docs/v2/lmx-followups.md

@@ -77,18 +77,36 @@ drive a full OPC UA session with username/password, then read an
 `IHostConnectivityProbe`-style "whoami" node to verify the role surfaced).
 That needs a test-only address-space node and is a separate PR.
 
-## 5. Full Galaxy live-service smoke test against the merged v2 stack
+## 5. Full Galaxy live-service smoke test against the merged v2 stack — **IN PROGRESS (PRs 36 + 37)**
 
-**Status**: Individual pieces have live smoke tests (PR 5 MXAccess, PR 13
-probe manager, PR 14 alarm tracker), but the full loop — OPC UA client →
-`OtOpcUaServer` → `GalaxyProxyDriver` (in-process) → named-pipe to
-Galaxy.Host subprocess → live MXAccess runtime → real Galaxy objects — has
-no single end-to-end smoke test.
+PR 36 shipped the prerequisites helper (`AvevaPrerequisites`) that probes
+every dependency a live smoke test needs and produces actionable skip
+messages.
 
-**To do**:
-- Test that spawns the full topology, discovers a deployed Galaxy object,
-  subscribes to one of its attributes, writes a value back, and asserts the
-  write round-tripped through MXAccess. Skip when ArchestrA isn't running.
+PR 37 shipped the live-stack smoke test project structure:
+`tests/Driver.Galaxy.Proxy.Tests/LiveStack/` with `LiveStackFixture` (connects
+to the *already-running* `OtOpcUaGalaxyHost` Windows service via named pipe;
+never spawns the Host process) and `LiveStackSmokeTests` covering:
+
+- Fixture initializes successfully (IPC handshake succeeds end-to-end).
+- Driver reports `DriverState.Healthy` post-handshake.
+- `DiscoverAsync` returns at least one variable from the live Galaxy.
+- `GetHostStatuses` reports at least one Platform/AppEngine host.
+- `ReadAsync` on a discovered variable round-trips through
+  Proxy → Host pipe → MXAccess → back without a BadInternalError.
+
+Shared secret + pipe name resolve from `OTOPCUA_GALAXY_SECRET` /
+`OTOPCUA_GALAXY_PIPE` env vars, falling back to reading the service's
+registry-stored Environment values (requires elevated test host).
+
+**Remaining**:
+- Install + run the `OtOpcUaGalaxyHost` + `OtOpcUa` services on the dev box
+  (`scripts/install/Install-Services.ps1`) so the skip-on-unready tests
+  actually execute and the smoke PR lands green.
+- Subscribe-and-receive-data-change fact (needs a known tag that actually
+  ticks; deferred until operators confirm a scratch tag exists).
+- Write-and-roundtrip fact (needs a test-only UDA or agreed scratch tag
+  so we can't accidentally mutate a process-critical value).
 
 ## 6. Second driver instance on the same server — **DONE (PR 32)**
 

tests/ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests/LiveStack/LiveStackConfig.cs

