main.c revision d677d4df80e0ea1c66c691f50867fedd63c6770a
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering/***
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering This file is part of systemd.
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering Copyright 2010 Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering systemd is free software; you can redistribute it and/or modify it
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering under the terms of the GNU Lesser General Public License as published by
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering the Free Software Foundation; either version 2.1 of the License, or
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering (at your option) any later version.
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering systemd is distributed in the hope that it will be useful, but
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering WITHOUT ANY WARRANTY; without even the implied warranty of
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering Lesser General Public License for more details.
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering You should have received a copy of the GNU Lesser General Public License
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering along with systemd; If not, see <http://www.gnu.org/licenses/>.
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering***/
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <stdio.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <errno.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <string.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <unistd.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <sys/types.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <sys/stat.h>
322345fdb9865ef2477fba8e4bdde0e1183ef505Lennart Poettering#include <getopt.h>
24710c48ed16be5fa461fbb303a744a907541dafLennart Poettering#include <signal.h>
71d35b6b5563817dfbe757ab9e3b9f018b2db491Thomas Hindoe Paaboel Andersen#include <sys/wait.h>
71d35b6b5563817dfbe757ab9e3b9f018b2db491Thomas Hindoe Paaboel Andersen#include <fcntl.h>
623a4c97b9175f95c4b1c6fc34e36c56f1e4ddbfLennart Poettering#include <sys/prctl.h>
39d8db043b599a7382f94bfc904d5e108af438bdLennart Poettering#include <sys/mount.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#ifdef HAVE_VALGRIND_VALGRIND_H
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <valgrind/valgrind.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#endif
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#ifdef HAVE_SECCOMP
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include <seccomp.h>
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#endif
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "sd-daemon.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "sd-messages.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "sd-bus.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "manager.h"
1716f6dcf54d4c181c2e2558e3d5414f54c8d9caLennart Poettering#include "log.h"
0dd25fb9f005d8ab7ac4bc10a609d00569f8c56aLennart Poettering#include "load-fragment.h"
24710c48ed16be5fa461fbb303a744a907541dafLennart Poettering#include "fdset.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "special.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "conf-parser.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "missing.h"
322345fdb9865ef2477fba8e4bdde0e1183ef505Lennart Poettering#include "label.h"
623a4c97b9175f95c4b1c6fc34e36c56f1e4ddbfLennart Poettering#include "pager.h"
322345fdb9865ef2477fba8e4bdde0e1183ef505Lennart Poettering#include "build.h"
1e43061b67336052b5b231840a38508a5397a363Michal Schmidt#include "strv.h"
a407657425a3e47fd2b559cd3bc800f791303f63Lennart Poettering#include "def.h"
a407657425a3e47fd2b559cd3bc800f791303f63Lennart Poettering#include "virt.h"
aea2429d6ec32261dbf6b9caa125fcc6ea9ea76aLennart Poettering#include "architecture.h"
aea2429d6ec32261dbf6b9caa125fcc6ea9ea76aLennart Poettering#include "watchdog.h"
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersen#include "path-util.h"
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersen#include "switch-root.h"
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersen#include "capability.h"
801ad6a6a9cd8fbd58b9f9c27f20dbb3c87d47ddLennart Poettering#include "killall.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "env-util.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "clock-util.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "fileio.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "dbus-manager.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "bus-error.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "bus-util.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "mount-setup.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "loopback-setup.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "hostname-setup.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "machine-id-setup.h"
f9ebb22ab4758bc5bbaaf8eeead74b5b4f81d5c3Lennart Poettering#include "selinux-setup.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "ima-setup.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#include "smack-setup.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#ifdef HAVE_KMOD
0dd25fb9f005d8ab7ac4bc10a609d00569f8c56aLennart Poettering#include "kmod-setup.h"
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering#endif
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poettering
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersenstatic enum {
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersen ACTION_RUN,
9df3ba6c6cb65eecec06f39dfe85a3596cedac4eTom Gundersen ACTION_HELP,
519ef04651b07a547f010d6462603669d7fde4e5Lennart Poettering ACTION_VERSION,
519ef04651b07a547f010d6462603669d7fde4e5Lennart Poettering ACTION_TEST,
519ef04651b07a547f010d6462603669d7fde4e5Lennart Poettering ACTION_DUMP_CONFIGURATION_ITEMS,
ad867662936a4c7ab2c7116d804c272338801231Lennart Poettering ACTION_DONE
51323288fc628a5cac50914df915545d685b793eLennart Poettering} arg_action = ACTION_RUN;
011696f76233486bc56c266b18a328924f70269cLennart Poetteringstatic char *arg_default_unit = NULL;
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poetteringstatic SystemdRunningAs arg_running_as = _SYSTEMD_RUNNING_AS_INVALID;
2c27fbca2d88214bd305272308a370a962818f1eLennart Poetteringstatic bool arg_dump_core = true;
74b2466e14a1961bf3ac0e8a60cfaceec705bd59Lennart Poetteringstatic bool arg_crash_shell = false;
1716f6dcf54d4c181c2e2558e3d5414f54c8d9caLennart Poetteringstatic int arg_crash_chvt = -1;
1716f6dcf54d4c181c2e2558e3d5414f54c8d9caLennart Poetteringstatic bool arg_confirm_spawn = false;
0db4c90afd7d9c7c8884bf8b3ec459edc74a03daDaniel Mackstatic ShowStatus arg_show_status = _SHOW_STATUS_UNSET;
623a4c97b9175f95c4b1c6fc34e36c56f1e4ddbfLennart Poetteringstatic bool arg_switched_root = false;
623a4c97b9175f95c4b1c6fc34e36c56f1e4ddbfLennart Poetteringstatic int arg_no_pager = -1;
ec2c5e4398f9d65e5dfe61530f2556224733d1e6Lennart Poetteringstatic char ***arg_join_controllers = NULL;
