CI / Test (ubuntu-latest) (push) Canceled after 0s
CI / Test (windows-latest) (push) Canceled after 0s
CI / Storage tests on PostgreSQL (push) Canceled after 0s
CI / Complement (federation off) (push) Canceled after 0s
CI / golangci-lint (push) Canceled after 0s
CI / Smoke test the binary (push) Canceled after 0s
Implements PLAN.md Phase 5: the Server-Server API. Warp joins rooms on other servers and lets their users join its rooms, invites across servers with the invitee's server countersigning, rejects remote invites (locally if the inviter is unreachable), and exchanges events, typing, receipts and to-device messages with other homeservers — all behind the runtime switch, which now starts and stops real outbound traffic. The protocol lives in internal/federation with no knowledge of rooms: X-Matrix signing (fuzzed parser), the spec's discovery algorithm keeping the dial address and the TLS name apart, a signed client that refuses every request while the switch is off or the policy denies the destination, and a key ring that caches other servers' keys for at most a week, with optional notaries. The roomserver takes events from elsewhere through the spec's intake: signatures and hashes (a hash mismatch keeps the redacted form), the rules against the event's own auth events, against the state before it — with /get_missing_events for small gaps and /state_ids for wide ones — and against the current state, where a failure is a soft failure. Events held without their state are outliers: a column on roomserver_events, because the current state of a room joined over federation is almost all outliers, and excluded from every timeline query. Forks are resolved at the room's head. Membership across servers is make_join/send_join, make_leave/send_leave and countersigned invites, in both directions; m.room.server_acl is enforced on everything another server sends. History before a remote join is backfilled when /messages pages past it, stored at negative stream positions. internal/federationapi is the running component: the inbound endpoints, and an outbound queue that persists what it owes each destination, sends it in transactions of 50 PDUs and 100 EDUs, backs off from ten seconds to six hours, survives restarts, and drops events persisted while federation is off rather than delivering a backlog in one burst. Its Module is what the switch starts and stops. Also: a TLS listener for port 8448, trusted key servers, key validity and an outbound CA bundle in the config; per-destination health and backoff reset in the admin API; remote profiles, aliases and directories in the Client-Server API; a Complement federation profile. See docs/federation.md, including its known gaps (no partial-state joins, no key rotation). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
260 lines
8.5 KiB
Go
260 lines
8.5 KiB
Go
package federation
|
|
|
|
import (
|
|
"context"
|
|
"encoding/json"
|
|
"errors"
|
|
"io"
|
|
"net"
|
|
"net/http"
|
|
"net/http/httptest"
|
|
"strings"
|
|
"sync/atomic"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/dariuszjaniak/warp/server/internal/matrix/signing"
|
|
"github.com/dariuszjaniak/warp/server/internal/storage"
|
|
_ "github.com/dariuszjaniak/warp/server/internal/storage/drivers"
|
|
)
|
|
|
|
// testStore opens a migrated SQLite database for the key cache.
|
|
func testStore(t *testing.T) storage.FederationStore {
|
|
t.Helper()
|
|
db, err := storage.Open(t.Context(), storage.Config{
|
|
Driver: storage.DriverSQLite,
|
|
ConnectionString: "file:" + strings.ReplaceAll(t.TempDir(), `\`, "/") + "/warp.db",
|
|
}, nil)
|
|
if err != nil {
|
|
t.Fatalf("open: %v", err)
|
|
}
|
|
t.Cleanup(func() { _ = db.Close() })
|
|
if err := db.Migrate(t.Context()); err != nil {
|
|
t.Fatalf("migrate: %v", err)
|
|
}
|
|
return db.Federation()
|
|
}
|
|
|
|
// remote is a fake homeserver: a TLS server answering for a server name,
|
|
// with its own key.
|
|
type remote struct {
|
|
name string
|
|
key *signing.KeyPair
|
|
srv *httptest.Server
|
|
mux *http.ServeMux
|
|
keyHits atomic.Int32
|
|
|
|
// validFor is the validity its key response claims.
|
|
validFor time.Duration
|
|
}
|
|
|
|
func newRemote(t *testing.T, name string) *remote {
|
|
t.Helper()
|
|
key, err := signing.Generate("")
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
r := &remote{name: name, key: key, mux: http.NewServeMux(), validFor: time.Hour}
|
|
r.mux.HandleFunc("GET /_matrix/key/v2/server", func(w http.ResponseWriter, _ *http.Request) {
|
|
r.keyHits.Add(1)
|
|
doc, _ := json.Marshal(serverKeyResponse{
|
|
ServerName: r.name,
|
|
ValidUntilTS: time.Now().Add(r.validFor).UnixMilli(),
|
|
VerifyKeys: map[string]verifyKey{r.key.KeyID(): {Key: r.key.PublicBase64()}},
|
|
OldVerifyKeys: map[string]oldVerifyKey{},
|
|
})
|
|
signed, _ := signing.SignJSON(r.name, r.key, doc)
|
|
_, _ = w.Write(signed)
|
|
})
|
|
r.srv = httptest.NewTLSServer(r.mux)
|
|
t.Cleanup(r.srv.Close)
|
|
return r
|
|
}
|
|
|
|
// clientFor builds a client that resolves every remote in the list to its
|
|
// test server. The Host is "example.com", the name the test certificate is
|
|
// issued for, and the address is the listener: the dial-target path is
|
|
// exercised on every request.
