PR 4.5 — ReconnectSupervisor
State machine that drives GalaxyDriver's recovery from gw transport failure. Healthy → TransportLost → Reopening → Replaying → Healthy. Drivers report failure signals; the supervisor runs reopen + replay with capped exponential backoff (default 500ms → 30s) until both succeed. Files: - Runtime/ReconnectSupervisor.cs — state machine with snapshot, change event, last-error tracking, and a one-attempt-at-a-time recovery loop. Idempotent ReportTransportFailure: repeated failure reports during an in-flight recovery do not spawn parallel loops. Reopen + replay are caller-supplied callbacks (the driver injects them in the wire-up PR); reopen re-Registers the gw session, replay re-establishes every active subscription via gw's ReplaySubscriptionsCommand (mxaccessgw issue gw-3) or the SubscribeBulk fallback. Dispose cancels the loop cleanly. - Public StateTransition record + IsDegraded predicate the driver maps to DriverState.Degraded for health snapshots. Wiring (GalaxyDriver subscribes the supervisor to its EventPump's transport-failure signal, exposes IsDegraded through GetHealth(), routes reopen/replay callbacks through GalaxyMxSession + SubscriptionRegistry) lands in PR 4.W to avoid conflict with the parallel host-probe track (PR 4.7) and align the wire-up with the rest of Phase 4's plumbing. 9 supervisor tests (full state-machine traversal, retry-until-success on both reopen and replay failures, idempotent failure reports, last-error propagation, Dispose mid-recovery, post-dispose throws, fast-path Healthy WaitForHealthy). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Joseph Doherty
@@ -0,0 +1,173 @@
|
||||
using Shouldly;
|
||||
using Xunit;
|
||||
using ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Runtime;
|
||||
|
||||
namespace ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Tests.Runtime;
|
||||
|
||||
/// <summary>
|
||||
/// Tests for <see cref="ReconnectSupervisor"/>'s state machine + backoff. Each
|
||||
/// scenario drives the supervisor with controllable reopen/replay callbacks and
|
||||
/// observes the resulting state transitions.
|
||||
/// </summary>
|
||||
public sealed class ReconnectSupervisorTests
|
||||
{
|
||||
private const int WaitMs = 2_000;
|
||||
|
||||
private static ReconnectOptions FastBackoff() =>
|
||||
new(InitialBackoffOverride: TimeSpan.FromMilliseconds(5),
|
||||
MaxBackoffOverride: TimeSpan.FromMilliseconds(20));
|
||||
|
||||
[Fact]
|
||||
public void InitialState_IsHealthy()
|
||||
{
|
||||
using var sup = new ReconnectSupervisor(_ => Task.CompletedTask, _ => Task.CompletedTask, FastBackoff());
|
||||
sup.CurrentState.ShouldBe(ReconnectSupervisor.State.Healthy);
|
||||
sup.IsDegraded.ShouldBeFalse();
|
||||
sup.LastError.ShouldBeNull();
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task ReportTransportFailure_DrivesThroughReopenReplay_BackToHealthy()
|
||||
{
|
||||
var transitions = new List<StateTransition>();
|
||||
var lockObj = new object();
|
||||
|
||||
using var sup = new ReconnectSupervisor(
|
||||
reopen: _ => Task.CompletedTask,
|
||||
replay: _ => Task.CompletedTask,
|
||||
options: FastBackoff());
|
||||
|
||||
sup.StateChanged += (_, t) => { lock (lockObj) transitions.Add(t); };
|
||||
|
||||
sup.ReportTransportFailure(new IOException("transport drop"));
|
||||
await sup.WaitForHealthyAsync(new CancellationTokenSource(WaitMs).Token);
|
||||
|
||||
// Expected sequence: Healthy → TransportLost → Reopening → Replaying → Healthy.
|
||||
IReadOnlyList<StateTransition> snapshot;
|
||||
lock (lockObj) snapshot = [.. transitions];
|
||||
var states = snapshot.Select(t => t.Next).ToArray();
|
||||
|
||||
states.ShouldContain(ReconnectSupervisor.State.TransportLost);
|
||||
states.ShouldContain(ReconnectSupervisor.State.Reopening);
|
||||
states.ShouldContain(ReconnectSupervisor.State.Replaying);
|
||||
states[^1].ShouldBe(ReconnectSupervisor.State.Healthy);
|
||||
sup.IsDegraded.ShouldBeFalse();
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task ReopenFailure_RetriesUntilSuccess_StaysInReopeningBetweenAttempts()
|
||||
{
|
||||
var attempts = 0;
|
||||
using var sup = new ReconnectSupervisor(
|
||||
reopen: _ => { attempts++; return attempts < 3 ? Task.FromException(new IOException("nope")) : Task.CompletedTask; },
|
||||
replay: _ => Task.CompletedTask,
|
||||
options: FastBackoff());
|
||||
|
||||
sup.ReportTransportFailure(new IOException("kick off"));
|
||||
await sup.WaitForHealthyAsync(new CancellationTokenSource(WaitMs).Token);
|
||||
|
||||
attempts.ShouldBe(3);
|
||||
sup.CurrentState.ShouldBe(ReconnectSupervisor.State.Healthy);
|
||||
sup.LastError.ShouldBeNull(); // cleared on Healthy transition
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task ReplayFailure_RetriesEntireCycle()
|
||||
{
|
||||
var reopens = 0;
|
||||
var replays = 0;
|
||||
using var sup = new ReconnectSupervisor(
|
||||
reopen: _ => { reopens++; return Task.CompletedTask; },
|
||||
replay: _ => { replays++; return replays < 2 ? Task.FromException(new IOException("replay nope")) : Task.CompletedTask; },
|
||||
options: FastBackoff());
|
||||
|
||||
sup.ReportTransportFailure(new IOException("kick off"));
|
||||
await sup.WaitForHealthyAsync(new CancellationTokenSource(WaitMs).Token);
|
||||
|
||||
// First cycle: reopen succeeds, replay fails. Second cycle: both succeed.