f52e61da047d7fc74e83f12dbbf87e0cbcc51c73Lennart Poetteringstatic ExecOutput arg_default_std_output = EXEC_OUTPUT_JOURNAL;
a407657425a3e47fd2b559cd3bc800f791303f63Lennart Poetteringstatic ExecOutput arg_default_std_error = EXEC_OUTPUT_INHERIT;
a407657425a3e47fd2b559cd3bc800f791303f63Lennart Poetteringstatic usec_t arg_default_restart_usec = DEFAULT_RESTART_USEC;
a407657425a3e47fd2b559cd3bc800f791303f63Lennart Poetteringstatic usec_t arg_default_timeout_start_usec = DEFAULT_TIMEOUT_USEC;
4d506d6bb757af3b99e0876234c465e6898c5ea4Lennart Poetteringstatic usec_t arg_default_timeout_stop_usec = DEFAULT_TIMEOUT_USEC;
4d506d6bb757af3b99e0876234c465e6898c5ea4Lennart Poetteringstatic usec_t arg_default_start_limit_interval = DEFAULT_START_LIMIT_INTERVAL;
a51c10485af349eb15faa4d1a63b9818bcf3e589Lennart Poetteringstatic unsigned arg_default_start_limit_burst = DEFAULT_START_LIMIT_BURST;
a51c10485af349eb15faa4d1a63b9818bcf3e589Lennart Poetteringstatic usec_t arg_runtime_watchdog = 0;
801ad6a6a9cd8fbd58b9f9c27f20dbb3c87d47ddLennart Poetteringstatic usec_t arg_shutdown_watchdog = 10 * USEC_PER_MINUTE;
dc477e7385e8ab29efb8fadb72ec994077a105c6Lennart Poetteringstatic char **arg_default_environment = NULL;
edbcc1fdd94355c5cf22263ba2c1cfa4ec2eb010Lennart Poetteringstatic struct rlimit *arg_default_rlimit[_RLIMIT_MAX] = {};
edbcc1fdd94355c5cf22263ba2c1cfa4ec2eb010Lennart Poetteringstatic uint64_t arg_capability_bounding_set_drop = 0;
static nsec_t arg_timer_slack_nsec = NSEC_INFINITY;
static usec_t arg_default_timer_accuracy_usec = 1 * USEC_PER_MINUTE;
static Set* arg_syscall_archs = NULL;
static FILE* arg_serialization = NULL;
static bool arg_default_cpu_accounting = false;
static bool arg_default_blockio_accounting = false;
static bool arg_default_memory_accounting = false;
static usec_t arg_start_timeout_usec = DEFAULT_MANAGER_START_TIMEOUT_USEC;
static FailureAction arg_start_timeout_action = FAILURE_ACTION_POWEROFF_FORCE;
static char *arg_start_timeout_reboot_arg = NULL;
static void nop_handler(int sig) {}
static void pager_open_if_enabled(void) {
if (arg_no_pager <= 0)
return;
pager_open(false);
}
noreturn static void crash(int sig) {
if (getpid() != 1)
/* Pass this on immediately, if this is not PID 1 */
raise(sig);
else if (!arg_dump_core)
log_error("Caught <%s>, not dumping core.", signal_to_string(sig));
else {
struct sigaction sa = {
.sa_handler = nop_handler,
.sa_flags = SA_NOCLDSTOP|SA_RESTART,
};
pid_t pid;
/* We want to wait for the core process, hence let's enable SIGCHLD */
sigaction(SIGCHLD, &sa, NULL);
pid = fork();
if (pid < 0)
log_error("Caught <%s>, cannot fork for core dump: %m", signal_to_string(sig));
else if (pid == 0) {
struct rlimit rl = {};
/* Enable default signal handler for core dump */
zero(sa);
sa.sa_handler = SIG_DFL;
sigaction(sig, &sa, NULL);
/* Don't limit the core dump size */
rl.rlim_cur = RLIM_INFINITY;
rl.rlim_max = RLIM_INFINITY;
setrlimit(RLIMIT_CORE, &rl);
/* Just to be sure... */
chdir("/");
/* Raise the signal again */
raise(sig);
assert_not_reached("We shouldn't be here...");
_exit(1);
} else {
siginfo_t status;
int r;
/* Order things nicely. */
r = wait_for_terminate(pid, &status);
if (r < 0)
log_error("Caught <%s>, waitpid() failed: %s", signal_to_string(sig), strerror(-r));
else if (status.si_code != CLD_DUMPED)
log_error("Caught <%s>, core dump failed.", signal_to_string(sig));
else
log_error("Caught <%s>, dumped core as pid "PID_FMT".", signal_to_string(sig), pid);
}
}
if (arg_crash_chvt)
chvt(arg_crash_chvt);
if (arg_crash_shell) {
struct sigaction sa = {
.sa_handler = SIG_IGN,
.sa_flags = SA_NOCLDSTOP|SA_NOCLDWAIT|SA_RESTART,
};
pid_t pid;
log_info("Executing crash shell in 10s...");
sleep(10);
/* Let the kernel reap children for us */
assert_se(sigaction(SIGCHLD, &sa, NULL) == 0);
pid = fork();
if (pid < 0)
log_error("Failed to fork off crash shell: %m");
else if (pid == 0) {
make_console_stdio();
execl("/bin/sh", "/bin/sh", NULL);
log_error("execl() failed: %m");
_exit(1);
}
log_info("Successfully spawned crash shell as pid "PID_FMT".", pid);
}
log_info("Freezing execution.");
freeze();
}
static void install_crash_handler(void) {
struct sigaction sa = {
.sa_handler = crash,
.sa_flags = SA_NODEFER,
};
sigaction_many(&sa, SIGNALS_CRASH_HANDLER, -1);
}
static int console_setup(void) {
_cleanup_close_ int tty_fd = -1;
int r;
tty_fd = open_terminal("/dev/console", O_WRONLY|O_NOCTTY|O_CLOEXEC);
if (tty_fd < 0) {
log_error("Failed to open /dev/console: %s", strerror(-tty_fd));
return tty_fd;
}
/* We don't want to force text mode. plymouth may be showing
* pictures already from initrd. */
r = reset_terminal_fd(tty_fd, false);
if (r < 0) {
log_error("Failed to reset /dev/console: %s", strerror(-r));
return r;
}
return 0;
}
static int set_default_unit(const char *u) {
char *c;
assert(u);
c = strdup(u);
if (!c)
return -ENOMEM;
free(arg_default_unit);
arg_default_unit = c;
return 0;
}
static int parse_proc_cmdline_item(const char *key, const char *value) {
static const char * const rlmap[] = {
"emergency", SPECIAL_EMERGENCY_TARGET,
"-b", SPECIAL_EMERGENCY_TARGET,
"rescue", SPECIAL_RESCUE_TARGET,
"single", SPECIAL_RESCUE_TARGET,
"-s", SPECIAL_RESCUE_TARGET,
"s", SPECIAL_RESCUE_TARGET,
"S", SPECIAL_RESCUE_TARGET,
"1", SPECIAL_RESCUE_TARGET,
"2", SPECIAL_RUNLEVEL2_TARGET,
"3", SPECIAL_RUNLEVEL3_TARGET,
"4", SPECIAL_RUNLEVEL4_TARGET,
"5", SPECIAL_RUNLEVEL5_TARGET,
};
int r;
assert(key);
if (streq(key, "systemd.unit") && value) {
if (!in_initrd())
return set_default_unit(value);
} else if (streq(key, "rd.systemd.unit") && value) {
if (in_initrd())
return set_default_unit(value);
} else if (streq(key, "systemd.dump_core") && value) {
r = parse_boolean(value);
if (r < 0)
log_warning("Failed to parse dump core switch %s. Ignoring.", value);
else
arg_dump_core = r;
} else if (streq(key, "systemd.crash_shell") && value) {
r = parse_boolean(value);
if (r < 0)
log_warning("Failed to parse crash shell switch %s. Ignoring.", value);
else
arg_crash_shell = r;
} else if (streq(key, "systemd.crash_chvt") && value) {
if (safe_atoi(value, &r) < 0)
log_warning("Failed to parse crash chvt switch %s. Ignoring.", value);
else
arg_crash_chvt = r;
} else if (streq(key, "systemd.confirm_spawn") && value) {
r = parse_boolean(value);
if (r < 0)
log_warning("Failed to parse confirm spawn switch %s. Ignoring.", value);
else
arg_confirm_spawn = r;
} else if (streq(key, "systemd.show_status") && value) {
r = parse_show_status(value, &arg_show_status);
if (r < 0)
log_warning("Failed to parse show status switch %s. Ignoring.", value);
} else if (streq(key, "systemd.default_standard_output") && value) {
r = exec_output_from_string(value);
if (r < 0)
log_warning("Failed to parse default standard output switch %s. Ignoring.", value);
else
arg_default_std_output = r;
} else if (streq(key, "systemd.default_standard_error") && value) {
r = exec_output_from_string(value);
if (r < 0)
log_warning("Failed to parse default standard error switch %s. Ignoring.", value);
else
arg_default_std_error = r;
} else if (streq(key, "systemd.setenv") && value) {
