Add the identity command and serve it on an opt-in HTTP info API

Build slackhq/nebula#1821 ("who is this vpn address?" over a local
HTTP+JSON socket) on the command registry, so the lookup is one command
on every surface and the socket is one more local API.

identity
- `identity [-address <vpn addr>[:port]]`: no address reports this node;
  with one, this node or the peer behind an established tunnel
- Certificate fields only: name, vpn addresses, networks, unsafe
  networks, groups, fingerprint, issuer, validity, cert version; a peer's
  fingerprint is the one nebula computed when it verified the cert
- Port accepted and stripped, 4in6 unmapped, so a RemoteAddr passes
  straight through
- No tunnel is a commands.NotFound; an empty -address is refused
- Lighthouse TXT lookup shares the owner and peer lookups, keeps answering
  with the initiating certificate, and no longer panics without a default cert
- Served on ssh and nebula ctl too

Info API
- Off until `local_api.info.enabled` with `socket` and optional quoted
  octal `socket_mode` (default "0600"); nothing derived
- Serves commands declaring Command.Identity plus help; others 403 naming
  their surfaces. TestInfoAPISurface pins the set to identity
- GET /v1 is the schema; GET /v1/<cmd>?<flag>=<value> runs one; body is
  the -json rendering via commands.Render; commands.PresentationFormat
  reads json/pretty from the query the way a line is read
- 400 commands.ParamError, 403 commands.WithheldError (what Resolve now
  returns for a withheld name, so help and dispatch type-check it too),
  404 NotFound or unknown, 500 Run error, refusal with no reason, or
  recovered panic with no detail; no redirects; every miss is JSON
- listen takes a mode: bind under umask 0077 then chmod; "0660" plus a
  searchable dir opens it to a group
- The chmod, at bind and on a mode-only reload, goes through chmodSocket,
  which never follows a link: a user who can write the socket's directory
  could otherwise swap the node for a link and have a root daemon put
  0660 or 0666 on any file. Only a socket we own is chmoded. On linux the
  node is held open O_PATH|O_NOFOLLOW, checked by fstat, and chmoded
  through /proc/self/fd, so the check and the chmod land on one inode;
  fchmodat2 is the fallback without /proc. On darwin and the BSDs it is
  an Lstat check and fchmodat(AT_SYMLINK_NOFOLLOW). clearStale Lstats
  too, so a dangling link is refused rather than followed by bind on
  darwin
- One endpoint type owns bind under mutex, bind-before-release move,
  context watch and stop/unlink; each server supplies what it serves
- Reloadable through the same reconcile shape as the control socket:
  mode applied in place, unparsable section keeps what runs, control
  socket path refused by name
- Wire returns Endpoints{Control, Info} with one Stop
- Windows unsupported for now (a pipe has no mode); warned once

Tests: identity lookups and shapes; API filter, status codes, panic
recovery, socket mode, rebind, cancellation; config parsing incl. bare
0660; Wire transitions; e2e of two nebulas with the info API resolving
the peer by address and port, through shared test.HTTPDo/HTTPGet helpers.
Goldens gain identity.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
Matt RichardsonandClaude Fable 5.1 committed 2026-10-21 12:26:09 -04:00
1 parent 92c67a52b2
commit 613c7e3061
38 files changed
+2937 -332

