diff --git a/multiplayer-next.md b/multiplayer-next.md index 38ca4f4a..5febe725 100644 --- a/multiplayer-next.md +++ b/multiplayer-next.md @@ -58,9 +58,10 @@ product policy are in [`docs/MATCHMAKING.md`](docs/MATCHMAKING.md). fencing; `server/agones` now submits and validates namespaced `GameServerAllocation` responses, including dynamic address/port data; `server/allocator` now requires provider allocation reconciliation into the - durable registry before returning an endpoint; durable roster publication - now verifies canonical join-authorisation signatures before exposing player - rows; live Agones integration remains. + durable registry before returning an endpoint; allocator-facing roster + publication now requires an allocated endpoint and verifies canonical + join-authorisation signatures before exposing player rows; live Agones + integration remains. - [ ] **IN PROGRESS:** Run the Go control plane against PostgreSQL/Redis with independently runnable API, matcher, allocator and maintenance roles. The `cmd/control-plane` API role now opens PostgreSQL, applies migrations, wires diff --git a/multiplayer-todo.md b/multiplayer-todo.md index fdb4e87a..f68e3815 100644 --- a/multiplayer-todo.md +++ b/multiplayer-todo.md @@ -1214,7 +1214,7 @@ the local/CI/community transport, not a silent production fallback. | 8.28 `[D:8.6,8.27]` | **IN PROGRESS.** Supervisor separates explicit process-ready from Agones Ready and never scrapes stdout; allocated mode refuses to mark Ready without a configured readiness probe | `server/supervisor/` tests prove Ready follows the probe and direct mode remains functional; `server_control.gd`, `agones_sdk.gd` and process-level smokes prove loopback `/ready`, `/health`, bearer-protected `/drain`, sidecar-shaped Health/Ready calls and drain admission fencing; detached-container and Health-reclaim integration remain | | 8.29 `[D:8.26,8.27]` | **IN PROGRESS.** Supervisor discovers and validates the Agones endpoint, propagates the actual dynamic `--port`, and exports `SDR_LISTEN_PORT`/`SDR_IP` only for Hosted-SDR while preserving an isolated ENet path | `server/supervisor/` tests cover invalid address/port rejection, dynamic port argument/env propagation and SDR-vs-ENet separation; real Agones dynamic/passthrough mapping, POP/cert/firewall/NAT and multi-match fixture remain | | 8.30 `[D:8.18,8.26,8.28,8.29]` | **IN PROGRESS.** Pure Go allocator filters Ready servers by region/build/protocol/transport, atomically claims one with idempotent allocation replay, and now owns the assignment-publication boundary; PostgreSQL adds durable GameServer registration and compatible `SKIP LOCKED` claims with request-digest fencing; `server/agones` submits and validates namespaced `GameServerAllocation` responses and dynamic endpoints; `server/allocator` reconciles provider success into durable state before exposing the endpoint | `server/domain/allocator.go`, `server/store/allocator_sql.go`, `server/agones/allocation.go`, `server/allocator/service.go`, `server/migrations/0004_allocator_registry.sql` and tests cover deterministic compatible selection, exhaustion, conflicting/identical allocation replay, unknown allocations, SQL claim ordering, invalid input, provider error/malformed response/IPv6 endpoint handling, durable-reconciliation failure isolation and assignment replay/conflict; `TestPostgreSQLAllocatorClaimReplayAndCapacityFence` now covers live registration/selection/replay/conflict/no-capacity when the disposable database gate is run; signed roster metadata, bounded cross-replica retry and live integration remain | -| 8.31 `[D:8.9,8.30]` | **IN PROGRESS.** Pure Go assignment gate requires Allocated state, exact allocation ID/match/server/region/build/protocol/transport compatibility, non-empty hosted endpoint and verified manifest signature before exposure; allocator publication cannot expose Ready state; durable roster persistence now verifies each canonical join-authorisation signature before publishing player rows | `server/domain/assignment.go`, `allocator.go`, `store/assignment_sql.go` plus adversarial fixtures cover early-connect, tampered signature/manifest, wrong compatibility, empty endpoint, unknown allocation, forged roster signature, valid signature and post-publication mutation rejection; Agones metadata watch, hosted-address registration, production signer and client-ticket publication remain | +| 8.31 `[D:8.9,8.30]` | **IN PROGRESS.** Pure Go assignment gate requires Allocated state, exact allocation ID/match/server/region/build/protocol/transport compatibility, non-empty hosted endpoint and verified manifest signature before exposure; allocator publication cannot expose Ready state; durable roster persistence now verifies each canonical join-authorisation signature before publishing player rows; allocator service gates roster publication on allocated state and endpoint presence | `server/domain/assignment.go`, `allocator.go`, `store/assignment_sql.go` plus adversarial fixtures cover early-connect, tampered signature/manifest, wrong compatibility, empty endpoint, unknown allocation, forged roster signature, valid signature, premature publication and post-publication mutation rejection; Agones metadata watch, hosted-address registration, production signer and client-ticket publication remain | | 8.32 `[D:8.2,8.26,8.30]` | **IN PROGRESS.