Files
ScadaBridge/tests/ZB.MOM.WW.ScadaBridge.Communication.Tests/PendingDeploymentPurgeActorTests.cs
T
Joseph Doherty 06f2df4f89 feat(deploy): wire periodic PendingDeployment purge + SQL Server same-id re-stage test
Notify-and-fetch follow-ups:

- PendingDeploymentPurgeActor: a central cluster singleton (not
  readiness-gated, best-effort) that sweeps expired PendingDeployment
  staging rows on CommunicationOptions.PendingDeploymentPurgeInterval
  (default 1h). Modeled on the kpi-history-recorder pattern: self-scheduling
  timer, per-tick DI scope -> IDeploymentManagerRepository, continue-on-error.
  Wired in AkkaHostedService.RegisterCentralActors (manager + proxy + drain);
  resolves the deferred TODO in DeploymentService. Correctness never depends
  on it (supersession bounds rows to <=1/instance; the fetch endpoint enforces
  the TTL), so it is deliberately absent from RequiredSingletonsHealthCheck.

- SQL Server integration test for StagePendingIfAbsentAsync re-staging an
  instance's OWN DeploymentId over an expired row against the real UNIQUE
  index on DeploymentId — confirms EF orders DELETE before INSERT in one
  SaveChanges (SQLite's constraint timing differs from SQL Server's). Plus
  a same-instance supersession variant on real SQL Server.

Tests: 2 TestKit actor tests + 2 SQL Server integration tests (both ran
green against the infra MSSQL container); 235 Communication + 15
PendingDeployment tests pass; Host builds 0 warnings.
2026-06-26 23:19:29 -04:00

99 lines
4.5 KiB
C#

using Akka.Actor;
using Akka.TestKit.Xunit2;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Logging.Abstractions;
using Microsoft.Extensions.Options;
using NSubstitute;
using ZB.MOM.WW.ScadaBridge.Commons.Interfaces.Repositories;
using ZB.MOM.WW.ScadaBridge.Communication.Actors;
namespace ZB.MOM.WW.ScadaBridge.Communication.Tests;
/// <summary>
/// Tests for <see cref="PendingDeploymentPurgeActor"/> — the notify-and-fetch central
/// singleton that sweeps expired PendingDeployment staging rows. The actor is thin glue
/// over <c>IDeploymentManagerRepository.PurgeExpiredPendingDeploymentsAsync</c> (whose
/// row-level semantics are covered by the repository tests); these tests pin the actor's
/// own policy: it opens a per-tick DI scope, calls the purge with a now-ish threshold,
/// and — being best-effort hygiene — survives a throwing purge and keeps ticking.
/// </summary>
/// <remarks>
/// The substitute's return is driven by a closure with an <see cref="System.Threading.Interlocked"/>
/// counter rather than NSubstitute's <c>ReceivedCalls()</c>, because the repeating
/// scheduler timer records calls on a background thread concurrently with the test
/// thread's assertions — querying NSubstitute's call list under that concurrency can
/// throw "collection was modified". A fenced counter is safe to read while the timer runs.
/// </remarks>
public class PendingDeploymentPurgeActorTests : TestKit
{
private IActorRef CreateActor(IDeploymentManagerRepository repo, TimeSpan interval)
{
var services = new ServiceCollection();
// Mirror AddConfigurationDatabase: IDeploymentManagerRepository is scoped, so the
// actor opens a fresh scope per tick and resolves the repository there.
services.AddScoped(_ => repo);
var sp = services.BuildServiceProvider();
var options = Options.Create(new CommunicationOptions { PendingDeploymentPurgeInterval = interval });
return Sys.ActorOf(Props.Create(() => new PendingDeploymentPurgeActor(
sp, options, NullLogger<PendingDeploymentPurgeActor>.Instance)));
}
[Fact]
public void Tick_PurgesExpiredRows_WithNowThreshold()
{
var calls = 0;
DateTimeOffset captured = default;
var repo = Substitute.For<IDeploymentManagerRepository>();
repo.PurgeExpiredPendingDeploymentsAsync(Arg.Any<DateTimeOffset>(), Arg.Any<CancellationToken>())
.Returns(ci =>
{
captured = ci.Arg<DateTimeOffset>();
Interlocked.Increment(ref calls);
return 3;
});
// Large interval so the PreStart timer never fires during the test — only the
// single manual tick below drives a purge, making the count deterministic.
var actor = CreateActor(repo, TimeSpan.FromHours(1));
var before = DateTimeOffset.UtcNow;
actor.Tell(PendingDeploymentPurgeActor.PurgeTick.Instance);
AwaitAssert(
() => Assert.Equal(1, Volatile.Read(ref calls)),
duration: TimeSpan.FromSeconds(3),
interval: TimeSpan.FromMilliseconds(50));
// The purge threshold is "now" — bounded by the window around the tick. Reading
// `captured` after the fenced counter read is safe (Interlocked establishes the
// happens-before with the write that precedes the increment).
var after = DateTimeOffset.UtcNow;
Assert.InRange(captured, before, after);
}
[Fact]
public void Tick_PurgeThrows_ActorSurvives_AndRetriesOnNextInterval()
{
var calls = 0;
var repo = Substitute.For<IDeploymentManagerRepository>();
repo.PurgeExpiredPendingDeploymentsAsync(Arg.Any<DateTimeOffset>(), Arg.Any<CancellationToken>())
.Returns<int>(_ =>
{
Interlocked.Increment(ref calls);
throw new InvalidOperationException("simulated purge failure");
});
// Fast interval so the self-scheduling timer fires repeatedly; the actor must
// swallow each per-tick exception and keep ticking (best-effort hygiene — a
// crash here would drop the singleton until failover).
CreateActor(repo, TimeSpan.FromMilliseconds(100));
AwaitAssert(
() => Assert.True(Volatile.Read(ref calls) >= 2,
$"expected >= 2 ticks despite failures, got {Volatile.Read(ref calls)}"),
duration: TimeSpan.FromSeconds(5),
interval: TimeSpan.FromMilliseconds(50));
}
}