No files matched your search

+106 -2
View File
@@ -288,6 +288,52 @@ func TestListenerDoesNotUnlinkOnClose(t *testing.T) {
assert.NoError(t, err, "closing the old listener left the newer daemon's socket in place")
}
// TestRebindRecreatesASocketRemovedFromUnderIt: logind clearing /run/user/<uid> on
// logout, or a unit cleaning a RuntimeDirectory= it shares, takes the node away while
// the listener stays open and unreachable. The next reload, which Rebinds onto the same
// path, has to notice and bind the socket again rather than return early because the
// path has not changed.
func TestRebindRecreatesASocketRemovedFromUnderIt(t *testing.T) {
path := testSocket(t)
s := NewServer(test.NewLogger(), basicRegistry())
require.NoError(t, s.Rebind(context.Background(), path, ownerOnly()))
t.Cleanup(s.Stop)
require.NoError(t, os.Remove(path))
_, err := (&Client{Path: path}).Describe()
require.Error(t, err, "nothing can reach the daemon with its node gone")
require.NoError(t, s.Rebind(context.Background(), path, ownerOnly()))
assert.Equal(t, path, s.Path())
_, err = (&Client{Path: path}).Describe()
require.NoError(t, err, "the plain reload brought the socket back")
// And a Rebind with the socket in place is still a no-op.
before, ok := identifySocket(path)
require.True(t, ok)
require.NoError(t, s.Rebind(context.Background(), path, ownerOnly()))
after, _ := identifySocket(path)
assert.Equal(t, before, after, "the node was left alone")
}
// TestStopLeavesAStrangersSocketAlone: once our node has been taken from under us and
// another daemon has bound the path, the node there is that daemon's, and our Stop must
// not unlink it by path.
func TestStopLeavesAStrangersSocketAlone(t *testing.T) {
path := testSocket(t)
s := NewServer(test.NewLogger(), basicRegistry())
require.NoError(t, s.Rebind(context.Background(), path, ownerOnly()))
require.NoError(t, os.Remove(path))
other, err := net.Listen(network, path)
require.NoError(t, err)
t.Cleanup(func() { other.Close() })
s.Stop()
_, err = os.Stat(path)
assert.NoError(t, err, "the newer daemon's socket is still in place")
}
// TestListenRefusesADirectoryOthersCanWrite: whoever can write the socket's directory
// can put a socket of their own at the path and answer the clients, who trust the path;
// so a directory writable by its group or by everyone is refused, and nothing is bound.
@@ -322,7 +368,7 @@ func TestListenRefusesADirectoryOwnedBySomebodyElse(t *testing.T) {
asUID(t, me)
ln, err := listen(filepath.Join(dir, "s.sock"), ownerOnly())
require.NoError(t, err, "the same directory, seen as our own, serves")
release(ln, filepath.Join(dir, "s.sock"))
release(ln, filepath.Join(dir, "s.sock"), socketID{})
}
// TestListenThroughASymlinkedDirectory: a link where the directory should be is judged
@@ -344,7 +390,7 @@ func TestListenThroughASymlinkedDirectory(t *testing.T) {
ln, err := listen(path, ownerOnly())
require.NoError(t, err, "our link to our directory")
release(ln, path)
release(ln, path, socketID{})
assert.NoFileExists(t, filepath.Join(real, "s.sock"))
asUID(t, me+1)
@@ -356,3 +402,61 @@ func TestListenThroughASymlinkedDirectory(t *testing.T) {
_, err = listen(path, ownerOnly())
require.ErrorContains(t, err, "writable by everyone")
}
// TestListenRefusesASymlinkAtThePath: a link where the socket should be is somebody
// else's doing, and it is refused before bind rather than followed. bind(2) on darwin
// follows a dangling link and creates the socket at its target, and the chmod after it
// would follow a link to anything, so neither the target nor the link is touched.
func TestListenRefusesASymlinkAtThePath(t *testing.T) {
path := testSocket(t)
require.NoError(t, os.Mkdir(filepath.Dir(path), 0700))
target := filepath.Join(filepath.Dir(path), "target")
require.NoError(t, os.Symlink(target, path))
_, err := listen(path, Access{Mode: 0666})
require.ErrorContains(t, err, "is not a socket")
_, statErr := os.Lstat(target)
assert.True(t, os.IsNotExist(statErr), "a dangling link's target is not created")
require.NoError(t, os.WriteFile(target, nil, 0600))
_, err = listen(path, Access{Mode: 0666})
require.ErrorContains(t, err, "is not a socket")
fi, err := os.Stat(target)
require.NoError(t, err)
assert.Equal(t, os.FileMode(0600), fi.Mode().Perm(), "the file the link names keeps its mode")
assert.True(t, fi.Mode().IsRegular())
}
// TestChmodSocketRefusesWhatIsNotOurSocket: the mode is set through a look that does not
// follow links, so a link, a plain file, or a socket of another user's at the path is
// reported instead of chmoded. The last is seen as another user by changing who we are.
func TestChmodSocketRefusesWhatIsNotOurSocket(t *testing.T) {
dir := filepath.Join(privateBase(t), "n")
require.NoError(t, os.Mkdir(dir, 0700))
file := filepath.Join(dir, "file")
require.NoError(t, os.WriteFile(file, nil, 0600))
link := filepath.Join(dir, "link")
require.NoError(t, os.Symlink(file, link))
require.ErrorContains(t, applyAccess(link, Access{Mode: 0666}), "a symlink")
require.ErrorContains(t, applyAccess(file, Access{Mode: 0666}), "a regular file")
fi, err := os.Stat(file)
require.NoError(t, err)
assert.Equal(t, os.FileMode(0600), fi.Mode().Perm())
sock := filepath.Join(dir, "s.sock")
ln, err := net.Listen(network, sock)
require.NoError(t, err)
defer ln.Close()
require.NoError(t, applyAccess(sock, Access{Mode: 0660}))
fi, err = os.Stat(sock)
require.NoError(t, err)
assert.Equal(t, os.FileMode(0660), fi.Mode().Perm())
asUID(t, currentUID()+1)
require.ErrorContains(t, applyAccess(sock, Access{Mode: 0666}), "not by us")
fi, err = os.Stat(sock)
require.NoError(t, err)
assert.Equal(t, os.FileMode(0660), fi.Mode().Perm())
}