@@ -0,0 +1,75 @@
+using System.Runtime.InteropServices;
+using System.Runtime.Versioning;
+using Microsoft.Win32;
+
+namespace ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests.LiveStack;
+
+/// <summary>
+///     Resolves the pipe name + shared secret the live <see cref="GalaxyProxyDriver"/> needs
+///     to connect to a running <c>OtOpcUaGalaxyHost</c> Windows service. Two sources are
+///     consulted, first match wins:
+///     <list type="number">
+///         <item>Explicit env vars (<c>OTOPCUA_GALAXY_PIPE</c>, <c>OTOPCUA_GALAXY_SECRET</c>) — lets CI / benchwork override.</item>
+///         <item>The service's per-process <c>Environment</c> registry values under
+///               <c>HKLM\SYSTEM\CurrentControlSet\Services\OtOpcUaGalaxyHost</c> — what
+///               <c>Install-Services.ps1</c> writes at install time. Requires the test to run as a
+///               principal with read access to that registry key (typically Administrators).</item>
+///     </list>
+/// </summary>
+/// <remarks>
+///     Explicitly NOT baked-in-to-source: the shared secret is rotated per install (the
+///     installer generates 32 random bytes and stores the base64 string). A hard-coded secret
+///     in tests would diverge from production the moment someone re-installed the service.
+/// </remarks>
+public sealed record LiveStackConfig(string PipeName, string SharedSecret, string? Source)
+{
+    public const string EnvPipeName = "OTOPCUA_GALAXY_PIPE";
+    public const string EnvSharedSecret = "OTOPCUA_GALAXY_SECRET";
+    public const string ServiceRegistryKey =
+        @"SYSTEM\CurrentControlSet\Services\OtOpcUaGalaxyHost";
+    public const string DefaultPipeName = "OtOpcUaGalaxy";
+
+    public static LiveStackConfig? Resolve()
+    {
+        var envPipe = Environment.GetEnvironmentVariable(EnvPipeName);
+        var envSecret = Environment.GetEnvironmentVariable(EnvSharedSecret);
+        if (!string.IsNullOrWhiteSpace(envPipe) && !string.IsNullOrWhiteSpace(envSecret))
+            return new LiveStackConfig(envPipe, envSecret, "env vars");
+
+        if (!RuntimeInformation.IsOSPlatform(OSPlatform.Windows))
+            return null;
+
+        return FromServiceRegistry();
+    }
+
+    [SupportedOSPlatform("windows")]
+    private static LiveStackConfig? FromServiceRegistry()
+    {
+        try
+        {
+            using var key = Registry.LocalMachine.OpenSubKey(ServiceRegistryKey);
+            if (key is null) return null;
+            var env = key.GetValue("Environment") as string[];
+            if (env is null || env.Length == 0) return null;
+
+            string? pipe = null, secret = null;
+            foreach (var line in env)
+            {
+                var eq = line.IndexOf('=');
+                if (eq <= 0) continue;
+                var name = line[..eq];
+                var value = line[(eq + 1)..];
+                if (name.Equals(EnvPipeName, StringComparison.OrdinalIgnoreCase)) pipe = value;
+                else if (name.Equals(EnvSharedSecret, StringComparison.OrdinalIgnoreCase)) secret = value;
+            }
+
+            if (string.IsNullOrWhiteSpace(secret)) return null;
+            return new LiveStackConfig(pipe ?? DefaultPipeName, secret, "service registry");
+        }
+        catch
+        {
+            // Access denied / key missing / malformed — caller gets null and surfaces a Skip.
+            return null;
+        }
+    }
+}

tests/ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests/LiveStack/LiveStackFixture.cs