if (env_assignment_is_valid(value)) {
char **env;
env = strv_env_set(arg_default_environment, value);
if (env)
arg_default_environment = env;
else
log_warning("Setting environment variable '%s' failed, ignoring: %s", value, strerror(ENOMEM));
} else
log_warning("Environment variable name '%s' is not valid. Ignoring.", value);
} else if (streq(key, "quiet") && !value) {
log_set_max_level(LOG_NOTICE);
if (arg_show_status == _SHOW_STATUS_UNSET)
arg_show_status = SHOW_STATUS_AUTO;
} else if (streq(key, "debug") && !value) {
/* Note that log_parse_environment() handles 'debug'
* too, and sets the log level to LOG_DEBUG. */
if (detect_container(NULL) > 0)
log_set_target(LOG_TARGET_CONSOLE);
} else if (!in_initrd() && !value) {
unsigned i;
/* SysV compatibility */
for (i = 0; i < ELEMENTSOF(rlmap); i += 2)
if (streq(key, rlmap[i]))
return set_default_unit(rlmap[i+1]);
}
return 0;
}
#define DEFINE_SETTER(name, func, descr) \
static int name(const char *unit, \
const char *filename, \
unsigned line, \
const char *section, \
unsigned section_line, \
const char *lvalue, \
int ltype, \
const char *rvalue, \
void *data, \
void *userdata) { \
\
int r; \
\
assert(filename); \
assert(lvalue); \
assert(rvalue); \
\
r = func(rvalue); \
if (r < 0) \
log_syntax(unit, LOG_ERR, filename, line, -r, \
"Invalid " descr "'%s': %s", \
rvalue, strerror(-r)); \
\
return 0; \
}
DEFINE_SETTER(config_parse_level2, log_set_max_level_from_string, "log level")
DEFINE_SETTER(config_parse_target, log_set_target_from_string, "target")
DEFINE_SETTER(config_parse_color, log_show_color_from_string, "color" )
DEFINE_SETTER(config_parse_location, log_show_location_from_string, "location")
static int config_parse_cpu_affinity2(
const char *unit,
const char *filename,
unsigned line,
const char *section,
unsigned section_line,
const char *lvalue,
int ltype,
const char *rvalue,
void *data,
void *userdata) {
const char *word, *state;
size_t l;
cpu_set_t *c = NULL;
unsigned ncpus = 0;
assert(filename);
assert(lvalue);
assert(rvalue);
FOREACH_WORD_QUOTED(word, l, rvalue, state) {
char *t;
int r;
unsigned cpu;
if (!(t = strndup(word, l)))
return log_oom();
r = safe_atou(t, &cpu);
free(t);
if (!c)
if (!(c = cpu_set_malloc(&ncpus)))
return log_oom();
if (r < 0 || cpu >= ncpus) {
log_syntax(unit, LOG_ERR, filename, line, -r,
"Failed to parse CPU affinity '%s'", rvalue);
CPU_FREE(c);
return -EBADMSG;
}
CPU_SET_S(cpu, CPU_ALLOC_SIZE(ncpus), c);
}
if (!isempty(state))
log_syntax(unit, LOG_ERR, filename, line, EINVAL,
"Trailing garbage, ignoring.");
if (c) {
if (sched_setaffinity(0, CPU_ALLOC_SIZE(ncpus), c) < 0)
log_warning_unit(unit, "Failed to set CPU affinity: %m");
CPU_FREE(c);
}
return 0;
}
static int config_parse_show_status(
const char* unit,
const char *filename,
unsigned line,
const char *section,
unsigned section_line,
const char *lvalue,
int ltype,
const char *rvalue,
void *data,
void *userdata) {
int k;
ShowStatus *b = data;
assert(filename);
assert(lvalue);
assert(rvalue);
assert(data);
k = parse_show_status(rvalue, b);
if (k < 0) {
log_syntax(unit, LOG_ERR, filename, line, -k,
"Failed to parse show status setting, ignoring: %s", rvalue);
return 0;
}
return 0;
}
static void strv_free_free(char ***l) {
char ***i;
if (!l)
return;
for (i = l; *i; i++)
strv_free(*i);
free(l);
}
static void free_join_controllers(void) {
strv_free_free(arg_join_controllers);
arg_join_controllers = NULL;
}
static int config_parse_join_controllers(const char *unit,
const char *filename,
unsigned line,
const char *section,
unsigned section_line,
const char *lvalue,
int ltype,
const char *rvalue,
void *data,
void *userdata) {
unsigned n = 0;
const char *word, *state;
size_t length;
assert(filename);
assert(lvalue);
assert(rvalue);
free_join_controllers();
FOREACH_WORD_QUOTED(word, length, rvalue, state) {
char *s, **l;
s = strndup(word, length);
if (!s)
return log_oom();
l = strv_split(s, ",");
free(s);
strv_uniq(l);
if (strv_length(l) <= 1) {
strv_free(l);
continue;
}
if (!arg_join_controllers) {
arg_join_controllers = new(char**, 2);
if (!arg_join_controllers) {
strv_free(l);
return log_oom();
}
arg_join_controllers[0] = l;
arg_join_controllers[1] = NULL;
n = 1;
} else {
char ***a;
char ***t;
t = new0(char**, n+2);
if (!t) {
strv_free(l);
return log_oom();
}
n = 0;
for (a = arg_join_controllers; *a; a++) {
if (strv_overlap(*a, l)) {
if (strv_extend_strv(&l, *a) < 0) {
strv_free(l);
strv_free_free(t);
return log_oom();
}
} else {
char **c;
c = strv_copy(*a);
if (!c) {
strv_free(l);
strv_free_free(t);
return log_oom();
}
t[n++] = c;
}
}
t[n++] = strv_uniq(l);
strv_free_free(arg_join_controllers);
arg_join_controllers = t;
}
}
if (!isempty(state))
log_syntax(unit, LOG_ERR, filename, line, EINVAL,
"Trailing garbage, ignoring.");
return 0;
}
static int parse_config_file(void) {
const ConfigTableItem items[] = {
{ "Manager", "LogLevel", config_parse_level2, 0, NULL },
{ "Manager", "LogTarget", config_parse_target, 0, NULL },
{ "Manager", "LogColor", config_parse_color, 0, NULL },
{ "Manager", "LogLocation", config_parse_location, 0, NULL },
{ "Manager", "DumpCore", config_parse_bool, 0, &arg_dump_core },
{ "Manager", "CrashShell", config_parse_bool, 0, &arg_crash_shell },
{ "Manager", "ShowStatus", config_parse_show_status, 0, &arg_show_status },
{ "Manager", "CrashChVT", config_parse_int, 0, &arg_crash_chvt },
{ "Manager", "CPUAffinity", config_parse_cpu_affinity2, 0, NULL },
{ "Manager", "JoinControllers", config_parse_join_controllers, 0, &arg_join_controllers },
{ "Manager", "RuntimeWatchdogSec", config_parse_sec, 0, &arg_runtime_watchdog },
{ "Manager", "ShutdownWatchdogSec", config_parse_sec, 0, &arg_shutdown_watchdog },
{ "Manager", "CapabilityBoundingSet", config_parse_bounding_set, 0, &arg_capability_bounding_set_drop },
#ifdef HAVE_SECCOMP
{ "Manager", "SystemCallArchitectures", config_parse_syscall_archs, 0, &arg_syscall_archs },
#endif
{ "Manager", "TimerSlackNSec", config_parse_nsec, 0, &arg_timer_slack_nsec },
{ "Manager", "DefaultTimerAccuracySec", config_parse_sec, 0, &arg_default_timer_accuracy_usec },
{ "Manager", "DefaultStandardOutput", config_parse_output, 0, &arg_default_std_output },
{ "Manager", "DefaultStandardError", config_parse_output, 0, &arg_default_std_error },
{ "Manager", "DefaultTimeoutStartSec", config_parse_sec, 0, &arg_default_timeout_start_usec },
{ "Manager", "DefaultTimeoutStopSec", config_parse_sec, 0, &arg_default_timeout_stop_usec },
{ "Manager", "DefaultRestartSec", config_parse_sec, 0, &arg_default_restart_usec },
{ "Manager", "DefaultStartLimitInterval", config_parse_sec, 0, &arg_default_start_limit_interval },
{ "Manager", "DefaultStartLimitBurst", config_parse_unsigned, 0, &arg_default_start_limit_burst },
{ "Manager", "DefaultEnvironment", config_parse_environ, 0, &arg_default_environment },
{ "Manager", "DefaultLimitCPU", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_CPU] },
{ "Manager", "DefaultLimitFSIZE", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_FSIZE] },
{ "Manager", "DefaultLimitDATA", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_DATA] },
