machine-dbus.c revision ecb0573db18142bed8115584ca251c001c31427e
/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
/***
This file is part of systemd.
Copyright 2011 Lennart Poettering
systemd is free software; you can redistribute it and/or modify it
under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 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
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with systemd; If not, see <http://www.gnu.org/licenses/>.
***/
#include <errno.h>
#include <string.h>
#include <sys/mount.h>
/* When we include libgen.h because we need dirname() we immediately
* undefine basename() since libgen.h defines it as a macro to the POSIX
* version which is really broken. We prefer GNU basename(). */
#include <libgen.h>
#undef basename
#include "alloc-util.h"
#include "bus-common-errors.h"
#include "bus-internal.h"
#include "bus-label.h"
#include "bus-util.h"
#include "copy.h"
#include "env-util.h"
#include "fd-util.h"
#include "fileio.h"
#include "formats-util.h"
#include "fs-util.h"
#include "in-addr-util.h"
#include "local-addresses.h"
#include "machine-dbus.h"
#include "machine.h"
#include "mkdir.h"
#include "path-util.h"
#include "process-util.h"
#include "strv.h"
#include "terminal-util.h"
#include "user-util.h"
static int property_get_id(
sd_bus *bus,
const char *path,
const char *interface,
const char *property,
sd_bus_message *reply,
void *userdata,
sd_bus_error *error) {
Machine *m = userdata;
assert(bus);
assert(reply);
assert(m);
return sd_bus_message_append_array(reply, 'y', &m->id, 16);
}
static int property_get_state(
sd_bus *bus,
const char *path,
const char *interface,
const char *property,
sd_bus_message *reply,
void *userdata,
sd_bus_error *error) {
Machine *m = userdata;
const char *state;
int r;
assert(bus);
assert(reply);
assert(m);
state = machine_state_to_string(machine_get_state(m));
r = sd_bus_message_append_basic(reply, 's', state);
if (r < 0)
return r;
return 1;
}
static int property_get_netif(
sd_bus *bus,
const char *path,
const char *interface,
const char *property,
sd_bus_message *reply,
void *userdata,
sd_bus_error *error) {
Machine *m = userdata;
assert(bus);
assert(reply);
assert(m);
assert_cc(sizeof(int) == sizeof(int32_t));
return sd_bus_message_append_array(reply, 'i', m->netif, m->n_netif * sizeof(int));
}
static BUS_DEFINE_PROPERTY_GET_ENUM(property_get_class, machine_class, MachineClass);
int bus_machine_method_terminate(sd_bus_message *message, void *userdata, sd_bus_error *error) {
Machine *m = userdata;
int r;
assert(message);
assert(m);
r = bus_verify_polkit_async(
message,
CAP_KILL,
"org.freedesktop.machine1.manage-machines",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
r = machine_stop(m);
if (r < 0)
return r;
return sd_bus_reply_method_return(message, NULL);
}
int bus_machine_method_kill(sd_bus_message *message, void *userdata, sd_bus_error *error) {
Machine *m = userdata;
const char *swho;
int32_t signo;
KillWho who;
int r;
assert(message);
assert(m);
r = sd_bus_message_read(message, "si", &swho, &signo);
if (r < 0)
return r;
if (isempty(swho))
who = KILL_ALL;
else {
who = kill_who_from_string(swho);
if (who < 0)
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Invalid kill parameter '%s'", swho);
}
if (signo <= 0 || signo >= _NSIG)
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Invalid signal %i", signo);
r = bus_verify_polkit_async(
message,
CAP_KILL,
"org.freedesktop.machine1.manage-machines",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
r = machine_kill(m, who, signo);
if (r < 0)
return r;
return sd_bus_reply_method_return(message, NULL);
}
int bus_machine_method_get_addresses(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *reply = NULL;
Machine *m = userdata;
int r;
assert(message);
assert(m);
r = sd_bus_message_new_method_return(message, &reply);
if (r < 0)
return r;
r = sd_bus_message_open_container(reply, 'a', "(iay)");
if (r < 0)
return r;
switch (m->class) {
case MACHINE_HOST: {
_cleanup_free_ struct local_address *addresses = NULL;
