From 981f849796512a549b12cfe2e00c4b7d17c67bd7 Mon Sep 17 00:00:00 2001
From: Aaron Stannard <aaron@petabridge.com>
Date: Sat, 9 May 2026 08:41:02 -0500
Subject: [PATCH] fix(tests): resolve passivation test race condition and
 missing DI registrations

Two issues prevented reliable CI on the passivation-abort integration tests:

1. Race condition in New_user_message_during_passivation_aborts_shutdown:
   AwaitAssertAsync was gating only on _fakeChatClient.CallCount == 1 (LLM
   called), then immediately checking TurnCount. Because the in-memory journal
   persists asynchronously, TurnCount could still be 0 at the moment CallCount
   first reaches 1. The fix polls with a witness probe until both CallCount == 1
   and TurnCount == 1 are true before the final subscriber re-join.

2. Missing DI registrations in LlmSessionTestBase: EffectivePolicyDefaults
   (required by ReminderManagerActor) and BackgroundJobDefinitionStore (required
   by BackgroundJobManagerActor) were not registered, causing actor initialization
   failures on every test in every LlmSessionTestBase subclass. Added both to
   ConfigureServices so the actors start cleanly across all session integration
   tests.
---
 .../Sessions/LlmSessionIntegrationTests.cs    | 26 +++++++++++++++----
 .../Sessions/LlmSessionTestBase.cs            |  3 +++
 2 files changed, 24 insertions(+), 5 deletions(-)

diff --git a/src/Netclaw.Actors.Tests/Sessions/LlmSessionIntegrationTests.cs b/src/Netclaw.Actors.Tests/Sessions/LlmSessionIntegrationTests.cs
index 7436518f..41e2b293 100644
--- a/src/Netclaw.Actors.Tests/Sessions/LlmSessionIntegrationTests.cs
+++ b/src/Netclaw.Actors.Tests/Sessions/LlmSessionIntegrationTests.cs
@@ -1417,11 +1417,27 @@ await sessionManager.Ask<SessionJoined>(new JoinSession
         }, TimeSpan.FromSeconds(3), cancellationToken: TestContext.Current.CancellationToken);
 
         Assert.Equal(sessionId, ack.SessionId);
-        await AwaitAssertAsync(() =>
-        {
-            Assert.Equal(1, _fakeChatClient.CallCount);
-            return Task.CompletedTask;
-        }, TimeSpan.FromSeconds(3), TimeSpan.FromMilliseconds(100), cancellationToken: TestContext.Current.CancellationToken);
+
+        // Poll until the turn is fully persisted. Checking CallCount alone is not
+        // sufficient — the in-memory journal persists asynchronously, so TurnCount
+        // can still be 0 at the moment CallCount first reaches 1. CallCount is
+        // intentionally NOT asserted inside the retry loop: if retries push it above
+        // the expected value, a strict equality check would loop forever rather than
+        // failing fast. Assert it once after the loop, when the actor is idle.
+        // JoinSession is idempotent for the same subscriber (Dictionary keyed by
+        // IActorRef), so repeated calls with the witness probe are safe.
+        var witness = CreateTestProbe("passivation-witness");
+        await AwaitAssertAsync(async () =>
+        {
+            var peek = await sessionManager.Ask<SessionJoined>(new JoinSession
+            {
+                SessionId = sessionId,
+                Subscriber = witness,
+                Filter = OutputFilter.TextOnly
+            }, TimeSpan.FromSeconds(1), cancellationToken: TestContext.Current.CancellationToken);
+            Assert.Equal(1, peek.TurnCount);
+        }, TimeSpan.FromSeconds(5), TimeSpan.FromMilliseconds(100), cancellationToken: TestContext.Current.CancellationToken);
+        Assert.Equal(1, _fakeChatClient.CallCount);
 
         var rejoined = await sessionManager.Ask<SessionJoined>(new JoinSession
         {
diff --git a/src/Netclaw.Actors.Tests/Sessions/LlmSessionTestBase.cs b/src/Netclaw.Actors.Tests/Sessions/LlmSessionTestBase.cs
index 4a9eba9e..59e6c83f 100644
--- a/src/Netclaw.Actors.Tests/Sessions/LlmSessionTestBase.cs
+++ b/src/Netclaw.Actors.Tests/Sessions/LlmSessionTestBase.cs
@@ -9,6 +9,7 @@
 using Microsoft.Extensions.DependencyInjection;
 using Microsoft.Extensions.Hosting;
 using Netclaw.Actors.Hosting;
+using Netclaw.Actors.Jobs;
 using Netclaw.Configuration;
 
 namespace Netclaw.Actors.Tests.Sessions;
@@ -30,6 +31,8 @@ protected sealed override void ConfigureServices(HostBuilderContext context, ISe
     {
         services.AddSingleton<IModelCapabilityResolver>(new FakeCapabilityResolver());
         services.AddTestNetclawPaths();
+        services.AddSingleton(SecurityPolicyDefaults.Resolve(null));
+        services.AddSingleton<BackgroundJobDefinitionStore>();
         ConfigureSessionServices(services);
         services.AddLlmSessionCompositeRecords();
     }