{ "Manager", "DefaultLimitSTACK", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_STACK] },
{ "Manager", "DefaultLimitCORE", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_CORE] },
{ "Manager", "DefaultLimitRSS", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_RSS] },
{ "Manager", "DefaultLimitNOFILE", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_NOFILE] },
{ "Manager", "DefaultLimitAS", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_AS] },
{ "Manager", "DefaultLimitNPROC", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_NPROC] },
{ "Manager", "DefaultLimitMEMLOCK", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_MEMLOCK] },
{ "Manager", "DefaultLimitLOCKS", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_LOCKS] },
{ "Manager", "DefaultLimitSIGPENDING", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_SIGPENDING] },
{ "Manager", "DefaultLimitMSGQUEUE", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_MSGQUEUE] },
{ "Manager", "DefaultLimitNICE", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_NICE] },
{ "Manager", "DefaultLimitRTPRIO", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_RTPRIO] },
{ "Manager", "DefaultLimitRTTIME", config_parse_limit, 0, &arg_default_rlimit[RLIMIT_RTTIME] },
{ "Manager", "DefaultCPUAccounting", config_parse_bool, 0, &arg_default_cpu_accounting },
{ "Manager", "DefaultBlockIOAccounting", config_parse_bool, 0, &arg_default_blockio_accounting },
{ "Manager", "DefaultMemoryAccounting", config_parse_bool, 0, &arg_default_memory_accounting },
{ "Manager", "StartTimeoutSec", config_parse_sec, 0, &arg_start_timeout_usec },
{ "Manager", "StartTimeoutAction", config_parse_failure_action, 0, &arg_start_timeout_action },
{ "Manager", "StartTimeoutRebootArgument",config_parse_string, 0, &arg_start_timeout_reboot_arg },
{}
};
const char *fn;
fn = arg_running_as == SYSTEMD_SYSTEM ? PKGSYSCONFDIR "/system.conf" : PKGSYSCONFDIR "/user.conf";
config_parse(NULL, fn, NULL,
"Manager\0",
config_item_table_lookup, items,
false, false, true, NULL);
return 0;
}
static int parse_argv(int argc, char *argv[]) {
enum {
ARG_LOG_LEVEL = 0x100,
ARG_LOG_TARGET,
ARG_LOG_COLOR,
ARG_LOG_LOCATION,
ARG_UNIT,
ARG_SYSTEM,
ARG_USER,
ARG_TEST,
ARG_NO_PAGER,
ARG_VERSION,
ARG_DUMP_CONFIGURATION_ITEMS,
ARG_DUMP_CORE,
ARG_CRASH_SHELL,
ARG_CONFIRM_SPAWN,
ARG_SHOW_STATUS,
ARG_DESERIALIZE,
ARG_SWITCHED_ROOT,
ARG_DEFAULT_STD_OUTPUT,
ARG_DEFAULT_STD_ERROR
};
static const struct option options[] = {
{ "log-level", required_argument, NULL, ARG_LOG_LEVEL },
{ "log-target", required_argument, NULL, ARG_LOG_TARGET },
{ "log-color", optional_argument, NULL, ARG_LOG_COLOR },
{ "log-location", optional_argument, NULL, ARG_LOG_LOCATION },
{ "unit", required_argument, NULL, ARG_UNIT },
{ "system", no_argument, NULL, ARG_SYSTEM },
{ "user", no_argument, NULL, ARG_USER },
{ "test", no_argument, NULL, ARG_TEST },
{ "no-pager", no_argument, NULL, ARG_NO_PAGER },
{ "help", no_argument, NULL, 'h' },
{ "version", no_argument, NULL, ARG_VERSION },
{ "dump-configuration-items", no_argument, NULL, ARG_DUMP_CONFIGURATION_ITEMS },
{ "dump-core", optional_argument, NULL, ARG_DUMP_CORE },
{ "crash-shell", optional_argument, NULL, ARG_CRASH_SHELL },
{ "confirm-spawn", optional_argument, NULL, ARG_CONFIRM_SPAWN },
{ "show-status", optional_argument, NULL, ARG_SHOW_STATUS },
{ "deserialize", required_argument, NULL, ARG_DESERIALIZE },
{ "switched-root", no_argument, NULL, ARG_SWITCHED_ROOT },
{ "default-standard-output", required_argument, NULL, ARG_DEFAULT_STD_OUTPUT, },
{ "default-standard-error", required_argument, NULL, ARG_DEFAULT_STD_ERROR, },
{}
};
int c, r;
assert(argc >= 1);
assert(argv);
if (getpid() == 1)
opterr = 0;
while ((c = getopt_long(argc, argv, "hDbsz:", options, NULL)) >= 0)
switch (c) {
case ARG_LOG_LEVEL:
r = log_set_max_level_from_string(optarg);
if (r < 0) {
log_error("Failed to parse log level %s.", optarg);
return r;
}
break;
case ARG_LOG_TARGET:
r = log_set_target_from_string(optarg);
if (r < 0) {
log_error("Failed to parse log target %s.", optarg);
return r;
}
break;
case ARG_LOG_COLOR:
if (optarg) {
r = log_show_color_from_string(optarg);
if (r < 0) {
log_error("Failed to parse log color setting %s.", optarg);
return r;
}
} else
log_show_color(true);
break;
case ARG_LOG_LOCATION:
if (optarg) {
r = log_show_location_from_string(optarg);
if (r < 0) {
log_error("Failed to parse log location setting %s.", optarg);
return r;
}
} else
log_show_location(true);
break;
case ARG_DEFAULT_STD_OUTPUT:
r = exec_output_from_string(optarg);
if (r < 0) {
log_error("Failed to parse default standard output setting %s.", optarg);
return r;
} else
arg_default_std_output = r;
break;
case ARG_DEFAULT_STD_ERROR:
r = exec_output_from_string(optarg);
if (r < 0) {
log_error("Failed to parse default standard error output setting %s.", optarg);
return r;
} else
arg_default_std_error = r;
break;
case ARG_UNIT:
r = set_default_unit(optarg);
if (r < 0) {
log_error("Failed to set default unit %s: %s", optarg, strerror(-r));
return r;
}
break;
case ARG_SYSTEM:
arg_running_as = SYSTEMD_SYSTEM;
break;
case ARG_USER:
arg_running_as = SYSTEMD_USER;
break;
case ARG_TEST:
arg_action = ACTION_TEST;
if (arg_no_pager < 0)
arg_no_pager = true;
break;
case ARG_NO_PAGER:
arg_no_pager = true;
break;
case ARG_VERSION:
arg_action = ACTION_VERSION;
break;
case ARG_DUMP_CONFIGURATION_ITEMS:
arg_action = ACTION_DUMP_CONFIGURATION_ITEMS;
break;
case ARG_DUMP_CORE:
r = optarg ? parse_boolean(optarg) : 1;
if (r < 0) {
log_error("Failed to parse dump core boolean %s.", optarg);
return r;
}
arg_dump_core = r;
break;
case ARG_CRASH_SHELL:
r = optarg ? parse_boolean(optarg) : 1;
if (r < 0) {
log_error("Failed to parse crash shell boolean %s.", optarg);
return r;
}
arg_crash_shell = r;
break;
case ARG_CONFIRM_SPAWN:
r = optarg ? parse_boolean(optarg) : 1;
if (r < 0) {
log_error("Failed to parse confirm spawn boolean %s.", optarg);
return r;
}
arg_confirm_spawn = r;
break;
case ARG_SHOW_STATUS:
if (optarg) {
r = parse_show_status(optarg, &arg_show_status);
if (r < 0) {
log_error("Failed to parse show status boolean %s.", optarg);
return r;
}
} else
arg_show_status = SHOW_STATUS_YES;
break;
case ARG_DESERIALIZE: {
int fd;
FILE *f;
r = safe_atoi(optarg, &fd);
if (r < 0 || fd < 0) {
log_error("Failed to parse deserialize option %s.", optarg);
return r < 0 ? r : -EINVAL;
}
fd_cloexec(fd, true);
f = fdopen(fd, "r");
if (!f) {
log_error("Failed to open serialization fd: %m");
return -errno;
}
if (arg_serialization)
fclose(arg_serialization);
arg_serialization = f;
break;
}
case ARG_SWITCHED_ROOT:
arg_switched_root = true;
break;
case 'h':
arg_action = ACTION_HELP;
if (arg_no_pager < 0)
arg_no_pager = true;
break;
case 'D':
log_set_max_level(LOG_DEBUG);
break;
case 'b':
case 's':
case 'z':
/* Just to eat away the sysvinit kernel
* cmdline args without getopt() error
* messages that we'll parse in
* parse_proc_cmdline_word() or ignore. */
case '?':
if (getpid() != 1)
return -EINVAL;
else
return 0;
default:
assert_not_reached("Unhandled option code.");
