Implements PLAN.md Phase 7: the server half of end-to-end encryption.
Devices upload identity, one-time and fallback keys; other devices query
and claim them, across servers too, and every one-time key is handed out
exactly once. Cross-signing, signature uploads, device list tracking and
server-side key backup complete the set.
internal/keyserver is the component. Keys are stored in canonical JSON
so that re-uploading the same key is a byte comparison, and signatures
uploaded later are kept apart from the key they sign and merged in on
read, so a key's own JSON stays what its owner uploaded. Device keys must
be for the caller's own device; cross-signing keys must be the
uploader's, carry the right usage and, for self- and user-signing keys,
be signed by the master key. Signatures are verified against the key as
stored. Replacing a master key asks for the password; a first upload or
an upload of the keys already stored does not, as the spec allows. A
signature of another user's master key is shown only to the user who
made it and never federated.
Every write that changes what /keys/query answers moves the device list
stream in the same transaction, one position per device, and wakes the
syncs of everyone sharing a room. The stream is the sixth field of a sync
token; older tokens still parse. The syncapi works out device_lists
changed and left from the stream and from membership moves in the
window, skipping invites and knocks, and reports
device_one_time_keys_count and device_unused_fallback_key_types, the
latter as [] when empty. GET /keys/changes uses the same logic.
Over federation: /user/keys/query, /user/keys/claim and
/user/devices/{userId}, and m.device_list_update and m.signing_key_update
both ways. Remote keys are not cached; an update about a remote user
sharing a room moves them in the stream so local clients re-query.
Device display names are not sent to other servers. With federation off,
remote users are reported in failures at once.
Deleting a device deletes its keys, through a new userapi
DeviceObserver. Key backup keeps the better copy of each session by the
spec's rule, never reuses a deleted version's number, and answers writes
to a stale version with M_WRONG_ROOM_KEYS_VERSION.
Storage is migration 0008 for sqlite and postgres, with a PocketBase
stub. The Complement target sets gain the key, device list, fallback,
cross-signing and backup tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
139 lines
4.6 KiB
Go
139 lines
4.6 KiB
Go
package setup
|
|
|
|
import (
|
|
"net/http"
|
|
"time"
|
|
|
|
"github.com/dariuszjaniak/warp/server/internal/buildinfo"
|
|
"github.com/dariuszjaniak/warp/server/internal/httputil"
|
|
)
|
|
|
|
// SpecVersions is what GET /_matrix/client/versions reports.
|
|
//
|
|
// Keep it honest: a version belongs here once the endpoints a client
|
|
// actually uses from it work. Accounts, sessions, devices, profiles,
|
|
// account data, rooms, the event stream, media (Phase 6) and end-to-end
|
|
// encryption (Phase 7) all do, which is enough for a client to be used.
|
|
// Advertising v1.11 or later is also what makes a client use the
|
|
// authenticated media endpoints, the only download endpoints Warp serves
|
|
// by default. What is still missing is push (Phase 8); a client discovers
|
|
// that from the endpoint it needs rather than from this list, which is why
|
|
// its absence does not make a version listed here a lie.
|
|
var SpecVersions = []string{"v1.13", "v1.14"}
|
|
|
|
// UnstableFeatures advertises MSC support. Simplified Sliding Sync
|
|
// (MSC4186) is Phase 10.
|
|
var UnstableFeatures = map[string]bool{}
|
|
|
|
type versionsResponse struct {
|
|
Versions []string `json:"versions"`
|
|
UnstableFeatures map[string]bool `json:"unstable_features,omitempty"`
|
|
}
|
|
|
|
// wellKnownClient is the /.well-known/matrix/client document.
|
|
type wellKnownClient struct {
|
|
Homeserver wellKnownBaseURL `json:"m.homeserver"`
|
|
IdentityServer *wellKnownBaseURL `json:"m.identity_server,omitempty"`
|
|
}
|
|
|
|
type wellKnownBaseURL struct {
|
|
BaseURL string `json:"base_url"`
|
|
}
|
|
|
|
type wellKnownServer struct {
|
|
Server string `json:"m.server"`
|
|
}
|
|
|
|
type healthResponse struct {
|
|
Status string `json:"status"`
|
|
Version string `json:"version"`
|
|
Database string `json:"database"`
|
|
Error string `json:"error,omitempty"`
|
|
}
|
|
|
|
// routes registers every endpoint this server serves.