** Provider-neutral FleetAutoscaler baseline preserves a two-process Ready buffer, caps warm capacity, and leaves Allocated scale-down independent of the Ready floor; Fleet image references remain digest-pinned for current/rollback pre-pull | `deploy/k8s/base/fleet-autoscaler.yaml` and manifest tests cover Fleet ownership, Buffer policy and floor/cap invariants; regional on-demand node pools/failure domains, pre-pull rollout, warm-allocation p95/p99 and N+1 certification remain | | 8.33 `[D:8.26,8.32]` | **IN PROGRESS.** Fleet scheduling now requires on-demand capacity and spreads Ready processes across zones with skew 1; the autoscaler preserves the two-process Ready floor | `deploy/k8s/base/fleet.yaml` and manifest tests reject interruptible placement and single-zone concentration structurally; regional node pools, forced node-loss testing and measured N+1 headroom remain | | 8.34 `[D:8.28,8.29]` | Native x86_64 benchmark of boot-to-process-ready and assignment-ready, p99 CPU/RSS/network and 60 Hz ticks; limits/node cap with 30% headroom | Measurements replace old estimates and certify density with no tick backlog | diff --git a/server/allocator/service.go b/server/allocator/service.go index fe96a5c1..dbceed35 100644 --- a/server/allocator/service.go +++ b/server/allocator/service.go @@ -18,12 +18,27 @@ type Durable interface { RecordProviderAllocation(context.Context, domain.Allocation, time.Time) (domain.Allocation, error) } +type RosterPublisher interface { + PublishRoster(context.Context, domain.Assignment, []domain.SignedJoinAuthorisation, func([]byte, []byte) bool) error +} + type Service struct { Provider Provider Durable Durable + Roster RosterPublisher Now func() time.Time } +func (s Service) PublishRoster(ctx context.Context, assignment domain.Assignment, roster []domain.SignedJoinAuthorisation, verify func([]byte, []byte) bool) error { + if s.Roster == nil { + return errNotConfigured + } + if assignment.Allocation.State != domain.ServerAllocated || assignment.Endpoint == "" { + return domain.ErrManifestRejected + } + return s.Roster.PublishRoster(ctx, assignment, roster, verify) +} + func (s Service) Allocate(ctx context.Context, request domain.AllocationRequest, labels map[string]string) (agones.AllocatedServer, error) { if s.Provider == nil || s.Durable == nil || s.Now == nil { return agones.AllocatedServer{}, errNotConfigured diff --git a/server/allocator/service_test.go b/server/allocator/service_test.go index 1e1f4aa9..84cac80a 100644 --- a/server/allocator/service_test.go +++ b/server/allocator/service_test.go @@ -27,6 +27,16 @@ type durableSpy struct { err error } +type rosterSpy struct { + calls int + err error +} + +func (r *rosterSpy) PublishRoster(_ context.Context, _ domain.Assignment, _ []domain.SignedJoinAuthorisation, _ func([]byte, []byte) bool) error { + r.calls++ + return r.err +} + func (d *durableSpy) RecordProviderAllocation(_ context.Context, allocation domain.Allocation, _ time.Time) (domain.Allocation, error) { d.calls++ d.allocation = allocation @@ -52,3 +62,16 @@ func TestServiceDoesNotReturnProviderResultAfterDurableFailure(t *testing.T) { t.Fatalf("result=%+v err=%v calls=%d", result, err, durable.calls) } } + +func TestServicePublishesRosterOnlyForAllocatedAssignment(t *testing.T) { + roster := &rosterSpy{} + service := Service{Roster: roster} + assignment := domain.Assignment{Allocation: domain.Allocation{State: domain.ServerAllocated}, Endpoint: "127.0.0.1:7777"} + if err := service.PublishRoster(context.Background(), assignment, []domain.SignedJoinAuthorisation{{Signature: []byte("sig")}}, func([]byte, []byte) bool { return true }); err != nil || roster.calls != 1 { + t.Fatalf("publish err=%v calls=%d", err, roster.calls) + } + assignment.Allocation.State = domain.ServerReady + if err := service.PublishRoster(context.Background(), assignment, nil, nil); err != domain.ErrManifestRejected || roster.calls != 1 { + t.Fatalf("premature publish err=%v calls=%d", err, roster.calls) + } +} diff --git a/server/store/assignment_sql.go b/server/store/assignment_sql.go index 31830b2a..74a1cd5f 100644 --- a/server/store/assignment_sql.go +++ b/server/store/assignment_sql.go @@ -32,6 +32,12 @@ type DurableAssignment struct { Revision uint64 } +type PostgresRosterStore struct{ DB *sql.DB } + +func (s PostgresRosterStore) PublishRoster(ctx context.Context, assignment domain.Assignment, roster []domain.SignedJoinAuthorisation, verify func([]byte, []byte) bool) error { + return SaveVerifiedAssignmentRoster(ctx, s.DB, assignment, roster, verify) +} + const AssignmentUpsertSQL = `INSERT INTO assignments (match_id, player_id, allocation_id, server_id, slot, region, client_build, protocol_version, transport, endpoint, join_authorisation, manifest_digest,