}
if (optind < argc && getpid() != 1) {
/* Hmm, when we aren't run as init system
* let's complain about excess arguments */
log_error("Excess arguments.");
return -EINVAL;
}
return 0;
}
static int help(void) {
printf("%s [OPTIONS...]\n\n"
"Starts up and maintains the system or user services.\n\n"
" -h --help Show this help\n"
" --test Determine startup sequence, dump it and exit\n"
" --no-pager Do not pipe output into a pager\n"
" --dump-configuration-items Dump understood unit configuration items\n"
" --unit=UNIT Set default unit\n"
" --system Run a system instance, even if PID != 1\n"
" --user Run a user instance\n"
" --dump-core[=0|1] Dump core on crash\n"
" --crash-shell[=0|1] Run shell on crash\n"
" --confirm-spawn[=0|1] Ask for confirmation when spawning processes\n"
" --show-status[=0|1] Show status updates on the console during bootup\n"
" --log-target=TARGET Set log target (console, journal, kmsg, journal-or-kmsg, null)\n"
" --log-level=LEVEL Set log level (debug, info, notice, warning, err, crit, alert, emerg)\n"
" --log-color[=0|1] Highlight important log messages\n"
" --log-location[=0|1] Include code location in log messages\n"
" --default-standard-output= Set default standard output for services\n"
" --default-standard-error= Set default standard error output for services\n",
program_invocation_short_name);
return 0;
}
static int version(void) {
puts(PACKAGE_STRING);
puts(SYSTEMD_FEATURES);
return 0;
}
static int prepare_reexecute(Manager *m, FILE **_f, FDSet **_fds, bool switching_root) {
FILE *f = NULL;
FDSet *fds = NULL;
int r;
assert(m);
assert(_f);
assert(_fds);
r = manager_open_serialization(m, &f);
if (r < 0) {
log_error("Failed to create serialization file: %s", strerror(-r));
goto fail;
}
/* Make sure nothing is really destructed when we shut down */
m->n_reloading ++;
bus_manager_send_reloading(m, true);
fds = fdset_new();
if (!fds) {
r = -ENOMEM;
log_error("Failed to allocate fd set: %s", strerror(-r));
goto fail;
}
r = manager_serialize(m, f, fds, switching_root);
if (r < 0) {
log_error("Failed to serialize state: %s", strerror(-r));
goto fail;
}
if (fseeko(f, 0, SEEK_SET) < 0) {
log_error("Failed to rewind serialization fd: %m");
goto fail;
}
r = fd_cloexec(fileno(f), false);
if (r < 0) {
log_error("Failed to disable O_CLOEXEC for serialization: %s", strerror(-r));
goto fail;
}
r = fdset_cloexec(fds, false);
if (r < 0) {
log_error("Failed to disable O_CLOEXEC for serialization fds: %s", strerror(-r));
goto fail;
}
*_f = f;
*_fds = fds;
return 0;
fail:
fdset_free(fds);
if (f)
fclose(f);
return r;
}
static int bump_rlimit_nofile(struct rlimit *saved_rlimit) {
struct rlimit nl;
int r;
assert(saved_rlimit);
/* Save the original RLIMIT_NOFILE so that we can reset it
* later when transitioning from the initrd to the main
* systemd or suchlike. */
if (getrlimit(RLIMIT_NOFILE, saved_rlimit) < 0) {
log_error("Reading RLIMIT_NOFILE failed: %m");
return -errno;
}
/* Make sure forked processes get the default kernel setting */
if (!arg_default_rlimit[RLIMIT_NOFILE]) {
struct rlimit *rl;
rl = newdup(struct rlimit, saved_rlimit, 1);
if (!rl)
return log_oom();
arg_default_rlimit[RLIMIT_NOFILE] = rl;
}
/* Bump up the resource limit for ourselves substantially */
nl.rlim_cur = nl.rlim_max = 64*1024;
r = setrlimit_closest(RLIMIT_NOFILE, &nl);
if (r < 0) {
log_error("Setting RLIMIT_NOFILE failed: %s", strerror(-r));
return r;
}
return 0;
}
static void test_mtab(void) {
static const char ok[] =
"/proc/self/mounts\0"
"/proc/mounts\0"
"../proc/self/mounts\0"
"../proc/mounts\0";
_cleanup_free_ char *p = NULL;
int r;
/* Check that /etc/mtab is a symlink to the right place or
* non-existing. But certainly not a file, or a symlink to
* some weird place... */
r = readlink_malloc("/etc/mtab", &p);
if (r == -ENOENT)
return;
if (r >= 0 && nulstr_contains(ok, p))
return;
log_warning("/etc/mtab is not a symlink or not pointing to /proc/self/mounts. "
"This is not supported anymore. "
"Please make sure to replace this file by a symlink to avoid incorrect or misleading mount(8) output.");
}
static void test_usr(void) {
/* Check that /usr is not a separate fs */
if (dir_is_empty("/usr") <= 0)
return;
log_warning("/usr appears to be on its own filesytem and is not already mounted. This is not a supported setup. "
"Some things will probably break (sometimes even silently) in mysterious ways. "
"Consult http://freedesktop.org/wiki/Software/systemd/separate-usr-is-broken for more information.");
}
static int initialize_join_controllers(void) {
/* By default, mount "cpu" + "cpuacct" together, and "net_cls"
* + "net_prio". We'd like to add "cpuset" to the mix, but
* "cpuset" does't really work for groups with no initialized
* attributes. */
arg_join_controllers = new(char**, 3);
if (!arg_join_controllers)
return -ENOMEM;
arg_join_controllers[0] = strv_new("cpu", "cpuacct", NULL);
arg_join_controllers[1] = strv_new("net_cls", "net_prio", NULL);
arg_join_controllers[2] = NULL;
if (!arg_join_controllers[0] || !arg_join_controllers[1]) {
free_join_controllers();
return -ENOMEM;
}
return 0;
}
static int enforce_syscall_archs(Set *archs) {
#ifdef HAVE_SECCOMP
scmp_filter_ctx *seccomp;
Iterator i;
void *id;
int r;
seccomp = seccomp_init(SCMP_ACT_ALLOW);
if (!seccomp)
return log_oom();
SET_FOREACH(id, arg_syscall_archs, i) {
r = seccomp_arch_add(seccomp, PTR_TO_UINT32(id) - 1);
if (r == -EEXIST)
continue;
if (r < 0) {
log_error("Failed to add architecture to seccomp: %s", strerror(-r));
goto finish;
}
}
r = seccomp_attr_set(seccomp, SCMP_FLTATR_CTL_NNP, 0);
if (r < 0) {
log_error("Failed to unset NO_NEW_PRIVS: %s", strerror(-r));
goto finish;
}
r = seccomp_load(seccomp);
if (r < 0)
log_error("Failed to add install architecture seccomp: %s", strerror(-r));
finish:
seccomp_release(seccomp);
return r;
#else
return 0;
#endif
}
static int status_welcome(void) {
_cleanup_free_ char *pretty_name = NULL, *ansi_color = NULL;
int r;
r = parse_env_file("/etc/os-release", NEWLINE,
"PRETTY_NAME", &pretty_name,
"ANSI_COLOR", &ansi_color,
NULL);
if (r == -ENOENT) {
r = parse_env_file("/usr/lib/os-release", NEWLINE,
"PRETTY_NAME", &pretty_name,
"ANSI_COLOR", &ansi_color,
NULL);
}
if (r < 0 && r != -ENOENT)
log_warning("Failed to read os-release file: %s", strerror(-r));
return status_printf(NULL, false, false,
"\nWelcome to \x1B[%sm%s\x1B[0m!\n",
isempty(ansi_color) ? "1" : ansi_color,
isempty(pretty_name) ? "Linux" : pretty_name);
}
static int write_container_id(void) {
const char *c;
c = getenv("container");
if (isempty(c))
return 0;
return write_string_file("/run/systemd/container", c);
}
int main(int argc, char *argv[]) {
Manager *m = NULL;
int r, retval = EXIT_FAILURE;
usec_t before_startup, after_startup;
char timespan[FORMAT_TIMESPAN_MAX];
FDSet *fds = NULL;
bool reexecute = false;
const char *shutdown_verb = NULL;
dual_timestamp initrd_timestamp = { 0ULL, 0ULL };
dual_timestamp userspace_timestamp = { 0ULL, 0ULL };
dual_timestamp kernel_timestamp = { 0ULL, 0ULL };
dual_timestamp security_start_timestamp = { 0ULL, 0ULL };
dual_timestamp security_finish_timestamp = { 0ULL, 0ULL };