|
|
func clientFor(t *testing.T, serverName string, key *signing.KeyPair, remotes ...*remote) *Client {
|
|
t.Helper()
|
|
byName := map[string]*remote{}
|
|
for _, r := range remotes {
|
|
byName[r.name] = r
|
|
}
|
|
var tlsConfig = remotes[0].srv.Client().Transport.(*http.Transport).TLSClientConfig.Clone()
|
|
return NewClient(ClientOptions{
|
|
ServerName: serverName,
|
|
Key: key,
|
|
TLSConfig: tlsConfig,
|
|
Enabled: func() bool { return true },
|
|
Resolve: func(_ context.Context, name string) (Resolved, error) {
|
|
r, ok := byName[name]
|
|
if !ok {
|
|
return Resolved{}, errors.New("no such test server")
|
|
}
|
|
_, port, _ := net.SplitHostPort(r.srv.Listener.Addr().String())
|
|
return Resolved{Addresses: []string{r.srv.Listener.Addr().String()}, Host: "example.com:" + port}, nil
|
|
},
|
|
})
|
|
}
|
|
|
|
func newKeyRing(t *testing.T, name string, remotes ...*remote) (*KeyRing, *signing.KeyPair) {
|
|
t.Helper()
|
|
key, err := signing.Generate("")
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
return NewKeyRing(KeyRingOptions{
|
|
ServerName: name,
|
|
Key: key,
|
|
Store: testStore(t),
|
|
Client: clientFor(t, name, key, remotes...),
|
|
}), key
|
|
}
|
|
|
|
func TestKeyRingFetchesAndCaches(t *testing.T) {
|
|
r := newRemote(t, "remote.example")
|
|
ring, _ := newKeyRing(t, "warp.example", r)
|
|
|
|
pub, err := ring.PublicKey(t.Context(), "remote.example", r.key.KeyID(), time.Now().UnixMilli())
|
|
if err != nil {
|
|
t.Fatalf("PublicKey: %v", err)
|
|
}
|
|
if !pub.Equal(r.key.Public) {
|
|
t.Error("fetched the wrong key")
|
|
}
|
|
for range 3 {
|
|
if _, err := ring.PublicKey(t.Context(), "remote.example", r.key.KeyID(), time.Now().UnixMilli()); err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
}
|
|
if n := r.keyHits.Load(); n != 1 {
|
|
t.Errorf("%d key fetches, want 1", n)
|
|
}
|
|
|
|
// A key needed for a time past its validity is fetched again, in case
|
|
// the server has extended it.
|
|
if _, err := ring.PublicKey(t.Context(), "remote.example", r.key.KeyID(),
|
|
time.Now().Add(2*time.Hour).UnixMilli()); err == nil {
|
|
t.Error("a key was accepted for a time past its validity")
|
|
}
|
|
if n := r.keyHits.Load(); n != 2 {
|
|
t.Errorf("%d key fetches after asking past validity, want 2", n)
|
|
}
|
|
|
|
// Validity is capped at a week, whatever the server claims.
|
|
r.validFor = 365 * 24 * time.Hour
|
|
ring2, _ := newKeyRing(t, "warp.example", r)
|
|
if _, err := ring2.PublicKey(t.Context(), "remote.example", r.key.KeyID(),
|
|
time.Now().Add(30*24*time.Hour).UnixMilli()); err == nil {
|
|
t.Error("a key was believed for longer than a week")
|
|
}
|
|
}
|
|
|
|
func TestKeyRingRefusesAForgedResponse(t *testing.T) {
|
|
r := newRemote(t, "remote.example")
|
|
impostor, _ := signing.Generate("")
|
|
r.mux = http.NewServeMux()
|
|
r.mux.HandleFunc("GET /_matrix/key/v2/server", func(w http.ResponseWriter, _ *http.Request) {
|
|
// Lists the real key but is signed with another one.