|
||||
reopens.ShouldBe(2);
|
||||
replays.ShouldBe(2);
|
||||
sup.CurrentState.ShouldBe(ReconnectSupervisor.State.Healthy);
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task RepeatedFailureReports_DuringRecovery_DoNotSpawnParallelLoops()
|
||||
{
|
||||
var attempts = 0;
|
||||
using var sup = new ReconnectSupervisor(
|
||||
reopen: async ct =>
|
||||
{
|
||||
attempts++;
|
||||
await Task.Delay(50, ct).ConfigureAwait(false);
|
||||
},
|
||||
replay: _ => Task.CompletedTask,
|
||||
options: FastBackoff());
|
||||
|
||||
sup.ReportTransportFailure(new IOException("first"));
|
||||
// Fire several more reports while reopen is in flight.
|
||||
for (var i = 0; i < 5; i++)
|
||||
{
|
||||
sup.ReportTransportFailure(new IOException($"rapid-{i}"));
|
||||
}
|
||||
|
||||
await sup.WaitForHealthyAsync(new CancellationTokenSource(WaitMs).Token);
|
||||
|
||||
// One Reopen call regardless of how many failures arrived during recovery.
|
||||
attempts.ShouldBe(1);
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task LastError_ReflectsMostRecentFailureCause()
|
||||
{
|
||||
using var sup = new ReconnectSupervisor(
|
||||
reopen: _ => Task.FromException(new IOException("reopen broke")),
|
||||
replay: _ => Task.CompletedTask,
|
||||
options: new ReconnectOptions(
|
||||
InitialBackoffOverride: TimeSpan.FromMilliseconds(5),
|
||||
MaxBackoffOverride: TimeSpan.FromMilliseconds(10)));
|
||||
|
||||
sup.ReportTransportFailure(new IOException("initial"));
|
||||
|
||||
// Allow the loop to attempt at least twice.
|
||||
await Task.Delay(100);
|
||||
sup.LastError.ShouldNotBeNull();
|
||||
sup.LastError.ShouldContain("reopen broke"); // updates from the loop's failed reopen attempts
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task Dispose_CancelsRunningRecoveryLoop_Cleanly()
|
||||
{
|
||||
var cancelled = false;
|
||||
var sup = new ReconnectSupervisor(
|
||||
reopen: async ct =>
|
||||
{
|
||||
try { await Task.Delay(10_000, ct).ConfigureAwait(false); }
|
||||
catch (OperationCanceledException) { cancelled = true; throw; }
|
||||
},
|
||||
replay: _ => Task.CompletedTask,
|
||||
options: FastBackoff());
|
||||
|
||||
sup.ReportTransportFailure(new IOException("kick off"));
|
||||
await Task.Delay(50); // let the loop start the long reopen
|
||||
Should.NotThrow(() => sup.Dispose());
|
||||
cancelled.ShouldBeTrue();
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void ReportTransportFailure_AfterDispose_Throws()
|
||||
{
|
||||
var sup = new ReconnectSupervisor(_ => Task.CompletedTask, _ => Task.CompletedTask, FastBackoff());
|
||||
sup.Dispose();
|
||||
Should.Throw<ObjectDisposedException>(() => sup.ReportTransportFailure(new IOException("x")));
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public async Task WaitForHealthy_ReturnsImmediately_WhenAlreadyHealthy()
|
||||
{
|
||||
using var sup = new ReconnectSupervisor(_ => Task.CompletedTask, _ => Task.CompletedTask, FastBackoff());
|
||||
// No failure reported — should be Healthy from the start.
|
||||
var deadline = DateTime.UtcNow.AddMilliseconds(50);
|
||||
await sup.WaitForHealthyAsync(new CancellationTokenSource(50).Token);
|
||||
DateTime.UtcNow.ShouldBeLessThan(deadline.AddMilliseconds(100)); // returned promptly
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user