struct local_address *a;
int n, i;
n = local_addresses(NULL, 0, AF_UNSPEC, &addresses);
if (n < 0)
return n;
for (a = addresses, i = 0; i < n; a++, i++) {
r = sd_bus_message_open_container(reply, 'r', "iay");
if (r < 0)
return r;
r = sd_bus_message_append(reply, "i", addresses[i].family);
if (r < 0)
return r;
r = sd_bus_message_append_array(reply, 'y', &addresses[i].address, FAMILY_ADDRESS_SIZE(addresses[i].family));
if (r < 0)
return r;
r = sd_bus_message_close_container(reply);
if (r < 0)
return r;
}
break;
}
case MACHINE_CONTAINER: {
_cleanup_close_pair_ int pair[2] = { -1, -1 };
_cleanup_free_ char *us = NULL, *them = NULL;
_cleanup_close_ int netns_fd = -1;
const char *p;
siginfo_t si;
pid_t child;
r = readlink_malloc("/proc/self/ns/net", &us);
if (r < 0)
return r;
p = procfs_file_alloca(m->leader, "ns/net");
r = readlink_malloc(p, &them);
if (r < 0)
return r;
if (streq(us, them))
return sd_bus_error_setf(error, BUS_ERROR_NO_PRIVATE_NETWORKING, "Machine %s does not use private networking", m->name);
r = namespace_open(m->leader, NULL, NULL, &netns_fd, NULL, NULL);
if (r < 0)
return r;
if (socketpair(AF_UNIX, SOCK_SEQPACKET, 0, pair) < 0)
return -errno;
child = fork();
if (child < 0)
return sd_bus_error_set_errnof(error, errno, "Failed to fork(): %m");
if (child == 0) {
_cleanup_free_ struct local_address *addresses = NULL;
struct local_address *a;
int i, n;
pair[0] = safe_close(pair[0]);
r = namespace_enter(-1, -1, netns_fd, -1, -1);
if (r < 0)
_exit(EXIT_FAILURE);
n = local_addresses(NULL, 0, AF_UNSPEC, &addresses);
if (n < 0)
_exit(EXIT_FAILURE);
for (a = addresses, i = 0; i < n; a++, i++) {
struct iovec iov[2] = {
{ .iov_base = &a->family, .iov_len = sizeof(a->family) },
{ .iov_base = &a->address, .iov_len = FAMILY_ADDRESS_SIZE(a->family) },
};
r = writev(pair[1], iov, 2);
if (r < 0)
_exit(EXIT_FAILURE);
}
pair[1] = safe_close(pair[1]);
_exit(EXIT_SUCCESS);
}
pair[1] = safe_close(pair[1]);
for (;;) {
int family;
ssize_t n;
union in_addr_union in_addr;
struct iovec iov[2];
struct msghdr mh = {
.msg_iov = iov,
.msg_iovlen = 2,
};
iov[0] = (struct iovec) { .iov_base = &family, .iov_len = sizeof(family) };
iov[1] = (struct iovec) { .iov_base = &in_addr, .iov_len = sizeof(in_addr) };
n = recvmsg(pair[0], &mh, 0);
if (n < 0)
return -errno;
if ((size_t) n < sizeof(family))
break;
r = sd_bus_message_open_container(reply, 'r', "iay");
if (r < 0)
return r;
r = sd_bus_message_append(reply, "i", family);
if (r < 0)
return r;
switch (family) {
case AF_INET:
if (n != sizeof(struct in_addr) + sizeof(family))
return -EIO;
r = sd_bus_message_append_array(reply, 'y', &in_addr.in, sizeof(in_addr.in));
break;
case AF_INET6:
if (n != sizeof(struct in6_addr) + sizeof(family))
return -EIO;
r = sd_bus_message_append_array(reply, 'y', &in_addr.in6, sizeof(in_addr.in6));
break;
}
if (r < 0)
return r;
r = sd_bus_message_close_container(reply);
if (r < 0)
return r;
}
r = wait_for_terminate(child, &si);
if (r < 0)
return sd_bus_error_set_errnof(error, r, "Failed to wait for child: %m");
if (si.si_code != CLD_EXITED || si.si_status != EXIT_SUCCESS)
return sd_bus_error_setf(error, SD_BUS_ERROR_FAILED, "Child died abnormally.");
break;
}
default:
return sd_bus_error_setf(error, SD_BUS_ERROR_NOT_SUPPORTED, "Requesting IP address data is only supported on container machines.");
}
r = sd_bus_message_close_container(reply);
if (r < 0)
return r;
return sd_bus_send(NULL, reply, NULL);
}
int bus_machine_method_get_os_release(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *reply = NULL;
_cleanup_strv_free_ char **l = NULL;
Machine *m = userdata;
char **k, **v;
int r;
assert(message);
assert(m);
switch (m->class) {
case MACHINE_HOST:
r = load_env_file_pairs(NULL, "/etc/os-release", NULL, &l);
if (r < 0)
return r;
break;
case MACHINE_CONTAINER: {
_cleanup_close_ int mntns_fd = -1, root_fd = -1;
_cleanup_close_pair_ int pair[2] = { -1, -1 };
_cleanup_fclose_ FILE *f = NULL;
siginfo_t si;
pid_t child;
r = namespace_open(m->leader, NULL, &mntns_fd, NULL, NULL, &root_fd);
if (r < 0)
return r;
