summaryrefslogtreecommitdiffstats
path: root/src/shutdown.c
blob: b78c74112a0c1930703e9c88963de4562cd246e6 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/

/***
  This file is part of systemd.

  Copyright 2010 ProFUSION embedded systems

  systemd 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.

  systemd is distributed in the hope that it will be useful, but
  WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  General Public License for more details.

  You should have received a copy of the GNU General Public License
  along with systemd; If not, see <http://www.gnu.org/licenses/>.
***/

#include <sys/mman.h>
#include <sys/types.h>
#include <sys/reboot.h>
#include <linux/reboot.h>
#include <sys/wait.h>
#include <dirent.h>
#include <errno.h>
#include <unistd.h>
#include <signal.h>
#include <stdbool.h>
#include <stdlib.h>
#include <string.h>

#include "log.h"
#include "umount.h"
#include "util.h"

#define TIMEOUT_USEC    (5 * USEC_PER_SEC)
#define FINALIZE_ATTEMPTS 50
#define FINALIZE_CRITICAL_ATTEMPTS 10

static bool ignore_proc(pid_t pid) {
        if (pid == 1)
                return true;

        /* TODO: add more ignore rules here: device-mapper, etc */

        return false;
}

static bool is_kernel_thread(pid_t pid)
{
        char buf[PATH_MAX];
        FILE *f;
        char c;
        size_t count;

        snprintf(buf, sizeof(buf), "/proc/%lu/cmdline", (unsigned long)pid);
        f = fopen(buf, "re");
        if (!f)
                return true; /* not really, but has the desired effect */

        count = fread(&c, 1, 1, f);
        fclose(f);
        return count != 1;
}

static int killall(int sign) {
        DIR *dir;
        struct dirent *d;
        unsigned int processes = 0;

        if ((dir = opendir("/proc")) == NULL)
                return -errno;

        while ((d = readdir(dir))) {
                pid_t pid;

                if (parse_pid(d->d_name, &pid) < 0)
                        continue;

                if (is_kernel_thread(pid))
                        continue;

                if (ignore_proc(pid))
                        continue;

                if (kill(pid, sign) == 0)
                        processes++;
                else
                        log_warning("Could not kill %d: %m", pid);
        }

        closedir(dir);

        return processes;
}

static int send_signal(int sign) {
        sigset_t mask, oldmask;
        usec_t until;
        int processes;
        struct timespec ts;

        assert_se(sigemptyset(&mask) == 0);
        assert_se(sigaddset(&mask, SIGCHLD) == 0);
        if (sigprocmask(SIG_BLOCK, &mask, &oldmask) != 0)
                return -errno;

        if (kill(-1, SIGSTOP) < 0)
                log_warning("Failed kill(-1, SIGSTOP): %m");

        processes = killall(sign);

        if (kill(-1, SIGCONT) < 0)
                log_warning("Failed kill(-1, SIGCONT): %m");

        if (processes <= 0)
                goto finish;

        until = now(CLOCK_MONOTONIC) + TIMEOUT_USEC;
        for (;;) {
                usec_t n = now(CLOCK_MONOTONIC);
                for (;;) {
                        pid_t pid = waitpid(-1, NULL, WNOHANG);
                        if (pid == 0)
                                break;
                        else if (pid < 0 && errno == ECHILD) {
                                processes = 0;
                                goto finish;
                        }

                        if (--processes == 0)
                                goto finish;
                }

                if (n >= until)
                        goto finish;

                timespec_store(&ts, until - n);
                if (sigtimedwait(&mask, NULL, &ts) != SIGCHLD)
                        log_warning("Failed: sigtimedwait did not return SIGCHLD: %m");
        }

finish:
        sigprocmask(SIG_SETMASK, &oldmask, NULL);

        return processes;
}

static int rescue_send_signal(int sign) {
        sigset_t mask, oldmask;
        usec_t until;
        struct timespec ts;
        int r;

        sigemptyset(&mask);
        sigaddset(&mask, SIGCHLD);
        if (sigprocmask(SIG_BLOCK, &mask, &oldmask) != 0)
                return -errno;

        if (kill(-1, SIGSTOP) < 0)
                log_warning("Failed kill(-1, SIGSTOP): %m");

        r = kill(-1, sign);
        if (r < 0)
                log_warning("Failed kill(-1, %d): %m", sign);

        if (kill(-1, SIGCONT) < 0)
                log_warning("Failed kill(-1, SIGCONT): %m");

        if (r < 0)
                goto finish;

        until = now(CLOCK_MONOTONIC) + TIMEOUT_USEC;
        for (;;) {
                usec_t n = now(CLOCK_MONOTONIC);
                for (;;) {
                        pid_t pid = waitpid(-1, NULL, WNOHANG);
                        if (pid == 0)
                                break;
                        else if (pid < 0 && errno == ECHILD)
                                goto finish;
                }

                if (n >= until)
                        goto finish;