static char systemd[] = "systemd";
bool skip_setup = false;
unsigned j;
bool loaded_policy = false;
bool arm_reboot_watchdog = false;
bool queue_default_job = false;
bool empty_etc = false;
char *switch_root_dir = NULL, *switch_root_init = NULL;
static struct rlimit saved_rlimit_nofile = { 0, 0 };
#ifdef HAVE_SYSV_COMPAT
if (getpid() != 1 && strstr(program_invocation_short_name, "init")) {
/* This is compatibility support for SysV, where
* calling init as a user is identical to telinit. */
errno = -ENOENT;
execv(SYSTEMCTL_BINARY_PATH, argv);
log_error("Failed to exec " SYSTEMCTL_BINARY_PATH ": %m");
return 1;
}
#endif
dual_timestamp_from_monotonic(&kernel_timestamp, 0);
dual_timestamp_get(&userspace_timestamp);
/* Determine if this is a reexecution or normal bootup. We do
* the full command line parsing much later, so let's just
* have a quick peek here. */
if (strv_find(argv+1, "--deserialize"))
skip_setup = true;
/* If we have switched root, do all the special setup
* things */
if (strv_find(argv+1, "--switched-root"))
skip_setup = false;
/* If we get started via the /sbin/init symlink then we are
called 'init'. After a subsequent reexecution we are then
called 'systemd'. That is confusing, hence let's call us
systemd right-away. */
program_invocation_short_name = systemd;
prctl(PR_SET_NAME, systemd);
saved_argv = argv;
saved_argc = argc;
log_show_color(isatty(STDERR_FILENO) > 0);
log_set_upgrade_syslog_to_journal(true);
/* Disable the umask logic */
if (getpid() == 1)
umask(0);
if (getpid() == 1 && detect_container(NULL) <= 0) {
/* Running outside of a container as PID 1 */
arg_running_as = SYSTEMD_SYSTEM;
make_null_stdio();
log_set_target(LOG_TARGET_KMSG);
log_open();
if (in_initrd())
initrd_timestamp = userspace_timestamp;
if (!skip_setup) {
mount_setup_early();
dual_timestamp_get(&security_start_timestamp);
if (selinux_setup(&loaded_policy) < 0)
goto finish;
if (ima_setup() < 0)
goto finish;
if (smack_setup(&loaded_policy) < 0)
goto finish;
dual_timestamp_get(&security_finish_timestamp);
}
if (label_init(NULL) < 0)
goto finish;
if (!skip_setup) {
if (clock_is_localtime() > 0) {
int min;
/*
* The very first call of settimeofday() also does a time warp in the kernel.
*
* In the rtc-in-local time mode, we set the kernel's timezone, and rely on
* external tools to take care of maintaining the RTC and do all adjustments.
* This matches the behavior of Windows, which leaves the RTC alone if the
* registry tells that the RTC runs in UTC.
*/
r = clock_set_timezone(&min);
if (r < 0)
log_error("Failed to apply local time delta, ignoring: %s", strerror(-r));
else
log_info("RTC configured in localtime, applying delta of %i minutes to system time.", min);
} else if (!in_initrd()) {
/*
* Do a dummy very first call to seal the kernel's time warp magic.
*
* Do not call this this from inside the initrd. The initrd might not
* carry /etc/adjtime with LOCAL, but the real system could be set up
* that way. In such case, we need to delay the time-warp or the sealing
* until we reach the real system.
*
* Do no set the kernel's timezone. The concept of local time cannot
* be supported reliably, the time will jump or be incorrect at every daylight
* saving time change. All kernel local time concepts will be treated
* as UTC that way.
*/
clock_reset_timewarp();
}
}
/* Set the default for later on, but don't actually
* open the logs like this for now. Note that if we
* are transitioning from the initrd there might still
* be journal fd open, and we shouldn't attempt
* opening that before we parsed /proc/cmdline which
* might redirect output elsewhere. */
log_set_target(LOG_TARGET_JOURNAL_OR_KMSG);
} else if (getpid() == 1) {
/* Running inside a container, as PID 1 */
arg_running_as = SYSTEMD_SYSTEM;
log_set_target(LOG_TARGET_CONSOLE);
log_close_console(); /* force reopen of /dev/console */
log_open();
/* For the later on, see above... */
log_set_target(LOG_TARGET_JOURNAL);
/* clear the kernel timestamp,
* because we are in a container */
kernel_timestamp.monotonic = 0ULL;
kernel_timestamp.realtime = 0ULL;
} else {
/* Running as user instance */
arg_running_as = SYSTEMD_USER;
log_set_target(LOG_TARGET_AUTO);
log_open();
/* clear the kernel timestamp,
* because we are not PID 1 */
kernel_timestamp.monotonic = 0ULL;
kernel_timestamp.realtime = 0ULL;
}
/* Initialize default unit */
r = set_default_unit(SPECIAL_DEFAULT_TARGET);
if (r < 0) {
log_error("Failed to set default unit %s: %s", SPECIAL_DEFAULT_TARGET, strerror(-r));
goto finish;
}
r = initialize_join_controllers();
if (r < 0)
goto finish;
/* Mount /proc, /sys and friends, so that /proc/cmdline and
* /proc/$PID/fd is available. */
if (getpid() == 1) {
r = mount_setup(loaded_policy);
if (r < 0)
goto finish;
}
/* Reset all signal handlers. */
assert_se(reset_all_signal_handlers() == 0);
ignore_signals(SIGNALS_IGNORE, -1);
if (parse_config_file() < 0)
goto finish;
if (arg_running_as == SYSTEMD_SYSTEM)
if (parse_proc_cmdline(parse_proc_cmdline_item) < 0)
goto finish;
/* Note that this also parses bits from the kernel command
* line, including "debug". */
log_parse_environment();
if (parse_argv(argc, argv) < 0)
goto finish;
if (arg_action == ACTION_TEST &&
geteuid() == 0) {
log_error("Don't run test mode as root.");
goto finish;
}
if (arg_running_as == SYSTEMD_USER &&
arg_action == ACTION_RUN &&
sd_booted() <= 0) {
log_error("Trying to run as user instance, but the system has not been booted with systemd.");
goto finish;
}
if (arg_running_as == SYSTEMD_SYSTEM &&
arg_action == ACTION_RUN &&
running_in_chroot() > 0) {
log_error("Cannot be run in a chroot() environment.");
goto finish;
}
if (arg_action == ACTION_TEST)
skip_setup = true;
pager_open_if_enabled();
if (arg_action == ACTION_HELP) {
retval = help();
goto finish;
} else if (arg_action == ACTION_VERSION) {
retval = version();
goto finish;
} else if (arg_action == ACTION_DUMP_CONFIGURATION_ITEMS) {
unit_dump_config_items(stdout);
retval = EXIT_SUCCESS;
goto finish;
} else if (arg_action == ACTION_DONE) {
retval = EXIT_SUCCESS;
goto finish;
}
if (arg_running_as == SYSTEMD_USER &&
!getenv("XDG_RUNTIME_DIR")) {
log_error("Trying to run as user instance, but $XDG_RUNTIME_DIR is not set.");
goto finish;
}
assert_se(arg_action == ACTION_RUN || arg_action == ACTION_TEST);
/* Close logging fds, in order not to confuse fdset below */
log_close();
/* Remember open file descriptors for later deserialization */
r = fdset_new_fill(&fds);
if (r < 0) {
log_error("Failed to allocate fd set: %s", strerror(-r));
goto finish;
} else
fdset_cloexec(fds, true);
if (arg_serialization)
assert_se(fdset_remove(fds, fileno(arg_serialization)) >= 0);
if (arg_running_as == SYSTEMD_SYSTEM)
/* Become a session leader if we aren't one yet. */
setsid();
/* Move out of the way, so that we won't block unmounts */
assert_se(chdir("/") == 0);
/* Reset the console, but only if this is really init and we
* are freshly booted */
if (arg_running_as == SYSTEMD_SYSTEM && arg_action == ACTION_RUN) {
/* If we are init, we connect stdin/stdout/stderr to
