core: store the invocation ID in the per-service keyring

Let's store the invocation ID in the per-service keyring as a root-owned key,
with strict access rights. This has the advantage over the environment-based ID
passing that it also works from SUID binaries (as they key cannot be overidden
by unprivileged code starting them), in contrast to the secure_getenv() based
mode.

The invocation ID is now passed in three different ways to a service:

- As environment variable $INVOCATION_ID. This is easy to use, but may be
  overriden by unprivileged code (which might be a bad or a good thing), which
  means it's incompatible with SUID code (see above).

- As extended attribute on the service cgroup. This cannot be overriden by
  unprivileged code, and may be queried safely from "outside" of a service.
  However, it is incompatible with containers right now, as unprivileged
  containers generally cannot set xattrs on cgroupfs.

- As "invocation_id" key in the kernel keyring. This has the benefit that the
  key cannot be changed by unprivileged service code, and thus is safe to
  access from SUID code (see above). But do note that service code can replace
  the session keyring with a fresh one that lacks the key. However in that case
  the key will not be owned by root, which is easily detectable. The keyring is
  also incompatible with containers right now, as it is not properly namespace
  aware (but this is being worked on), and thus most container managers mask
  the keyring-related system calls.

Ideally we'd only have one way to pass the invocation ID, but the different
ways all have limitations. The invocation ID hookup in journald is currently
only available on the host but not in containers, due to the mentioned
limitations.

How to verify the new invocation ID in the keyring:

 # elogind-run -t /bin/sh
 Running as unit: run-rd917366c04f847b480d486017f7239d6.service
 Press ^] three times within 1s to disconnect TTY.
 # keyctl show
 Session Keyring
  680208392 --alswrv      0     0  keyring: _ses
  250926536 ----s-rv      0     0   \_ user: invocation_id
 # keyctl request user invocation_id
 250926536
 # keyctl read 250926536
 16 bytes of data in key:
 9c96317c ac64495a a42b9cd7 4f3ff96b
 # echo $INVOCATION_ID
 9c96317cac64495aa42b9cd74f3ff96b
 # ^D

This creates a new transient service runnint a shell. Then verifies the
contents of the keyring, requests the invocation ID key, and reads its payload.
For comparison the invocation ID as passed via the environment variable is also
displayed.

diff --git a/src/libelogind/sd-id128/sd-id128.c b/src/libelogind/sd-id128/sd-id128.c
index 43cc86f43f5a25664892eb55e53d71ed43661448..14dfc31749d339525147abb5a431f72c5543e9e2 100644 (file)
--- a/src/libelogind/sd-id128/sd-id128.c
+++ b/src/libelogind/sd-id128/sd-id128.c
@@ -129,6 +129,105 @@ _public_ int sd_id128_get_boot(sd_id128_t *ret) {
         return 0;
 }
 
+static int get_invocation_from_keyring(sd_id128_t *ret) {
+
+        _cleanup_free_ char *description = NULL;
+        char *d, *p, *g, *u, *e;
+        unsigned long perms;
+        key_serial_t key;
+        size_t sz = 256;
+        uid_t uid;
+        gid_t gid;
+        int r, c;
+
+#define MAX_PERMS ((unsigned long) (KEY_POS_VIEW|KEY_POS_READ|KEY_POS_SEARCH| \
+                                    KEY_USR_VIEW|KEY_USR_READ|KEY_USR_SEARCH))
+
+        assert(ret);
+
+        key = request_key("user", "invocation_id", NULL, 0);
+        if (key == -1) {
+                /* Keyring support not available? No invocation key stored? */
+                if (IN_SET(errno, ENOSYS, ENOKEY))
+                        return 0;
+
+                return -errno;
+        }
+
+        for (;;) {
+                description = new(char, sz);
+                if (!description)
+                        return -ENOMEM;
+
+                c = keyctl(KEYCTL_DESCRIBE, key, (unsigned long) description, sz, 0);
+                if (c < 0)
+                        return -errno;
+
+                if ((size_t) c <= sz)
+                        break;
+
+                sz = c;
+                free(description);
+        }
+
+        /* The kernel returns a final NUL in the string, verify that. */
+        assert(description[c-1] == 0);
+
+        /* Chop off the final description string */
+        d = strrchr(description, ';');
+        if (!d)
+                return -EIO;
+        *d = 0;
+
+        /* Look for the permissions */
+        p = strrchr(description, ';');
+        if (!p)
+                return -EIO;
+
+        errno = 0;
+        perms = strtoul(p + 1, &e, 16);
+        if (errno > 0)
+                return -errno;
+        if (e == p + 1) /* Read at least one character */
+                return -EIO;
+        if (e != d) /* Must reached the end */
+                return -EIO;
+
+        if ((perms & ~MAX_PERMS) != 0)
+                return -EPERM;
+
+        *p = 0;
+
+        /* Look for the group ID */
+        g = strrchr(description, ';');
+        if (!g)
+                return -EIO;
+        r = parse_gid(g + 1, &gid);
+        if (r < 0)
+                return r;
+        if (gid != 0)
+                return -EPERM;
+        *g = 0;
+
+        /* Look for the user ID */
+        u = strrchr(description, ';');
+        if (!u)
+                return -EIO;
+        r = parse_uid(u + 1, &uid);
+        if (r < 0)
+                return r;
+        if (uid != 0)
+                return -EPERM;
+
+        c = keyctl(KEYCTL_READ, key, (unsigned long) ret, sizeof(sd_id128_t), 0);
+        if (c < 0)
+                return -errno;
+        if (c != sizeof(sd_id128_t))
+                return -EIO;
+
+        return 1;
+}
+
 _public_ int sd_id128_get_invocation(sd_id128_t *ret) {
         static thread_local sd_id128_t saved_invocation_id = {};
         int r;
@@ -136,15 +235,31 @@ _public_ int sd_id128_get_invocation(sd_id128_t *ret) {
         assert_return(ret, -EINVAL);
 
         if (sd_id128_is_null(saved_invocation_id)) {
-                const char *e;
 
-                e = secure_getenv("INVOCATION_ID");
-                if (!e)
-                        return -ENXIO;
+                /* We first try to read the invocation ID from the kernel keyring. This has the benefit that it is not
+                 * fakeable by unprivileged code. If the information is not available in the keyring, we use
+                 * $INVOCATION_ID but ignore the data if our process was called by less privileged code
+                 * (i.e. secure_getenv() instead of getenv()).
+                 *
+                 * The kernel keyring is only relevant for system services (as for user services we don't store the
+                 * invocation ID in the keyring, as there'd be no trust benefit in that). The environment variable is
+                 * primarily relevant for user services, and sufficiently safe as no privilege boundary is involved. */
 
-                r = sd_id128_from_string(e, &saved_invocation_id);
+                r = get_invocation_from_keyring(&saved_invocation_id);
                 if (r < 0)
                         return r;
+
+                if (r == 0) {
+                        const char *e;
+
+                        e = secure_getenv("INVOCATION_ID");
+                        if (!e)
+                                return -ENXIO;
+
+                        r = sd_id128_from_string(e, &saved_invocation_id);
+                        if (r < 0)
+                                return r;
+                }
         }
 
         *ret = saved_invocation_id;