if (socketpair(AF_UNIX, SOCK_SEQPACKET, 0, pair) < 0)
return -errno;
child = fork();
if (child < 0)
return sd_bus_error_set_errnof(error, errno, "Failed to fork(): %m");
if (child == 0) {
_cleanup_close_ int fd = -1;
pair[0] = safe_close(pair[0]);
r = namespace_enter(-1, mntns_fd, -1, -1, root_fd);
if (r < 0)
_exit(EXIT_FAILURE);
fd = open("/etc/os-release", O_RDONLY|O_CLOEXEC);
if (fd < 0) {
fd = open("/usr/lib/os-release", O_RDONLY|O_CLOEXEC);
if (fd < 0)
_exit(EXIT_FAILURE);
}
r = copy_bytes(fd, pair[1], (uint64_t) -1, false);
if (r < 0)
_exit(EXIT_FAILURE);
_exit(EXIT_SUCCESS);
}
pair[1] = safe_close(pair[1]);
f = fdopen(pair[0], "re");
if (!f)
return -errno;
pair[0] = -1;
r = load_env_file_pairs(f, "/etc/os-release", NULL, &l);
if (r < 0)
return r;
r = wait_for_terminate(child, &si);
if (r < 0)
return sd_bus_error_set_errnof(error, r, "Failed to wait for child: %m");
if (si.si_code != CLD_EXITED || si.si_status != EXIT_SUCCESS)
return sd_bus_error_setf(error, SD_BUS_ERROR_FAILED, "Child died abnormally.");
break;
}
default:
return sd_bus_error_setf(error, SD_BUS_ERROR_NOT_SUPPORTED, "Requesting OS release data is only supported on container machines.");
}
r = sd_bus_message_new_method_return(message, &reply);
if (r < 0)
return r;
r = sd_bus_message_open_container(reply, 'a', "{ss}");
if (r < 0)
return r;
STRV_FOREACH_PAIR(k, v, l) {
r = sd_bus_message_append(reply, "{ss}", *k, *v);
if (r < 0)
return r;
}
r = sd_bus_message_close_container(reply);
if (r < 0)
return r;
return sd_bus_send(NULL, reply, NULL);
}
int bus_machine_method_open_pty(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *reply = NULL;
_cleanup_free_ char *pty_name = NULL;
_cleanup_close_ int master = -1;
Machine *m = userdata;
int r;
assert(message);
assert(m);
r = bus_verify_polkit_async(
message,
CAP_SYS_ADMIN,
m->class == MACHINE_HOST ? "org.freedesktop.machine1.host-open-pty" : "org.freedesktop.machine1.open-pty",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
master = machine_openpt(m, O_RDWR|O_NOCTTY|O_CLOEXEC);
if (master < 0)
return master;
r = ptsname_namespace(master, &pty_name);
if (r < 0)
return r;
r = sd_bus_message_new_method_return(message, &reply);
if (r < 0)
return r;
r = sd_bus_message_append(reply, "hs", master, pty_name);
if (r < 0)
return r;
return sd_bus_send(NULL, reply, NULL);
}
static int container_bus_new(Machine *m, sd_bus_error *error, sd_bus **ret) {
int r;
assert(m);
assert(ret);
switch (m->class) {
case MACHINE_HOST:
*ret = NULL;
break;
case MACHINE_CONTAINER: {
_cleanup_(sd_bus_unrefp) sd_bus *bus = NULL;
char *address;
r = sd_bus_new(&bus);
if (r < 0)
return r;
if (asprintf(&address, "x-machine-kernel:pid=%1$" PID_PRI ";x-machine-unix:pid=%1$" PID_PRI, m->leader) < 0)
return -ENOMEM;
bus->address = address;
bus->bus_client = true;
bus->trusted = false;
bus->is_system = true;
r = sd_bus_start(bus);
if (r == -ENOENT)
return sd_bus_error_set_errnof(error, r, "There is no system bus in container %s.", m->name);
if (r < 0)
return r;
*ret = bus;
bus = NULL;
break;
}
default:
return -EOPNOTSUPP;
}
return 0;
}
int bus_machine_method_open_login(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *reply = NULL;
_cleanup_free_ char *pty_name = NULL;
_cleanup_(sd_bus_flush_close_unrefp) sd_bus *allocated_bus = NULL;
_cleanup_close_ int master = -1;
sd_bus *container_bus = NULL;
Machine *m = userdata;
const char *p, *getty;
int r;
assert(message);
assert(m);
r = bus_verify_polkit_async(
message,
CAP_SYS_ADMIN,
m->class == MACHINE_HOST ? "org.freedesktop.machine1.host-login" : "org.freedesktop.machine1.login",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
master = machine_openpt(m, O_RDWR|O_NOCTTY|O_CLOEXEC);
if (master < 0)
return master;
r = ptsname_namespace(master, &pty_name);
if (r < 0)
return r;
p = path_startswith(pty_name, "/dev/pts/");
if (!p)
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "PTS name %s is invalid", pty_name);
r = container_bus_new(m, error, &allocated_bus);
if (r < 0)
return r;
container_bus = allocated_bus ?: m->manager->bus;
