(errno));
return -1;
}
return 0;
}
static int mount_bind_ro(const char* src, const char* dst)
{
if (mount_bind_rw(src, dst)) {
return -1;
}
const char *errmsg;
if (mount(NULL, dst, NULL, MS_REMOUNT | MS_BIND | MS_RDONLY, NULL)) {
errmsg = strerror(errno);
goto fail;
}
if (!check_erofs(dst)) {
errmsg = "Still not readonly! This is supported from Linux 2.6.26, see mount(2).";
goto fail;
}
return 0;
fail:
fprintf(stderr, "remount,bind,ro %s: %s\n", dst, errmsg);
return -1;
}
struct child_args {
_Bool permit_writable;
_Bool has_tmp;
const char *oldroot;
const char *cd;
struct narg_optparam *map, *vol, *tmp;
char *const *argv;
};
static int child(void *arg)
{
const struct child_args *self = arg;
// The containing mountpoint must be marked private. How this is supposed
// to be done is AFAIK undocumented, but this trick, recursing from root,
// was taken from http://sourceforge.net/p/fuse/mailman/message/24957287/
if (mount(NULL, "/", NULL, MS_PRIVATE|MS_REC, NULL)) {
fprintf(stderr, "Failed to mark all mounts private: %s\n", strerror(errno));
return EXIT_CANNOT;
}
static const char newroot[] = TOSTRING(ROOTOVERLAY) "/dev/empty";
if (self->permit_writable
? mount_bind_rw(self->oldroot, newroot)
: mount_bind_ro(self->oldroot, newroot))
{
return EXIT_CANNOT;
}
if (chdir(newroot)) {
fprintf(stderr, "chdir: %s: %s\n", newroot, strerror(errno));
return EXIT_CANNOT;
}
if (mount_bind_ro(TOSTRING(ROOTOVERLAY) "/dev", "dev")) {
return EXIT_CANNOT;
}
for (unsigned i=0; i < self->map->paramc; i += 2) {
if (mount_bind_ro(self->map->paramv[i], self->map->paramv[i+1]+1)) {
return EXIT_CANNOT;
}
}
for (unsigned i=0; i < self->vol->paramc; i += 2) {
if (mount_bind_rw(self->vol->paramv[i], self->vol->paramv[i+1]+1)) {
return EXIT_CANNOT;
}
}
if (chroot(".")) {
fprintf(stderr, "chroot: %s: %s\n", newroot, strerror(errno));
return EXIT_CANNOT;
}
if (mount("none", "proc", "proc", MS_NOEXEC, NULL)) {
perror("mount proc");
return EXIT_CANNOT;
}
if (mount("none", "dev/pts", "devpts", MS_NOEXEC, NULL)) {
perror("mount devpts");
return EXIT_CANNOT;
}
for (unsigned i=0; i < self->tmp->paramc; i += 2) {
const char *dst = self->tmp->paramv[i+1];
if (mount("none", dst, "tmpfs", MS_NOEXEC, self->tmp->paramv[i])
|| chmod(dst, 0777))
{
perror(dst);
return EXIT_CANNOT;
}
}
if (!self->has_tmp) {
if (mount("none", "/tmp", "tmpfs", MS_NOEXEC, "size=2M")
|| chmod("/tmp", 0777))
{
perror("/tmp");
return EXIT_CANNOT;
}
}
// Drop effective uid
if (setuid(getuid())) {
perror("setuid");
return EXIT_CANNOT;
}
if (chdir(self->cd)) {
perror(self->cd);
return EXIT_CANNOT;
}
execvp(self->argv[0], self->argv);
// fail
int errval = errno;
if (errval == EACCES && isdir_pathname(self->argv[0])) {
errval = EISDIR;
}
fprintf(stderr, "exec: %s: %s\n", self->argv[0], strerror(errval));
switch(errval) {
case ELOOP:
case ENAMETOOLONG:
case ENOENT:
case ENOTDIR:
return EXIT_CMDNOTFOUND;
default:
return EXIT_CMDNOTEXEC;
}
}
const char *narg_strerror(unsigned err) {
switch (err) {
case NARG_ENOSUCHOPTION: return "No such option";
case NARG_EMISSINGPARAM: return "Missing parameter";
case NARG_EUNEXPECTEDPARAM: return "Unexpected parameter";
case NARG_EILSEQ: return strerror(EILSEQ);
}
return "Unknown error";
}
void help(const struct narg_optspec *optv, struct narg_optparam *ansv, unsigned optc) {
unsigned width = narg_terminalwidth(stdout);
flockfile(stdout);
narg_printopt_unlocked(stdout, width, optv, ansv, optc, dgettext, NULL, 2);
funlockfile(stdout);
}
int main(int argc, char *argv[])
{
enum {
OPT_HELP,
OPT_ROOT,
OPT_CD,
OPT_MAP,
OPT_VOL,
OPT_TMP,
OPT_ENV,
OPT_ENV_RM,
OPT_IGN,
OPT_MAX
};
static const struct narg_optspec optv[OPT_MAX] = {
{"h","help",NULL,"Show help text"},
{"r","rootfs"," DIR","Directory to use as root filesystem"},