* /dev/null and make sure we don't have a controlling
* tty. */
release_terminal();
if (getpid() == 1 && !skip_setup)
console_setup();
}
/* Open the logging devices, if possible and necessary */
log_open();
if (arg_show_status == _SHOW_STATUS_UNSET)
arg_show_status = SHOW_STATUS_YES;
/* Make sure we leave a core dump without panicing the
* kernel. */
if (getpid() == 1) {
install_crash_handler();
r = mount_cgroup_controllers(arg_join_controllers);
if (r < 0)
goto finish;
}
if (arg_running_as == SYSTEMD_SYSTEM) {
const char *virtualization = NULL;
log_info(PACKAGE_STRING " running in %ssystem mode. (" SYSTEMD_FEATURES ")",
arg_action == ACTION_TEST ? "test " : "" );
detect_virtualization(&virtualization);
if (virtualization)
log_info("Detected virtualization '%s'.", virtualization);
write_container_id();
log_info("Detected architecture '%s'.", architecture_to_string(uname_architecture()));
if (in_initrd())
log_info("Running in initial RAM disk.");
/* Let's check whether /etc is already populated. We
* don't actually really check for that, but use
* /etc/machine-id as flag file. This allows container
* managers and installers to provision a couple of
* files already. If the container manager wants to
* provision the machine ID itself it should pass
* $container_uuid to PID 1.*/
empty_etc = access("/etc/machine-id", F_OK) < 0;
if (empty_etc)
log_info("Running with unpopulated /etc.");
} else {
_cleanup_free_ char *t;
t = uid_to_name(getuid());
log_debug(PACKAGE_STRING " running in %suser mode for user "UID_FMT"/%s. (" SYSTEMD_FEATURES ")",
arg_action == ACTION_TEST ? " test" : "", getuid(), t);
}
if (arg_running_as == SYSTEMD_SYSTEM && !skip_setup) {
if (arg_show_status > 0 || plymouth_running())
status_welcome();
#ifdef HAVE_KMOD
kmod_setup();
#endif
hostname_setup();
machine_id_setup(NULL);
loopback_setup();
test_mtab();
test_usr();
}
if (arg_running_as == SYSTEMD_SYSTEM && arg_runtime_watchdog > 0)
watchdog_set_timeout(&arg_runtime_watchdog);
if (arg_timer_slack_nsec != NSEC_INFINITY)
if (prctl(PR_SET_TIMERSLACK, arg_timer_slack_nsec) < 0)
log_error("Failed to adjust timer slack: %m");
if (arg_capability_bounding_set_drop) {
r = capability_bounding_set_drop_usermode(arg_capability_bounding_set_drop);
if (r < 0) {
log_error("Failed to drop capability bounding set of usermode helpers: %s", strerror(-r));
goto finish;
}
r = capability_bounding_set_drop(arg_capability_bounding_set_drop, true);
if (r < 0) {
log_error("Failed to drop capability bounding set: %s", strerror(-r));
goto finish;
}
}
if (arg_syscall_archs) {
r = enforce_syscall_archs(arg_syscall_archs);
if (r < 0)
goto finish;
}
if (arg_running_as == SYSTEMD_USER) {
/* Become reaper of our children */
if (prctl(PR_SET_CHILD_SUBREAPER, 1) < 0) {
log_warning("Failed to make us a subreaper: %m");
if (errno == EINVAL)
log_info("Perhaps the kernel version is too old (< 3.4?)");
}
}
if (arg_running_as == SYSTEMD_SYSTEM) {
bump_rlimit_nofile(&saved_rlimit_nofile);
if (empty_etc) {
r = unit_file_preset_all(UNIT_FILE_SYSTEM, false, NULL, UNIT_FILE_PRESET_FULL, false, NULL, 0);
if (r < 0)
log_warning("Failed to populate /etc with preset unit settings, ignoring: %s", strerror(-r));
else
log_info("Populated /etc with preset unit settings.");
}
}
r = manager_new(arg_running_as, arg_action == ACTION_TEST, &m);
if (r < 0) {
log_error("Failed to allocate manager object: %s", strerror(-r));
goto finish;
}
m->confirm_spawn = arg_confirm_spawn;
m->default_timer_accuracy_usec = arg_default_timer_accuracy_usec;
m->default_std_output = arg_default_std_output;
m->default_std_error = arg_default_std_error;
m->default_restart_usec = arg_default_restart_usec;
m->default_timeout_start_usec = arg_default_timeout_start_usec;
m->default_timeout_stop_usec = arg_default_timeout_stop_usec;
m->default_start_limit_interval = arg_default_start_limit_interval;
m->default_start_limit_burst = arg_default_start_limit_burst;
m->default_cpu_accounting = arg_default_cpu_accounting;
m->default_blockio_accounting = arg_default_blockio_accounting;
m->default_memory_accounting = arg_default_memory_accounting;
m->runtime_watchdog = arg_runtime_watchdog;
m->shutdown_watchdog = arg_shutdown_watchdog;
m->start_timeout_usec = arg_start_timeout_usec;
m->start_timeout_action = arg_start_timeout_action;
free_and_strdup(&m->start_timeout_reboot_arg, arg_start_timeout_reboot_arg);
m->userspace_timestamp = userspace_timestamp;
m->kernel_timestamp = kernel_timestamp;
m->initrd_timestamp = initrd_timestamp;
m->security_start_timestamp = security_start_timestamp;
m->security_finish_timestamp = security_finish_timestamp;
manager_set_default_rlimits(m, arg_default_rlimit);
manager_environment_add(m, NULL, arg_default_environment);
manager_set_show_status(m, arg_show_status);
manager_set_first_boot(m, empty_etc);
/* Remember whether we should queue the default job */
queue_default_job = !arg_serialization || arg_switched_root;
before_startup = now(CLOCK_MONOTONIC);
r = manager_startup(m, arg_serialization, fds);
if (r < 0)
log_error("Failed to fully start up daemon: %s", strerror(-r));
/* This will close all file descriptors that were opened, but
* not claimed by any unit. */
fdset_free(fds);
fds = NULL;
if (arg_serialization) {
fclose(arg_serialization);
arg_serialization = NULL;
}
if (queue_default_job) {
_cleanup_bus_error_free_ sd_bus_error error = SD_BUS_ERROR_NULL;
Unit *target = NULL;
Job *default_unit_job;
log_debug("Activating default unit: %s", arg_default_unit);
r = manager_load_unit(m, arg_default_unit, NULL, &error, &target);
if (r < 0)
log_error("Failed to load default target: %s", bus_error_message(&error, r));
else if (target->load_state == UNIT_ERROR || target->load_state == UNIT_NOT_FOUND)
log_error("Failed to load default target: %s", strerror(-target->load_error));
else if (target->load_state == UNIT_MASKED)
log_error("Default target masked.");
if (!target || target->load_state != UNIT_LOADED) {
log_info("Trying to load rescue target...");
r = manager_load_unit(m, SPECIAL_RESCUE_TARGET, NULL, &error, &target);
if (r < 0) {
log_error("Failed to load rescue target: %s", bus_error_message(&error, r));
goto finish;
} else if (target->load_state == UNIT_ERROR || target->load_state == UNIT_NOT_FOUND) {
log_error("Failed to load rescue target: %s", strerror(-target->load_error));
goto finish;
} else if (target->load_state == UNIT_MASKED) {
log_error("Rescue target masked.");
goto finish;
}
}
assert(target->load_state == UNIT_LOADED);
if (arg_action == ACTION_TEST) {
printf("-> By units:\n");
manager_dump_units(m, stdout, "\t");
}
r = manager_add_job(m, JOB_START, target, JOB_ISOLATE, false, &error, &default_unit_job);
if (r == -EPERM) {
log_debug("Default target could not be isolated, starting instead: %s", bus_error_message(&error, r));
r = manager_add_job(m, JOB_START, target, JOB_REPLACE, false, &error, &default_unit_job);
if (r < 0) {
log_error("Failed to start default target: %s", bus_error_message(&error, r));
goto finish;
}
} else if (r < 0) {
log_error("Failed to isolate default target: %s", bus_error_message(&error, r));
goto finish;
}