getty = strjoina("container-getty@", p, ".service");
r = sd_bus_call_method(
container_bus,
"org.freedesktop.systemd1",
"/org/freedesktop/systemd1",
"org.freedesktop.systemd1.Manager",
"StartUnit",
error, NULL,
"ss", getty, "replace");
if (r < 0)
return r;
r = sd_bus_message_new_method_return(message, &reply);
if (r < 0)
return r;
r = sd_bus_message_append(reply, "hs", master, pty_name);
if (r < 0)
return r;
return sd_bus_send(NULL, reply, NULL);
}
int bus_machine_method_open_shell(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *reply = NULL, *tm = NULL;
_cleanup_free_ char *pty_name = NULL;
_cleanup_(sd_bus_flush_close_unrefp) sd_bus *allocated_bus = NULL;
sd_bus *container_bus = NULL;
_cleanup_close_ int master = -1, slave = -1;
_cleanup_strv_free_ char **env = NULL, **args = NULL;
Machine *m = userdata;
const char *p, *unit, *user, *path, *description, *utmp_id;
int r;
assert(message);
assert(m);
r = sd_bus_message_read(message, "ss", &user, &path);
if (r < 0)
return r;
if (isempty(user))
user = NULL;
if (isempty(path))
path = "/bin/sh";
if (!path_is_absolute(path))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Specified path '%s' is not absolute", path);
r = sd_bus_message_read_strv(message, &args);
if (r < 0)
return r;
if (strv_isempty(args)) {
args = strv_free(args);
args = strv_new(path, NULL);
if (!args)
return -ENOMEM;
args[0][0] = '-'; /* Tell /bin/sh that this shall be a login shell */
}
r = sd_bus_message_read_strv(message, &env);
if (r < 0)
return r;
if (!strv_env_is_valid(env))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Invalid environment assignments");
r = bus_verify_polkit_async(
message,
CAP_SYS_ADMIN,
m->class == MACHINE_HOST ? "org.freedesktop.machine1.host-shell" : "org.freedesktop.machine1.shell",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
master = machine_openpt(m, O_RDWR|O_NOCTTY|O_CLOEXEC);
if (master < 0)
return master;
r = ptsname_namespace(master, &pty_name);
if (r < 0)
return r;
p = path_startswith(pty_name, "/dev/pts/");
assert(p);
slave = machine_open_terminal(m, pty_name, O_RDWR|O_NOCTTY|O_CLOEXEC);
if (slave < 0)
return slave;
utmp_id = path_startswith(pty_name, "/dev/");
assert(utmp_id);
r = container_bus_new(m, error, &allocated_bus);
if (r < 0)
return r;
container_bus = allocated_bus ?: m->manager->bus;
r = sd_bus_message_new_method_call(
container_bus,
&tm,
"org.freedesktop.systemd1",
"/org/freedesktop/systemd1",
"org.freedesktop.systemd1.Manager",
"StartTransientUnit");
if (r < 0)
return r;
/* Name and mode */
unit = strjoina("container-shell@", p, ".service", NULL);
r = sd_bus_message_append(tm, "ss", unit, "fail");
if (r < 0)
return r;
/* Properties */
r = sd_bus_message_open_container(tm, 'a', "(sv)");
if (r < 0)
return r;
description = strjoina("Shell for User ", isempty(user) ? "root" : user);
r = sd_bus_message_append(tm,
"(sv)(sv)(sv)(sv)(sv)(sv)(sv)(sv)(sv)(sv)(sv)(sv)",
"Description", "s", description,
"StandardInputFileDescriptor", "h", slave,
"StandardOutputFileDescriptor", "h", slave,
"StandardErrorFileDescriptor", "h", slave,
"SendSIGHUP", "b", true,
"IgnoreSIGPIPE", "b", false,
"KillMode", "s", "mixed",
"TTYReset", "b", true,
"UtmpIdentifier", "s", utmp_id,
"UtmpMode", "s", "user",
"PAMName", "s", "login",
"WorkingDirectory", "s", "-~");
if (r < 0)
return r;
r = sd_bus_message_append(tm, "(sv)", "User", "s", isempty(user) ? "root" : user);
if (r < 0)
return r;
if (!strv_isempty(env)) {
r = sd_bus_message_open_container(tm, 'r', "sv");
if (r < 0)
return r;
r = sd_bus_message_append(tm, "s", "Environment");
if (r < 0)
return r;
r = sd_bus_message_open_container(tm, 'v', "as");
if (r < 0)
return r;
r = sd_bus_message_append_strv(tm, env);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
}
/* Exec container */
r = sd_bus_message_open_container(tm, 'r', "sv");
if (r < 0)
return r;
r = sd_bus_message_append(tm, "s", "ExecStart");
if (r < 0)
return r;
r = sd_bus_message_open_container(tm, 'v', "a(sasb)");
if (r < 0)
return r;
r = sd_bus_message_open_container(tm, 'a', "(sasb)");
if (r < 0)
