2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <stdio.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <errno.h>

8eea868708923a092ee85d6146ba4c04b7baea06Zbigniew Jędrzejewski-Szmek#include <string.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <unistd.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <sys/stat.h>

8eea868708923a092ee85d6146ba4c04b7baea06Zbigniew Jędrzejewski-Szmek#include <getopt.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <signal.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <fcntl.h>

8eea868708923a092ee85d6146ba4c04b7baea06Zbigniew Jędrzejewski-Szmek#include <sys/prctl.h>

2968644080fd103062f070e83edd620e0a58c44dZbigniew Jędrzejewski-Szmek#include <sys/mount.h>

#ifdef HAVE_VALGRIND_VALGRIND_H

#include <valgrind/valgrind.h>

#endif

#ifdef HAVE_SECCOMP

#include <seccomp.h>

#endif

#include "sd-daemon.h"

#include "sd-bus.h"

#include "log.h"

#include "fdset.h"

#include "special.h"

#include "conf-parser.h"

#include "missing.h"

#include "pager.h"

#include "build.h"

#include "strv.h"

#include "def.h"

#include "virt.h"

#include "architecture.h"

#include "watchdog.h"

#include "switch-root.h"

#include "capability.h"

#include "killall.h"

#include "env-util.h"

#include "clock-util.h"

#include "fileio.h"

#include "bus-error.h"

#include "bus-util.h"

#include "selinux-util.h"

#include "manager.h"

#include "dbus-manager.h"

#include "load-fragment.h"

#include "mount-setup.h"

#include "loopback-setup.h"

#include "hostname-setup.h"

#include "machine-id-setup.h"

#include "selinux-setup.h"

#include "ima-setup.h"

#include "smack-setup.h"

#include "kmod-setup.h"

static enum {

ACTION_RUN,

ACTION_HELP,

ACTION_VERSION,

ACTION_TEST,

ACTION_DUMP_CONFIGURATION_ITEMS,

ACTION_DONE

} arg_action = ACTION_RUN;

static char *arg_default_unit = NULL;

static SystemdRunningAs arg_running_as = _SYSTEMD_RUNNING_AS_INVALID;

static bool arg_dump_core = true;

static bool arg_crash_shell = false;

static int arg_crash_chvt = -1;

static bool arg_confirm_spawn = false;

static ShowStatus arg_show_status = _SHOW_STATUS_UNSET;

static bool arg_switched_root = false;

static int arg_no_pager = -1;

static char ***arg_join_controllers = NULL;

static ExecOutput arg_default_std_output = EXEC_OUTPUT_JOURNAL;

static ExecOutput arg_default_std_error = EXEC_OUTPUT_INHERIT;

static usec_t arg_default_restart_usec = DEFAULT_RESTART_USEC;

static usec_t arg_default_timeout_start_usec = DEFAULT_TIMEOUT_USEC;

static usec_t arg_default_timeout_stop_usec = DEFAULT_TIMEOUT_USEC;

static usec_t arg_default_start_limit_interval = DEFAULT_START_LIMIT_INTERVAL;

static unsigned arg_default_start_limit_burst = DEFAULT_START_LIMIT_BURST;

static usec_t arg_runtime_watchdog = 0;

static usec_t arg_shutdown_watchdog = 10 * USEC_PER_MINUTE;

static char **arg_default_environment = NULL;

static struct rlimit *arg_default_rlimit[_RLIMIT_MAX] = {};

static 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 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_emergency("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 = raw_clone(SIGCHLD, NULL);

if (pid < 0)

log_emergency_errno(errno, "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... */

(void) chdir("/");

/* Raise the signal again */

pid = raw_getpid();

kill(pid, sig); /* raise() would kill the parent */

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_emergency_errno(r, "Caught <%s>, waitpid() failed: %m", signal_to_string(sig));

else if (status.si_code != CLD_DUMPED)

log_emergency("Caught <%s>, core dump failed (child "PID_FMT", code=%s, status=%i/%s).",

signal_to_string(sig),

pid, sigchld_code_to_string(status.si_code),

status.si_status,

strna(status.si_code == CLD_EXITED

? exit_status_to_string(status.si_status, EXIT_STATUS_FULL)

: signal_to_string(status.si_status)));

else

log_emergency("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 = raw_clone(SIGCHLD, NULL);

if (pid < 0)

log_emergency_errno(errno, "Failed to fork off crash shell: %m");

else if (pid == 0) {

make_console_stdio();

execle("/bin/sh", "/bin/sh", NULL, environ);

log_emergency_errno(errno, "execle() failed: %m");

_exit(1);

} else

log_info("Successfully spawned crash shell as PID "PID_FMT".", pid);

}

log_emergency("Freezing execution.");

freeze();

}

static void install_crash_handler(void) {

static const struct sigaction sa = {

.sa_handler = crash,

.sa_flags = SA_NODEFER, /* So that we can raise the signal again from the signal handler */

};

int r;

/* We ignore the return value here, since, we don't mind if we

r = sigaction_many(&sa, SIGNALS_CRASH_HANDLER, -1);

if (r < 0)

log_debug_errno(r, "I had trouble setting up the crash handler, ignoring: %m");

