nspawn: implement configurable syscall whitelisting/blacklisting

Now that we have ported nspawn's seccomp code to the generic code in
seccomp-util, let's extend it to support whitelisting and blacklisting
of specific additional syscalls.

This uses similar syntax as PID1's support for system call filtering,
but in contrast to that always implements a blacklist (and not a
whitelist), as we prepopulate the filter with a blacklist, and the
unit's system call filter logic does not come with anything
prepopulated.

(Later on we might actually want to invert the logic here, and
whitelist rather than blacklist things, but at this point let's not do
that. In case we switch this over later, the syscall add/remove logic of
this commit should be compatible conceptually.)

Fixes: #5163

Replaces: #5944
This commit is contained in:
Lennart Poettering
2017-09-12 14:06:21 +02:00
parent 7609340e2f
commit 960e4569e1
10 changed files with 169 additions and 20 deletions
+52 -1
View File
@@ -208,6 +208,8 @@ static unsigned long arg_clone_ns_flags = CLONE_NEWIPC|CLONE_NEWPID|CLONE_NEWUTS
static MountSettingsMask arg_mount_settings = MOUNT_APPLY_APIVFS_RO;
static void *arg_root_hash = NULL;
static size_t arg_root_hash_size = 0;
static char **arg_syscall_whitelist = NULL;
static char **arg_syscall_blacklist = NULL;
static void help(void) {
printf("%s [OPTIONS...] [PATH] [ARGUMENTS...]\n\n"
@@ -267,6 +269,8 @@ static void help(void) {
" --capability=CAP In addition to the default, retain specified\n"
" capability\n"
" --drop-capability=CAP Drop the specified capability from the default set\n"
" --system-call-filter=LIST|~LIST\n"
" Permit/prohibit specific system calls\n"
" --kill-signal=SIGNAL Select signal to use for shutting down PID 1\n"
" --link-journal=MODE Link up guest journal, one of no, auto, guest, \n"
" host, try-guest, try-host\n"
@@ -431,6 +435,7 @@ static int parse_argv(int argc, char *argv[]) {
ARG_PRIVATE_USERS_CHOWN,
ARG_NOTIFY_READY,
ARG_ROOT_HASH,
ARG_SYSTEM_CALL_FILTER,
};
static const struct option options[] = {
@@ -482,6 +487,7 @@ static int parse_argv(int argc, char *argv[]) {
{ "pivot-root", required_argument, NULL, ARG_PIVOT_ROOT },
{ "notify-ready", required_argument, NULL, ARG_NOTIFY_READY },
{ "root-hash", required_argument, NULL, ARG_ROOT_HASH },
{ "system-call-filter", required_argument, NULL, ARG_SYSTEM_CALL_FILTER },
{}
};
@@ -1051,6 +1057,36 @@ static int parse_argv(int argc, char *argv[]) {
break;
}
case ARG_SYSTEM_CALL_FILTER: {
bool negative;
const char *items;
negative = optarg[0] == '~';
items = negative ? optarg + 1 : optarg;
for (;;) {
_cleanup_free_ char *word = NULL;
r = extract_first_word(&items, &word, NULL, 0);
if (r == 0)
break;
if (r == -ENOMEM)
return log_oom();
if (r < 0)
return log_error_errno(r, "Failed to parse system call filter: %m");
if (negative)
r = strv_extend(&arg_syscall_blacklist, word);
else
r = strv_extend(&arg_syscall_whitelist, word);
if (r < 0)
return log_oom();
}
arg_settings_mask |= SETTING_SYSCALL_FILTER;
break;
}
case '?':
return -EINVAL;
@@ -2606,7 +2642,7 @@ static int outer_child(
if (r < 0)
return r;
r = setup_seccomp(arg_caps_retain);
r = setup_seccomp(arg_caps_retain, arg_syscall_whitelist, arg_syscall_blacklist);
if (r < 0)
return r;
@@ -3111,6 +3147,21 @@ static int load_settings(void) {
if ((arg_settings_mask & SETTING_NOTIFY_READY) == 0)
arg_notify_ready = settings->notify_ready;
if ((arg_settings_mask & SETTING_SYSCALL_FILTER) == 0) {
if (!arg_settings_trusted && !strv_isempty(arg_syscall_whitelist))
log_warning("Ignoring SystemCallFilter= settings, file %s is not trusted.", p);
else {
strv_free(arg_syscall_whitelist);
strv_free(arg_syscall_blacklist);
arg_syscall_whitelist = settings->syscall_whitelist;
arg_syscall_blacklist = settings->syscall_blacklist;
settings->syscall_whitelist = settings->syscall_blacklist = NULL;
}
}
return 0;
}