return r;
r = sd_bus_message_open_container(tm, 'r', "sasb");
if (r < 0)
return r;
r = sd_bus_message_append(tm, "s", path);
if (r < 0)
return r;
r = sd_bus_message_append_strv(tm, args);
if (r < 0)
return r;
r = sd_bus_message_append(tm, "b", true);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
r = sd_bus_message_close_container(tm);
if (r < 0)
return r;
/* Auxiliary units */
r = sd_bus_message_append(tm, "a(sa(sv))", 0);
if (r < 0)
return r;
r = sd_bus_call(container_bus, tm, 0, error, NULL);
if (r < 0)
return r;
slave = safe_close(slave);
r = sd_bus_message_new_method_return(message, &reply);
if (r < 0)
return r;
r = sd_bus_message_append(reply, "hs", master, pty_name);
if (r < 0)
return r;
return sd_bus_send(NULL, reply, NULL);
}
int bus_machine_method_bind_mount(sd_bus_message *message, void *userdata, sd_bus_error *error) {
_cleanup_close_pair_ int errno_pipe_fd[2] = { -1, -1 };
char mount_slave[] = "/tmp/propagate.XXXXXX", *mount_tmp, *mount_outside, *p;
bool mount_slave_created = false, mount_slave_mounted = false,
mount_tmp_created = false, mount_tmp_mounted = false,
mount_outside_created = false, mount_outside_mounted = false;
const char *dest, *src;
Machine *m = userdata;
int read_only, make_directory;
pid_t child;
siginfo_t si;
int r;
assert(message);
assert(m);
if (m->class != MACHINE_CONTAINER)
return sd_bus_error_setf(error, SD_BUS_ERROR_NOT_SUPPORTED, "Bind mounting is only supported on container machines.");
r = sd_bus_message_read(message, "ssbb", &src, &dest, &read_only, &make_directory);
if (r < 0)
return r;
if (!path_is_absolute(src) || !path_is_safe(src))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Source path must be absolute and not contain ../.");
if (isempty(dest))
dest = src;
else if (!path_is_absolute(dest) || !path_is_safe(dest))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Destination path must be absolute and not contain ../.");
r = bus_verify_polkit_async(
message,
CAP_SYS_ADMIN,
"org.freedesktop.machine1.manage-machines",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
/* One day, when bind mounting /proc/self/fd/n works across
* namespace boundaries we should rework this logic to make
* use of it... */
p = strjoina("/run/systemd/nspawn/propagate/", m->name, "/");
if (laccess(p, F_OK) < 0)
return sd_bus_error_setf(error, SD_BUS_ERROR_NOT_SUPPORTED, "Container does not allow propagation of mount points.");
/* Our goal is to install a new bind mount into the container,
possibly read-only. This is irritatingly complex
unfortunately, currently.
First, we start by creating a private playground in /tmp,
that we can mount MS_SLAVE. (Which is necessary, since
MS_MOVE cannot be applied to mounts with MS_SHARED parent
mounts.) */
if (!mkdtemp(mount_slave))
return sd_bus_error_set_errnof(error, errno, "Failed to create playground %s: %m", mount_slave);
mount_slave_created = true;
if (mount(mount_slave, mount_slave, NULL, MS_BIND, NULL) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to make bind mount %s: %m", mount_slave);
goto finish;
}
mount_slave_mounted = true;
if (mount(NULL, mount_slave, NULL, MS_SLAVE, NULL) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to remount slave %s: %m", mount_slave);
goto finish;
}
/* Second, we mount the source directory to a directory inside
of our MS_SLAVE playground. */
mount_tmp = strjoina(mount_slave, "/mount");
if (mkdir(mount_tmp, 0700) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to create temporary mount point %s: %m", mount_tmp);
goto finish;
}
mount_tmp_created = true;
if (mount(src, mount_tmp, NULL, MS_BIND, NULL) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to overmount %s: %m", mount_tmp);
goto finish;
}
mount_tmp_mounted = true;
/* Third, we remount the new bind mount read-only if requested. */
if (read_only)
if (mount(NULL, mount_tmp, NULL, MS_BIND|MS_REMOUNT|MS_RDONLY, NULL) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to remount read-only %s: %m", mount_tmp);
goto finish;
}
/* Fourth, we move the new bind mount into the propagation