}

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)

return log_error_errno(tty_fd, "Failed to open /dev/console: %m");

/* We don't want to force text mode. plymouth may be showing

r = reset_terminal_fd(tty_fd, false);

if (r < 0)

return log_error_errno(r, "Failed to reset /dev/console: %m");

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_MULTI_USER_TARGET,

"3", SPECIAL_MULTI_USER_TARGET,

"4", SPECIAL_MULTI_USER_TARGET,

"5", SPECIAL_GRAPHICAL_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_errno(ENOMEM, "Setting environment variable '%s' failed, ignoring: %m", value);

} else

log_warning("Environment variable name '%s' is not valid. Ignoring.", value);

} else if (streq(key, "quiet") && !value) {

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'

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_unit_warning(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 },

{}

};

const char *fn, *conf_dirs_nulstr;

fn = arg_running_as == SYSTEMD_SYSTEM ? PKGSYSCONFDIR "/system.conf" : PKGSYSCONFDIR "/user.conf";

conf_dirs_nulstr = arg_running_as == SYSTEMD_SYSTEM ? CONF_DIRS_NULSTR("systemd/system.conf") : CONF_DIRS_NULSTR("systemd/user.conf");

config_parse_many(fn, conf_dirs_nulstr, "Manager\0",

config_item_table_lookup, items, false, 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)

return log_error_errno(r, "Failed to set default unit %s: %m", optarg);

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)

return log_error_errno(errno, "Failed to open serialization fd: %m");

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

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

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_errno(r, "Failed to create serialization file: %m");

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_errno(r, "Failed to allocate fd set: %m");

goto fail;

}

r = manager_serialize(m, f, fds, switching_root);

if (r < 0) {

log_error_errno(r, "Failed to serialize state: %m");

goto fail;

}

if (fseeko(f, 0, SEEK_SET) < 0) {

log_error_errno(errno, "Failed to rewind serialization fd: %m");

goto fail;

}

r = fd_cloexec(fileno(f), false);

if (r < 0) {

log_error_errno(r, "Failed to disable O_CLOEXEC for serialization: %m");

goto fail;

}

r = fdset_cloexec(fds, false);

if (r < 0) {

log_error_errno(r, "Failed to disable O_CLOEXEC for serialization fds: %m");

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

if (getrlimit(RLIMIT_NOFILE, saved_rlimit) < 0)

return log_error_errno(errno, "Reading RLIMIT_NOFILE failed: %m");

/* 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)

return log_error_errno(r, "Setting RLIMIT_NOFILE failed: %m");

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

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 filesystem 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"

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_errno(r, "Failed to add architecture to seccomp: %m");

goto finish;

}

}

r = seccomp_attr_set(seccomp, SCMP_FLTATR_CTL_NNP, 0);

if (r < 0) {

log_error_errno(r, "Failed to unset NO_NEW_PRIVS: %m");

goto finish;

}

r = seccomp_load(seccomp);

if (r < 0)

log_error_errno(r, "Failed to add install architecture seccomp: %m");

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_errno(r, "Failed to read os-release file: %m");

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 = DUAL_TIMESTAMP_NULL;

dual_timestamp userspace_timestamp = DUAL_TIMESTAMP_NULL;

dual_timestamp kernel_timestamp = DUAL_TIMESTAMP_NULL;

dual_timestamp security_start_timestamp = DUAL_TIMESTAMP_NULL;

dual_timestamp security_finish_timestamp = DUAL_TIMESTAMP_NULL;

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;

struct rlimit saved_rlimit_nofile = RLIMIT_MAKE_CONST(0);

const char *error_message = NULL;

#ifdef HAVE_SYSV_COMPAT

if (getpid() != 1 && strstr(program_invocation_short_name, "init")) {

/* This is compatibility support for SysV, where

errno = -ENOENT;

execv(SYSTEMCTL_BINARY_PATH, argv);

log_error_errno(errno, "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

if (strv_find(argv+1, "--deserialize"))

skip_setup = true;

/* If we have switched root, do all the special setup

if (strv_find(argv+1, "--switched-root"))

skip_setup = false;

/* If we get started via the /sbin/init symlink then we are

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 (mac_selinux_setup(&loaded_policy) < 0) {

error_message = "Failed to load SELinux policy";

goto finish;

} else if (ima_setup() < 0) {

error_message = "Failed to load IMA policy";

goto finish;

} else if (mac_smack_setup(&loaded_policy) < 0) {

error_message = "Failed to load SMACK policy";

goto finish;

}

dual_timestamp_get(&security_finish_timestamp);

}

if (mac_selinux_init(NULL) < 0) {

error_message = "Failed to initialize SELinux policy";

goto finish;

}

if (!skip_setup) {

if (clock_is_localtime() > 0) {

int min;

/*

r = clock_set_timezone(&min);

if (r < 0)

log_error_errno(r, "Failed to apply local time delta, ignoring: %m");

else

log_info("RTC configured in localtime, applying delta of %i minutes to system time.", min);

} else if (!in_initrd()) {

/*

clock_reset_timewarp();

}

}

/* Set the default for later on, but don't actually

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,

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,