                timespec_store(&ts, until - n);
                if (sigtimedwait(&mask, NULL, &ts) != SIGCHLD)
                        log_warning("Failed: sigtimedwait did not return SIGCHLD: %m");
        }

finish:
        sigprocmask(SIG_SETMASK, &oldmask, NULL);

        return r;
}


int main(int argc, char *argv[]) {
        int cmd, r, retries;
        bool need_umount = true, need_swapoff = true, need_loop_detach = true;

        log_parse_environment();
        log_set_target(LOG_TARGET_KMSG); /* syslog will die if not gone yet */
        log_open();

        if (getpid() != 1) {
                log_error("Not executed by init (pid-1).");
                r = -EPERM;
                goto error;
        }

        if (argc != 2) {
                log_error("Invalid number of arguments.");
                r = -EINVAL;
                goto error;
        }

        if (streq(argv[1], "reboot"))
                cmd = RB_AUTOBOOT;
        else if (streq(argv[1], "poweroff"))
                cmd = RB_POWER_OFF;
        else if (streq(argv[1], "halt"))
                cmd = RB_HALT_SYSTEM;
        else if (streq(argv[1], "kexec"))
                cmd = LINUX_REBOOT_CMD_KEXEC;
        else {
                log_error("Unknown action '%s'.", argv[1]);
                r = -EINVAL;
                goto error;
        }

        /* lock us into memory */
        if (mlockall(MCL_CURRENT|MCL_FUTURE) != 0)
                log_warning("Cannot lock process memory: %m");

        log_info("Sending SIGTERM to processes");
        r = send_signal(SIGTERM);
        if (r < 0)
                log_warning("Cannot send SIGTERM to all process: %s", strerror(r));

        log_info("Sending SIGKILL to processes");
        r = send_signal(SIGKILL);
        if (r < 0)
                log_warning("Cannot send SIGKILL to all process: %s", strerror(r));


        /* preventing that we won't block umounts */
        if (chdir("/") != 0)
                log_warning("Cannot chdir(\"/\"): %m. Unmounts likely to fail.");

        /* umount all mountpoints, swaps, and loopback devices */
        retries = FINALIZE_ATTEMPTS;
        while (need_umount || need_swapoff || need_loop_detach) {
                if (need_umount) {
                        log_info("Unmounting filesystems.");
                        r = umount_all();
                        if (r == 0)
                                need_umount = false;
                        else if (r > 0)
                                log_warning("Not all filesystems unmounted, %d left.", r);
                        else
                                log_error("Error unmounting filesystems: %s", strerror(-r));
                }

                if (need_swapoff) {
                        log_info("Disabling swaps.");
                        r = swapoff_all();
                        if (r == 0)
                                need_swapoff = false;
                        else if (r > 0)
                                log_warning("Not all swaps are off, %d left.", r);
                        else
                                log_error("Error turning off swaps: %s", strerror(-r));
                }

                if (need_loop_detach) {
                        log_info("Detaching loop devices.");
                        r = loopback_detach_all();
                        if (r == 0)
                                need_loop_detach = false;
                        else if (r > 0)
                                log_warning("Not all loop devices detached, %d left.", r);
                        else
                                log_error("Error detaching loop devices: %s", strerror(-r));

                }

                if (need_umount || need_swapoff || need_loop_detach) {
                        retries--;

                        if (retries <= FINALIZE_CRITICAL_ATTEMPTS) {
                                log_warning("Approaching critical level to finalize filesystem and devices, try to kill all processes.");
                                rescue_send_signal(SIGTERM);
                                rescue_send_signal(SIGKILL);
                        }

                        if (retries > 0)
                                log_info("Action still required, %d tries left", retries);
                        else {
                                log_error("Tried enough but still action required need_umount=%d, need_swapoff=%d, need_loop_detach=%d", need_umount, need_swapoff, need_loop_detach);
                                r = -EBUSY;
                                goto error;
                        }
                }
        }

        sync();

        if (cmd == LINUX_REBOOT_CMD_KEXEC) {
                /* we cheat and exec kexec to avoid doing all its work */
                pid_t pid = fork();
                if (pid < 0) {
                        log_error("Could not fork: %m. Falling back to reboot.");
                        cmd = RB_AUTOBOOT;
                } else if (pid > 0) {
                        waitpid(pid, NULL, 0);
                        log_warning("Failed %s -e -x -f. Falling back to reboot", KEXEC_BINARY_PATH);
                        cmd = RB_AUTOBOOT;
                } else {
                        const char *args[5] = {KEXEC_BINARY_PATH, "-e", "-f", "-x", NULL};
                        execv(args[0], (char * const *) args);
                        return EXIT_FAILURE;
                }
        }

        reboot(cmd);
        r = errno;

  error:
        sync();
        if (r < 0)
                r = -r;
        log_error("Critical error while doing system shutdown: %s", strerror(r));
        freeze();
        return EXIT_FAILURE;
}