* directory. This way it will appear there read-only
* right-away. */
mount_outside = strjoina("/run/systemd/nspawn/propagate/", m->name, "/XXXXXX");
if (!mkdtemp(mount_outside)) {
r = sd_bus_error_set_errnof(error, errno, "Cannot create propagation directory %s: %m", mount_outside);
goto finish;
}
mount_outside_created = true;
if (mount(mount_tmp, mount_outside, NULL, MS_MOVE, NULL) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to move %s to %s: %m", mount_tmp, mount_outside);
goto finish;
}
mount_outside_mounted = true;
mount_tmp_mounted = false;
(void) rmdir(mount_tmp);
mount_tmp_created = false;
(void) umount(mount_slave);
mount_slave_mounted = false;
(void) rmdir(mount_slave);
mount_slave_created = false;
if (pipe2(errno_pipe_fd, O_CLOEXEC|O_NONBLOCK) < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to create pipe: %m");
goto finish;
}
child = fork();
if (child < 0) {
r = sd_bus_error_set_errnof(error, errno, "Failed to fork(): %m");
goto finish;
}
if (child == 0) {
const char *mount_inside;
int mntfd;
const char *q;
errno_pipe_fd[0] = safe_close(errno_pipe_fd[0]);
q = procfs_file_alloca(m->leader, "ns/mnt");
mntfd = open(q, O_RDONLY|O_NOCTTY|O_CLOEXEC);
if (mntfd < 0) {
r = log_error_errno(errno, "Failed to open mount namespace of leader: %m");
goto child_fail;
}
if (setns(mntfd, CLONE_NEWNS) < 0) {
r = log_error_errno(errno, "Failed to join namespace of leader: %m");
goto child_fail;
}
if (make_directory)
(void) mkdir_p(dest, 0755);
/* Fifth, move the mount to the right place inside */
mount_inside = strjoina("/run/systemd/nspawn/incoming/", basename(mount_outside));
if (mount(mount_inside, dest, NULL, MS_MOVE, NULL) < 0) {
r = log_error_errno(errno, "Failed to mount: %m");
goto child_fail;
}
_exit(EXIT_SUCCESS);
child_fail:
(void) write(errno_pipe_fd[1], &r, sizeof(r));
errno_pipe_fd[1] = safe_close(errno_pipe_fd[1]);
_exit(EXIT_FAILURE);
}
errno_pipe_fd[1] = safe_close(errno_pipe_fd[1]);
r = wait_for_terminate(child, &si);
if (r < 0) {
r = sd_bus_error_set_errnof(error, r, "Failed to wait for child: %m");
goto finish;
}
if (si.si_code != CLD_EXITED) {
r = sd_bus_error_setf(error, SD_BUS_ERROR_FAILED, "Child died abnormally.");
goto finish;
}
if (si.si_status != EXIT_SUCCESS) {
if (read(errno_pipe_fd[0], &r, sizeof(r)) == sizeof(r))
r = sd_bus_error_set_errnof(error, r, "Failed to mount: %m");
else
r = sd_bus_error_setf(error, SD_BUS_ERROR_FAILED, "Child failed.");
goto finish;
}
r = sd_bus_reply_method_return(message, NULL);
finish:
if (mount_outside_mounted)
umount(mount_outside);
if (mount_outside_created)
rmdir(mount_outside);
if (mount_tmp_mounted)
umount(mount_tmp);
if (mount_tmp_created)
rmdir(mount_tmp);
if (mount_slave_mounted)
umount(mount_slave);
if (mount_slave_created)
rmdir(mount_slave);
return r;
}
static int machine_operation_done(sd_event_source *s, const siginfo_t *si, void *userdata) {
_cleanup_(sd_bus_error_free) sd_bus_error error = SD_BUS_ERROR_NULL;
MachineOperation *o = userdata;
int r;
assert(o);
assert(si);
o->pid = 0;
if (si->si_code != CLD_EXITED) {
r = sd_bus_error_setf(&error, SD_BUS_ERROR_FAILED, "Child died abnormally.");
goto fail;
}
if (si->si_status != EXIT_SUCCESS) {
if (read(o->errno_fd, &r, sizeof(r)) == sizeof(r))
r = sd_bus_error_set_errnof(&error, r, "%m");
else
r = sd_bus_error_setf(&error, SD_BUS_ERROR_FAILED, "Child failed.");
goto fail;
}
r = sd_bus_reply_method_return(o->message, NULL);
if (r < 0)
log_error_errno(r, "Failed to reply to message: %m");
machine_operation_unref(o);
return 0;
fail:
r = sd_bus_reply_method_error(o->message, &error);
if (r < 0)
log_error_errno(r, "Failed to reply to message: %m");
machine_operation_unref(o);
return 0;
}
int bus_machine_method_copy(sd_bus_message *message, void *userdata, sd_bus_error *error) {
const char *src, *dest, *host_path, *container_path, *host_basename, *host_dirname, *container_basename, *container_dirname;
_cleanup_close_pair_ int errno_pipe_fd[2] = { -1, -1 };
_cleanup_close_ int hostfd = -1;
Machine *m = userdata;
MachineOperation *o;
bool copy_from;