{"C", NULL, "DIR","Working directory"},
{"m","map"," SRC DST","Mount SRC to DST read-only"},
{"v","vol"," SRC DST","Mount SRC to DST read-write"},
{"t","tmpfs"," OPT DST","Mount tmpfs with options OPT at DST"},
{"e","env"," ENV val","Set environment variable ENV to val"},
{"E", NULL, "ENV","Unset environment variable ENV"},
{NULL,"",&narg_metavar.ignore_rest,"Don\'t interpret further arguments as options"}
};
struct narg_optparam ansv[OPT_MAX] = {
[OPT_HELP] = {0, NULL},
[OPT_ROOT] = {1, (const char*[]){"/"}},
[OPT_CD ] = {1, (const char*[]){"/"}},
[OPT_MAP] = {0, NULL},
[OPT_VOL] = {0, NULL},
[OPT_TMP] = {0, NULL},
[OPT_ENV] = {0, NULL},
[OPT_ENV_RM] = {0, NULL},
[OPT_IGN] = {0, NULL}
};
struct narg_result nargres = narg_findopt(argv, optv, ansv, OPT_MAX, 1, 2);
if (nargres.err) {
fprintf(stderr, "%s: %s\n", argv[nargres.arg], narg_strerror(nargres.err));
return EXIT_CANNOT;
}
if (ansv[OPT_HELP].paramc) {
ansv[OPT_HELP].paramc = 0;
help(optv, ansv, OPT_MAX);
return EXIT_SUCCESS;
}
for (unsigned i=0; i < ansv[OPT_ENV].paramc; i += 2) {
const char *env = ansv[OPT_ENV].paramv[i];
const char *val = ansv[OPT_ENV].paramv[i+1];
if (setenv(env, val, ~0)) {
fprintf(stderr, "setenv %s=%s: %s\n", env, val, strerror(errno));
return EXIT_CANNOT;
}
}
for (unsigned i=0; i < ansv[OPT_ENV_RM].paramc; i += 1) {
const char *env = ansv[OPT_ENV_RM].paramv[i];
if (unsetenv(env)) {
fprintf(stderr, "unsetenv %s: %s\n", env, strerror(errno));
return EXIT_CANNOT;
}
}
struct child_args barnebok;
barnebok.permit_writable = 0;
barnebok.has_tmp = 0;
barnebok.oldroot = ansv[OPT_ROOT].paramv[0];
barnebok.cd = ansv[OPT_CD].paramv[0];
barnebok.map = ansv+OPT_MAP;
barnebok.vol = ansv+OPT_VOL;
barnebok.tmp = ansv+OPT_TMP;
barnebok.argv = argv + nargres.arg + 1; // += optional + positional args
if (argc - nargres.arg < 2) {
printf(
"Usage: %s run|build [OPTIONS] argv\n"
, argv[0]);
return EXIT_CANNOT;
}
if (0 == strcmp(argv[nargres.arg], "run")) {
} else if (0 == strcmp(argv[nargres.arg], "build")) {
barnebok.permit_writable = ~0;
} else {
fputs("Action must be \"run\" or \"build\" for now. TODO: enter\n", stderr);
return EXIT_CANNOT;
}
for (unsigned o=OPT_MAP; o <= OPT_TMP; ++o) {
for (unsigned i=0; i < ansv[o].paramc; i += 2) {
const char *dst = ansv[o].paramv[i+1];
if (dst[0] != '/') {
fprintf(stderr, "--%s destinations must be absolute\n", optv[o].longopt);
return EXIT_CANNOT;
}
if (0 == strcmp(dst+1, "tmp")) {
barnebok.has_tmp = ~0;
}
}
}
int sigfail = start_handling_signals();
if (sigfail) {
fprintf(stderr, "sigaction(sig=%d): %s\n", sigfail, strerror(errno));
return EXIT_CANNOT;
}
// Apparently, this is just the stack size before exec.
// Apparently, 3 4K pages are needed to print to stderr (1 for stdout).
const unsigned initial_stack_size = 16*4096;
char *stack = mmap(
NULL, initial_stack_size,
PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_ANONYMOUS,
-1, 0
);
if (stack == MAP_FAILED) {
perror("mmap");
return EXIT_CANNOT;
}
pid_t pid = clone(
child, stack + initial_stack_size,
CLONE_VFORK|CLONE_VM|CLONE_NEWNS|CLONE_NEWPID|SIGCHLD,
&barnebok
);
int ret = EXIT_CANNOT;
if (pid == -1) {
if (!is_suid(argv[0])) {
fprintf(stderr, "No suid. Please check that the program is installed correctly.\n");
} else {
fprintf(stderr, "clone: %s\n", strerror(errno));
}
} else do {
int status;
if (-1 == wait(&status)) {
if (errno == EINTR) {
int deliver = sigrid;
if (kill(pid, deliver)) {
fprintf(stderr, "Failed to deliver signal %s\n", strsignal(deliver));
}
continue;
}
perror("wait");
} else if (WIFEXITED(status)) {
ret = WEXITSTATUS(status);
} else if (WIFSIGNALED(status)) {
psignal(WTERMSIG(status), barnebok.argv[0]);
}
} while (0);
if (munmap(stack, initial_stack_size)) {
perror("munmap");
}
return ret;
}