|
|
doc, _ := json.Marshal(serverKeyResponse{
|
|
ServerName: r.name,
|
|
ValidUntilTS: time.Now().Add(time.Hour).UnixMilli(),
|
|
VerifyKeys: map[string]verifyKey{r.key.KeyID(): {Key: impostor.PublicBase64()}},
|
|
})
|
|
signed, _ := signing.SignJSON(r.name, r.key, doc)
|
|
_, _ = w.Write(signed)
|
|
})
|
|
r.srv.Config.Handler = r.mux
|
|
|
|
ring, _ := newKeyRing(t, "warp.example", r)
|
|
if _, err := ring.PublicKey(t.Context(), "remote.example", r.key.KeyID(), 0); !errors.Is(err, ErrUnknownKey) {
|
|
t.Errorf("PublicKey with a forged response = %v, want ErrUnknownKey", err)
|
|
}
|
|
}
|
|
|
|
func TestOwnKeyResponseVerifies(t *testing.T) {
|
|
ring, key := newKeyRing(t, "warp.example", newRemote(t, "unused.example"))
|
|
doc, err := ring.OwnKeyResponse()
|
|
if err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
if err := signing.VerifyJSON("warp.example", key.KeyID(), key.Public, doc); err != nil {
|
|
t.Errorf("our own key response does not verify: %v", err)
|
|
}
|
|
var parsed serverKeyResponse
|
|
if err := json.Unmarshal(doc, &parsed); err != nil {
|
|
t.Fatal(err)
|
|
}
|
|
if parsed.VerifyKeys[key.KeyID()].Key != key.PublicBase64() {
|
|
t.Errorf("verify_keys = %+v", parsed.VerifyKeys)
|
|
}
|
|
}
|
|
|
|
func TestSignedRequestIsAuthenticatedEndToEnd(t *testing.T) {
|
|
// The remote here is the *sender*: a key server for remote.example,
|
|
// and a client signing as remote.example to warp.example.
|
|
sender := newRemote(t, "remote.example")
|
|
receiver := newRemote(t, "warp.example")
|
|
ring := NewKeyRing(KeyRingOptions{
|
|
ServerName: "warp.example",
|
|
Key: receiver.key,
|
|
Store: testStore(t),
|
|
Client: clientFor(t, "warp.example", receiver.key, sender),
|
|
})
|
|
|
|
var gotOrigin atomic.Value
|
|
receiver.mux.HandleFunc("PUT /_matrix/federation/v1/send/{txn}", func(w http.ResponseWriter, r *http.Request) {
|
|
body, _ := io.ReadAll(r.Body)
|
|
origin, err := ring.AuthenticateRequest(r.Context(), r, body)
|
|
if err != nil {
|
|
http.Error(w, `{"errcode":"M_UNAUTHORIZED","error":"`+err.Error()+`"}`, http.StatusUnauthorized)
|
|
return
|
|
}
|
|
gotOrigin.Store(origin)
|
|
_, _ = w.Write([]byte(`{"pdus":{}}`))
|
|
})
|
|
|
|
client := clientFor(t, "remote.example", sender.key, receiver)
|
|
if _, err := client.SendTransaction(t.Context(), "warp.example", "txn/1 with spaces", Transaction{
|
|
Origin: "remote.example", OriginServerTS: 1, PDUs: []json.RawMessage{},
|
|
}); err != nil {
|
|
t.Fatalf("SendTransaction: %v", err)
|
|
}
|
|
if got, _ := gotOrigin.Load().(string); got != "remote.example" {
|
|
t.Errorf("authenticated origin = %q", got)
|
|
}
|
|
|
|
// A request signed by a key the origin does not publish is refused.
|
|
stranger, _ := signing.Generate("")
|
|
forged := clientFor(t, "remote.example", stranger, receiver)
|
|
_, err := forged.SendTransaction(t.Context(), "warp.example", "2", Transaction{Origin: "remote.example"})
|
|
var httpErr *HTTPError
|
|
if !errors.As(err, &httpErr) || httpErr.Status != http.StatusUnauthorized {
|
|
t.Errorf("forged request = %v, want 401", err)
|
|
}
|
|
}
|
|
|
|
func TestClientRefusesWhileDisabledOrDenied(t *testing.T) {
|
|
r := newRemote(t, "remote.example")
|
|
key, _ := signing.Generate("")
|
|
enabled := false
|
|
client := clientFor(t, "warp.example", key, r)
|
|
client.enabled = func() bool { return enabled }
|
|
|
|
if _, err := client.Version(t.Context(), "remote.example"); !errors.Is(err, ErrDisabled) {
|
|
t.Errorf("Version while disabled = %v, want ErrDisabled", err)
|
|
}
|
|
enabled = true
|
|
client.policy = NewPolicy(nil, []string{"remote.example"})
|
|
if _, err := client.Version(t.Context(), "remote.example"); !errors.Is(err, ErrNotAllowed) {
|
|
t.Errorf("Version to a denied server = %v, want ErrNotAllowed", err)
|
|
}
|
|
if _, err := client.Version(t.Context(), "warp.example"); !errors.Is(err, ErrSelf) {
|
|
t.Errorf("Version to ourselves = %v, want ErrSelf", err)
|
|
}
|
|
if r.keyHits.Load() != 0 {
|
|
t.Error("a refused request reached the network")
|
|
}
|
|
}
|