pid_t child;
char *t;
int r;
assert(message);
assert(m);
if (m->n_operations >= MACHINE_OPERATIONS_MAX)
return sd_bus_error_setf(error, SD_BUS_ERROR_LIMITS_EXCEEDED, "Too many ongoing copies.");
if (m->class != MACHINE_CONTAINER)
return sd_bus_error_setf(error, SD_BUS_ERROR_NOT_SUPPORTED, "Copying files is only supported on container machines.");
r = sd_bus_message_read(message, "ss", &src, &dest);
if (r < 0)
return r;
if (!path_is_absolute(src))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Source path must be absolute.");
if (isempty(dest))
dest = src;
else if (!path_is_absolute(dest))
return sd_bus_error_setf(error, SD_BUS_ERROR_INVALID_ARGS, "Destination path must be absolute.");
r = bus_verify_polkit_async(
message,
CAP_SYS_ADMIN,
"org.freedesktop.machine1.manage-machines",
NULL,
false,
UID_INVALID,
&m->manager->polkit_registry,
error);
if (r < 0)
return r;
if (r == 0)
return 1; /* Will call us back */
copy_from = strstr(sd_bus_message_get_member(message), "CopyFrom");
if (copy_from) {
container_path = src;
host_path = dest;
} else {
host_path = src;
container_path = dest;
}
host_basename = basename(host_path);
t = strdupa(host_path);
host_dirname = dirname(t);
container_basename = basename(container_path);
t = strdupa(container_path);
container_dirname = dirname(t);
hostfd = open(host_dirname, O_CLOEXEC|O_RDONLY|O_NOCTTY|O_DIRECTORY);
if (hostfd < 0)
return sd_bus_error_set_errnof(error, errno, "Failed to open host directory %s: %m", host_dirname);
if (pipe2(errno_pipe_fd, O_CLOEXEC|O_NONBLOCK) < 0)
return sd_bus_error_set_errnof(error, errno, "Failed to create pipe: %m");
child = fork();
if (child < 0)
return sd_bus_error_set_errnof(error, errno, "Failed to fork(): %m");
if (child == 0) {
int containerfd;
const char *q;
int mntfd;
errno_pipe_fd[0] = safe_close(errno_pipe_fd[0]);
q = procfs_file_alloca(m->leader, "ns/mnt");
mntfd = open(q, O_RDONLY|O_NOCTTY|O_CLOEXEC);
if (mntfd < 0) {
r = log_error_errno(errno, "Failed to open mount namespace of leader: %m");
goto child_fail;
}
if (setns(mntfd, CLONE_NEWNS) < 0) {
r = log_error_errno(errno, "Failed to join namespace of leader: %m");
goto child_fail;
}
containerfd = open(container_dirname, O_CLOEXEC|O_RDONLY|O_NOCTTY|O_DIRECTORY);
if (containerfd < 0) {
r = log_error_errno(errno, "Failed top open destination directory: %m");
goto child_fail;
}
if (copy_from)
r = copy_tree_at(containerfd, container_basename, hostfd, host_basename, true);
else
r = copy_tree_at(hostfd, host_basename, containerfd, container_basename, true);
hostfd = safe_close(hostfd);
containerfd = safe_close(containerfd);
if (r < 0) {
r = log_error_errno(r, "Failed to copy tree: %m");
goto child_fail;
}
_exit(EXIT_SUCCESS);
child_fail:
(void) write(errno_pipe_fd[1], &r, sizeof(r));
errno_pipe_fd[1] = safe_close(errno_pipe_fd[1]);
_exit(EXIT_FAILURE);
}
errno_pipe_fd[1] = safe_close(errno_pipe_fd[1]);
/* Copying might take a while, hence install a watch the
* child, and return */
o = new0(MachineOperation, 1);
if (!o)
return log_oom();
o->pid = child;
o->message = sd_bus_message_ref(message);
o->errno_fd = errno_pipe_fd[0];
errno_pipe_fd[0] = -1;
r = sd_event_add_child(m->manager->event, &o->event_source, child, WEXITED, machine_operation_done, o);
if (r < 0) {
machine_operation_unref(o);
return log_oom();
}
LIST_PREPEND(operations, m->operations, o);
m->n_operations++;
o->machine = m;
return 1;
}
const sd_bus_vtable machine_vtable[] = {
SD_BUS_VTABLE_START(0),
SD_BUS_PROPERTY("Name", "s", NULL, offsetof(Machine, name), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("Id", "ay", property_get_id, 0, SD_BUS_VTABLE_PROPERTY_CONST),
BUS_PROPERTY_DUAL_TIMESTAMP("Timestamp", offsetof(Machine, timestamp), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("Service", "s", NULL, offsetof(Machine, service), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("Unit", "s", NULL, offsetof(Machine, unit), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("Scope", "s", NULL, offsetof(Machine, unit), SD_BUS_VTABLE_PROPERTY_CONST|SD_BUS_VTABLE_HIDDEN),