m->default_unit_job_id = default_unit_job->id;
after_startup = now(CLOCK_MONOTONIC);
log_full(arg_action == ACTION_TEST ? LOG_INFO : LOG_DEBUG,
"Loaded units and determined initial transaction in %s.",
format_timespan(timespan, sizeof(timespan), after_startup - before_startup, 100 * USEC_PER_MSEC));
if (arg_action == ACTION_TEST) {
printf("-> By jobs:\n");
manager_dump_jobs(m, stdout, "\t");
retval = EXIT_SUCCESS;
goto finish;
}
}
for (;;) {
r = manager_loop(m);
if (r < 0) {
log_error("Failed to run mainloop: %s", strerror(-r));
goto finish;
}
switch (m->exit_code) {
case MANAGER_EXIT:
retval = EXIT_SUCCESS;
log_debug("Exit.");
goto finish;
case MANAGER_RELOAD:
log_info("Reloading.");
r = manager_reload(m);
if (r < 0)
log_error("Failed to reload: %s", strerror(-r));
break;
case MANAGER_REEXECUTE:
if (prepare_reexecute(m, &arg_serialization, &fds, false) < 0)
goto finish;
reexecute = true;
log_notice("Reexecuting.");
goto finish;
case MANAGER_SWITCH_ROOT:
/* Steal the switch root parameters */
switch_root_dir = m->switch_root;
switch_root_init = m->switch_root_init;
m->switch_root = m->switch_root_init = NULL;
if (!switch_root_init)
if (prepare_reexecute(m, &arg_serialization, &fds, true) < 0)
goto finish;
reexecute = true;
log_notice("Switching root.");
goto finish;
case MANAGER_REBOOT:
case MANAGER_POWEROFF:
case MANAGER_HALT:
case MANAGER_KEXEC: {
static const char * const table[_MANAGER_EXIT_CODE_MAX] = {
[MANAGER_REBOOT] = "reboot",
[MANAGER_POWEROFF] = "poweroff",
[MANAGER_HALT] = "halt",
[MANAGER_KEXEC] = "kexec"
};
assert_se(shutdown_verb = table[m->exit_code]);
arm_reboot_watchdog = m->exit_code == MANAGER_REBOOT;
log_notice("Shutting down.");
goto finish;
}
default:
assert_not_reached("Unknown exit code.");
}
}
finish:
pager_close();
if (m) {
manager_free(m);
m = NULL;
}
for (j = 0; j < ELEMENTSOF(arg_default_rlimit); j++) {
free(arg_default_rlimit[j]);
arg_default_rlimit[j] = NULL;
}
free(arg_default_unit);
arg_default_unit = NULL;
free_join_controllers();
strv_free(arg_default_environment);
arg_default_environment = NULL;
set_free(arg_syscall_archs);
arg_syscall_archs = NULL;
free(arg_start_timeout_reboot_arg);
arg_start_timeout_reboot_arg = NULL;
label_finish();
if (reexecute) {
const char **args;
unsigned i, args_size;
/* Close and disarm the watchdog, so that the new
* instance can reinitialize it, but doesn't get
* rebooted while we do that */
watchdog_close(true);
/* Reset the RLIMIT_NOFILE to the kernel default, so
* that the new systemd can pass the kernel default to
* its child processes */
if (saved_rlimit_nofile.rlim_cur > 0)
setrlimit(RLIMIT_NOFILE, &saved_rlimit_nofile);
if (switch_root_dir) {
/* Kill all remaining processes from the
* initrd, but don't wait for them, so that we
* can handle the SIGCHLD for them after
* deserializing. */
broadcast_signal(SIGTERM, false, true);
/* And switch root with MS_MOVE, because we remove the old directory afterwards and detach it. */
r = switch_root(switch_root_dir, "/mnt", true, MS_MOVE);
if (r < 0)
log_error("Failed to switch root, trying to continue: %s", strerror(-r));
}
args_size = MAX(6, argc+1);
args = newa(const char*, args_size);
if (!switch_root_init) {
char sfd[16];
/* First try to spawn ourselves with the right
* path, and with full serialization. We do
* this only if the user didn't specify an
* explicit init to spawn. */
assert(arg_serialization);
assert(fds);
snprintf(sfd, sizeof(sfd), "%i", fileno(arg_serialization));
char_array_0(sfd);
i = 0;
args[i++] = SYSTEMD_BINARY_PATH;
if (switch_root_dir)
args[i++] = "--switched-root";
args[i++] = arg_running_as == SYSTEMD_SYSTEM ? "--system" : "--user";
args[i++] = "--deserialize";
args[i++] = sfd;
args[i++] = NULL;
/* do not pass along the environment we inherit from the kernel or initrd */
if (switch_root_dir)
clearenv();
assert(i <= args_size);
execv(args[0], (char* const*) args);
}
/* Try the fallback, if there is any, without any
* serialization. We pass the original argv[] and
* envp[]. (Well, modulo the ordering changes due to
* getopt() in argv[], and some cleanups in envp[],
* but let's hope that doesn't matter.) */
if (arg_serialization) {
fclose(arg_serialization);
arg_serialization = NULL;
}
if (fds) {
fdset_free(fds);
fds = NULL;
}
/* Reopen the console */
make_console_stdio();
for (j = 1, i = 1; j < (unsigned) argc; j++)
args[i++] = argv[j];
args[i++] = NULL;
assert(i <= args_size);
/* Reenable any blocked signals, especially important
* if we switch from initial ramdisk to init=... */
reset_all_signal_handlers();
reset_signal_mask();
if (switch_root_init) {
args[0] = switch_root_init;
execv(args[0], (char* const*) args);
log_warning("Failed to execute configured init, trying fallback: %m");
}
args[0] = "/sbin/init";
execv(args[0], (char* const*) args);
if (errno == ENOENT) {
log_warning("No /sbin/init, trying fallback");
args[0] = "/bin/sh";
args[1] = NULL;
execv(args[0], (char* const*) args);
log_error("Failed to execute /bin/sh, giving up: %m");
} else
log_warning("Failed to execute /sbin/init, giving up: %m");
}
if (arg_serialization) {
fclose(arg_serialization);
arg_serialization = NULL;
}
if (fds) {
fdset_free(fds);
fds = NULL;
}
#ifdef HAVE_VALGRIND_VALGRIND_H
/* If we are PID 1 and running under valgrind, then let's exit
* here explicitly. valgrind will only generate nice output on
* exit(), not on exec(), hence let's do the former not the
* latter here. */
if (getpid() == 1 && RUNNING_ON_VALGRIND)
return 0;
#endif
if (shutdown_verb) {
char log_level[DECIMAL_STR_MAX(int) + 1];
const char* command_line[9] = {
SYSTEMD_SHUTDOWN_BINARY_PATH,
shutdown_verb,
"--log-level", log_level,
"--log-target",
};
unsigned pos = 5;
_cleanup_strv_free_ char **env_block = NULL;
assert(command_line[pos] == NULL);
env_block = strv_copy(environ);
snprintf(log_level, sizeof(log_level), "%d", log_get_max_level());
switch (log_get_target()) {
case LOG_TARGET_KMSG:
case LOG_TARGET_JOURNAL_OR_KMSG:
case LOG_TARGET_SYSLOG_OR_KMSG:
command_line[pos++] = "kmsg";
break;
case LOG_TARGET_CONSOLE:
default:
command_line[pos++] = "console";
break;
};
if (log_get_show_color())
command_line[pos++] = "--log-color";
if (log_get_show_location())
command_line[pos++] = "--log-location";
assert(pos < ELEMENTSOF(command_line));
if (arm_reboot_watchdog && arg_shutdown_watchdog > 0) {
char *e;
/* If we reboot let's set the shutdown
* watchdog and tell the shutdown binary to
* repeatedly ping it */
watchdog_set_timeout(&arg_shutdown_watchdog);
watchdog_close(false);
/* Tell the binary how often to ping, ignore failure */
if (asprintf(&e, "WATCHDOG_USEC="USEC_FMT, arg_shutdown_watchdog) > 0)
strv_push(&env_block, e);
} else
watchdog_close(true);
/* Avoid the creation of new processes forked by the
* kernel; at this point, we will not listen to the
* signals anyway */
if (detect_container(NULL) <= 0)
cg_uninstall_release_agent(SYSTEMD_CGROUP_CONTROLLER);
execve(SYSTEMD_SHUTDOWN_BINARY_PATH, (char **) command_line, env_block);
log_error("Failed to execute shutdown binary, %s: %m",
getpid() == 1 ? "freezing" : "quitting");
}
if (getpid() == 1)
freeze();
return retval;
}