fix(GWC-28): stamp gateway envelope sequence at write, not construction

CreateEnvelope stamped Sequence with an interlocked increment when the
envelope was built, so two concurrent InvokeAsync callers could take 1
and 2 and then enqueue in the order 2, 1 — non-monotonic on the wire,
breaking gateway.md's "monotonic per sender" contract. Benign today
(neither side validates inbound sequence, old GWC-10 still open) but it
would fault healthy sessions the moment worker-side validation lands.

WriteLoopAsync now stamps immediately before _writer.WriteAsync. It is
the outbound channel's single consumer (SingleReader = true), so wire
order and stamp order are the same thing by construction and
_nextSequence drops to a plain ulong with no interlocking. This mirrors
the worker's WRK-04 fix, which the gateway half never received.

Also adds TST-28: a [Theory] pinning that GatewayHello.MaxFrameBytes
carries the configured worker-frame maximum (default + 2 MiB override).
The adoption half is asserted only in the Windows-only worker suite, so
a regression to sending 0 — "older gateway, use default" to the worker —
would silently downgrade the negotiated IPC-02 limit with every CI test
still green. Mutation-checked (hard-coded 0 fails both cases).

Tests: WorkerClientTests.ConcurrentInvokesEmitStrictlyIncreasingSequences
OnTheWire (32 parallel invokes; failed 3/3 pre-fix) and
.StartAsync_SendsGatewayHelloWithConfiguredMaxFrameBytes.
This commit is contained in:
Joseph Doherty
2026-08-07 06:15:20 -04:00
parent 6092172694
commit f27eb28063
3 changed files with 90 additions and 5 deletions
@@ -40,7 +40,9 @@ public sealed class WorkerClient : IWorkerClient
private readonly ConcurrentDictionary<string, PendingCommand> _pendingCommands = new(StringComparer.Ordinal);
private readonly SemaphoreSlim _pendingCommandSlots;
private readonly CancellationTokenSource _stopCts = new();
private long _nextSequence;
// Touched only by WriteLoopAsync — the single consumer of _outboundEnvelopes — so it needs no
// interlocking. See WriteLoopAsync for why the stamp happens there rather than at construction.
private ulong _nextSequence;
private WorkerClientState _state;
private DateTimeOffset _lastHeartbeatAt;
private int? _processId;
@@ -404,6 +406,13 @@ public sealed class WorkerClient : IWorkerClient
{
await foreach (WorkerEnvelope envelope in _outboundEnvelopes.Reader.ReadAllAsync(_stopCts.Token).ConfigureAwait(false))
{
// GWC-28: stamp the sequence at the point of writing, not when the envelope is built.
// Stamping at construction let two concurrent InvokeAsync callers take 1 and 2 and then
// enqueue in the order 2, 1 — non-monotonic on the wire, breaking gateway.md's
// "monotonic per sender" contract. This loop is the channel's single consumer
// (SingleReader = true), so wire order and stamp order are the same thing here and
// _nextSequence needs no interlocking. Mirrors the worker's WRK-04 fix.
envelope.Sequence = unchecked(++_nextSequence);
await _writer.WriteAsync(envelope, _stopCts.Token).ConfigureAwait(false);
}
}
@@ -1072,11 +1081,13 @@ public sealed class WorkerClient : IWorkerClient
string correlationId,
Action<WorkerEnvelope> setBody)
{
// Sequence is deliberately left unset here: WriteLoopAsync stamps it immediately before the
// frame goes out, so the numbers are monotonic in wire order however the callers interleave
// between construction and enqueue (GWC-28, mirroring the worker's WRK-04 fix).
WorkerEnvelope envelope = new()
{
ProtocolVersion = _connection.FrameOptions.ProtocolVersion,
SessionId = SessionId,
Sequence = (ulong)Interlocked.Increment(ref _nextSequence),
CorrelationId = correlationId,
};
setBody(envelope);