SD_BUS_PROPERTY("Leader", "u", NULL, offsetof(Machine, leader), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("Class", "s", property_get_class, offsetof(Machine, class), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("RootDirectory", "s", NULL, offsetof(Machine, root_directory), SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("NetworkInterfaces", "ai", property_get_netif, 0, SD_BUS_VTABLE_PROPERTY_CONST),
SD_BUS_PROPERTY("State", "s", property_get_state, 0, 0),
SD_BUS_METHOD("Terminate", NULL, NULL, bus_machine_method_terminate, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("Kill", "si", NULL, bus_machine_method_kill, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("GetAddresses", NULL, "a(iay)", bus_machine_method_get_addresses, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("GetOSRelease", NULL, "a{ss}", bus_machine_method_get_os_release, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("OpenPTY", NULL, "hs", bus_machine_method_open_pty, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("OpenLogin", NULL, "hs", bus_machine_method_open_login, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("OpenShell", "ssasas", "hs", bus_machine_method_open_shell, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("BindMount", "ssbb", NULL, bus_machine_method_bind_mount, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("CopyFrom", "ss", NULL, bus_machine_method_copy, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_METHOD("CopyTo", "ss", NULL, bus_machine_method_copy, SD_BUS_VTABLE_UNPRIVILEGED),
SD_BUS_VTABLE_END
};
int machine_object_find(sd_bus *bus, const char *path, const char *interface, void *userdata, void **found, sd_bus_error *error) {
Manager *m = userdata;
Machine *machine;
int r;
assert(bus);
assert(path);
assert(interface);
assert(found);
assert(m);
if (streq(path, "/org/freedesktop/machine1/machine/self")) {
_cleanup_(sd_bus_creds_unrefp) sd_bus_creds *creds = NULL;
sd_bus_message *message;
pid_t pid;
message = sd_bus_get_current_message(bus);
if (!message)
return 0;
r = sd_bus_query_sender_creds(message, SD_BUS_CREDS_PID, &creds);
if (r < 0)
return r;
r = sd_bus_creds_get_pid(creds, &pid);
if (r < 0)
return r;
r = manager_get_machine_by_pid(m, pid, &machine);
if (r <= 0)
return 0;
} else {
_cleanup_free_ char *e = NULL;
const char *p;
p = startswith(path, "/org/freedesktop/machine1/machine/");
if (!p)
return 0;
e = bus_label_unescape(p);
if (!e)
return -ENOMEM;
machine = hashmap_get(m->machines, e);
if (!machine)
return 0;
}
*found = machine;
return 1;
}
char *machine_bus_path(Machine *m) {
_cleanup_free_ char *e = NULL;
assert(m);
e = bus_label_escape(m->name);
if (!e)
return NULL;
return strappend("/org/freedesktop/machine1/machine/", e);
}
int machine_node_enumerator(sd_bus *bus, const char *path, void *userdata, char ***nodes, sd_bus_error *error) {
_cleanup_strv_free_ char **l = NULL;
Machine *machine = NULL;
Manager *m = userdata;
Iterator i;
int r;
assert(bus);
assert(path);
assert(nodes);
HASHMAP_FOREACH(machine, m->machines, i) {
char *p;
p = machine_bus_path(machine);
if (!p)
return -ENOMEM;
r = strv_consume(&l, p);
if (r < 0)
return r;
}
*nodes = l;
l = NULL;
return 1;
}
int machine_send_signal(Machine *m, bool new_machine) {
_cleanup_free_ char *p = NULL;
assert(m);
p = machine_bus_path(m);
if (!p)
return -ENOMEM;
return sd_bus_emit_signal(
m->manager->bus,
"/org/freedesktop/machine1",
"org.freedesktop.machine1.Manager",
new_machine ? "MachineNew" : "MachineRemoved",
"so", m->name, p);
}
int machine_send_create_reply(Machine *m, sd_bus_error *error) {
_cleanup_(sd_bus_message_unrefp) sd_bus_message *c = NULL;
_cleanup_free_ char *p = NULL;
assert(m);
if (!m->create_message)
return 0;
c = m->create_message;
m->create_message = NULL;
if (error)
return sd_bus_reply_method_error(c, error);
/* Update the machine state file before we notify the client
* about the result. */
machine_save(m);
p = machine_bus_path(m);
if (!p)
return -ENOMEM;
return sd_bus_reply_method_return(c, "o", p);
}