|
|
func (s *Server) routes() *httputil.Router {
|
|
r := httputil.NewRouter(s.metrics.Instrument)
|
|
|
|
r.HandleFunc("GET /_matrix/client/versions", s.handleVersions)
|
|
r.HandleFunc("GET /.well-known/matrix/client", s.handleWellKnownClient)
|
|
|
|
// The server-side delegation document is part of the federation
|
|
// surface, so it goes through the gate and disappears with it.
|
|
r.Handle("GET /.well-known/matrix/server", s.federation.ProtectFunc(s.handleWellKnownServer))
|
|
|
|
s.clientAPI.RegisterRoutes(r)
|
|
if s.fedAPI != nil {
|
|
s.fedAPI.RegisterRoutes(r)
|
|
}
|
|
s.adminAPI.RegisterRoutes(r)
|
|
|
|
r.HandleFunc("GET /healthz", s.handleHealth)
|
|
if s.cfg.Metrics.Enabled {
|
|
r.Handle("GET "+s.cfg.Metrics.Path, s.metrics.Handler())
|
|
}
|
|
return r
|
|
}
|
|
|
|
// handleVersions implements GET /_matrix/client/versions.
|
|
func (s *Server) handleVersions(w http.ResponseWriter, _ *http.Request) error {
|
|
httputil.CacheFor(w, time.Hour)
|
|
httputil.WriteJSON(w, http.StatusOK, versionsResponse{
|
|
Versions: SpecVersions,
|
|
UnstableFeatures: UnstableFeatures,
|
|
})
|
|
return nil
|
|
}
|
|
|
|
// handleWellKnownClient implements GET /.well-known/matrix/client, the
|
|
// document a client fetches to turn "@me:warp.example" into an API URL.
|
|
func (s *Server) handleWellKnownClient(w http.ResponseWriter, _ *http.Request) error {
|
|
doc := wellKnownClient{
|
|
Homeserver: wellKnownBaseURL{BaseURL: s.cfg.ClientBaseURL()},
|
|
}
|
|
if url := s.cfg.Global.WellKnown.IdentityServerBaseURL; url != "" {
|
|
doc.IdentityServer = &wellKnownBaseURL{BaseURL: url}
|
|
}
|
|
httputil.CacheFor(w, time.Hour)
|
|
httputil.WriteJSON(w, http.StatusOK, doc)
|
|
return nil
|
|
}
|
|
|
|
// handleWellKnownServer implements GET /.well-known/matrix/server. With no
|
|
// address configured there is nothing to delegate and the right answer is
|
|
// 404: other servers then fall back to SRV and A record resolution.
|
|
func (s *Server) handleWellKnownServer(w http.ResponseWriter, _ *http.Request) error {
|
|
address := s.cfg.Global.WellKnown.ServerAddress
|
|
if address == "" {
|
|
return httputil.NotFound("No server delegation is configured")
|
|
}
|
|
httputil.CacheFor(w, time.Hour)
|
|
httputil.WriteJSON(w, http.StatusOK, wellKnownServer{Server: address})
|
|
return nil
|
|
}
|
|
|
|
// handleHealth implements GET /healthz: a liveness and readiness check in
|
|
// one. It reports 503 when the database is unreachable, which is what a
|
|
// load balancer needs to stop sending traffic here.
|
|
func (s *Server) handleHealth(w http.ResponseWriter, r *http.Request) error {
|
|
resp := healthResponse{
|
|
Status: "ok",
|
|
Version: buildinfo.Version,
|
|
Database: s.db.Driver(),
|
|
}
|
|
status := http.StatusOK
|
|
|
|
ctx, cancel := contextWithTimeout(r.Context(), 2*time.Second)
|
|
defer cancel()
|
|
if err := s.db.Ping(ctx); err != nil {
|
|
resp.Status = "unavailable"
|
|
resp.Error = "database unreachable"
|
|
status = http.StatusServiceUnavailable
|
|
httputil.LoggerFrom(r.Context()).Error("health check failed", "error", err)
|
|
}
|
|
|
|
httputil.NoCache(w)
|
|
httputil.WriteJSON(w, status, resp)
|
|
return nil
|
|
}
|