@@ -0,0 +1,120 @@
+using System.Threading;
+using System.Threading.Tasks;
+using Xunit;
+using ZB.MOM.WW.OtOpcUa.Driver.Galaxy.TestSupport;
+
+namespace ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests.LiveStack;
+
+/// <summary>
+///     Connects a single <see cref="GalaxyProxyDriver"/> to the already-running
+///     <c>OtOpcUaGalaxyHost</c> Windows service for the lifetime of a test class. Uses
+///     <see cref="AvevaPrerequisites"/> to decide whether to proceed; on failure,
+///     <see cref="SkipReason"/> is populated and each test calls <see cref="SkipIfUnavailable"/>
+///     to translate that into <c>Assert.Skip</c>.
+/// </summary>
+/// <remarks>
+///     <para>
+///         <b>Does NOT spawn the Host process.</b> Production deploys <c>OtOpcUaGalaxyHost</c>
+///         as a standalone Windows service — spawning a second instance from a test would
+///         bypass the COM-apartment + service-account setup and fail differently than
+///         production (see <c>project_galaxy_host_service.md</c> memory).
+///     </para>
+///     <para>
+///         <b>Shared-secret handling</b>: read from <see cref="LiveStackConfig"/> — env vars
+///         first, then the service's registry-stored <c>Environment</c> values. Requires
+///         the test process to have read access to
+///         <c>HKLM\SYSTEM\CurrentControlSet\Services\OtOpcUaGalaxyHost</c>; on a dev box
+///         that typically means running the test host elevated, or exporting
+///         <c>OTOPCUA_GALAXY_SECRET</c> out-of-band.
+///     </para>
+/// </remarks>
+public sealed class LiveStackFixture : IAsyncLifetime
+{
+    public GalaxyProxyDriver? Driver { get; private set; }
+
+    public string? SkipReason { get; private set; }
+
+    public PrerequisiteReport? PrerequisiteReport { get; private set; }
+
+    public LiveStackConfig? Config { get; private set; }
+
+    public async ValueTask InitializeAsync()
+    {
+        // 1. AVEVA + OtOpcUa service state — actionable diagnostic if anything is missing.
+        using var cts = new CancellationTokenSource(TimeSpan.FromSeconds(10));
+        PrerequisiteReport = await AvevaPrerequisites.CheckAllAsync(
+            new AvevaPrerequisites.Options { CheckGalaxyHostPipe = true, CheckHistorian = false },
+            cts.Token);
+
+        if (!PrerequisiteReport.IsLivetestReady)
+        {
+            SkipReason = PrerequisiteReport.SkipReason;
+            return;
+        }
+
+        // 2. Secret / pipe-name resolution. If the service is running but we can't discover its
+        //    env vars from registry (non-elevated test host), a clear message beats a silent
+        //    connect-rejected failure 10 seconds later.
+        Config = LiveStackConfig.Resolve();
+        if (Config is null)
+        {
+            SkipReason =
+                $"Cannot resolve shared secret. Set {LiveStackConfig.EnvSharedSecret} (and optionally " +
+                $"{LiveStackConfig.EnvPipeName}) in the environment, or run the test host elevated so it " +
+                $"can read HKLM\\{LiveStackConfig.ServiceRegistryKey}\\Environment.";
+            return;
+        }
+
+        // 3. Connect. InitializeAsync does the pipe connect + handshake; a 5-second
+        //    ConnectTimeout gives enough headroom for a service that just started.
+        Driver = new GalaxyProxyDriver(new GalaxyProxyOptions
+        {
+            DriverInstanceId = "live-stack-smoke",
+            PipeName = Config.PipeName,
+            SharedSecret = Config.SharedSecret,
+            ConnectTimeout = TimeSpan.FromSeconds(5),
+        });
+
+        try
+        {
+            await Driver.InitializeAsync(driverConfigJson: "{}", CancellationToken.None);
+        }
+        catch (Exception ex)
+        {
+            SkipReason =
+                $"Connected to named pipe '{Config.PipeName}' but GalaxyProxyDriver.InitializeAsync failed: " +
+                $"{ex.GetType().Name}: {ex.Message}. Common causes: shared secret mismatch (rotated after last install), " +
+                $"service account SID not in pipe ACL (installer sets OTOPCUA_ALLOWED_SID to the service account — " +
+                $"test must run as that user), or Host's backend couldn't connect to ZB.";
+            Driver.Dispose();
+            Driver = null;
+            return;
+        }
+    }
+
+    public async ValueTask DisposeAsync()
+    {
+        if (Driver is not null)
+        {
+            try { await Driver.ShutdownAsync(CancellationToken.None); } catch { /* best-effort */ }
+            Driver.Dispose();
+        }
+    }
+
+    /// <summary>
+    ///     Translate <see cref="SkipReason"/> into <c>Assert.Skip</c>. Tests call this at the
+    ///     top of every fact so a fixture init failure shows up as a cleanly-skipped test with
+    ///     the full prerequisites report, not a cascading NullReferenceException on
+    ///     <see cref="Driver"/>.
+    /// </summary>
+    public void SkipIfUnavailable()
+    {
+        if (SkipReason is not null) Assert.Skip(SkipReason);
+    }
+}
+
+[CollectionDefinition(Name)]
+public sealed class LiveStackCollection : ICollectionFixture<LiveStackFixture>
+{
+    public const string Name = "LiveStack";
+}

tests/ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests/LiveStack/LiveStackSmokeTests.cs

@@ -0,0 +1,147 @@
+using System.Collections.Generic;
+using System.Linq;
+using System.Threading;
+using System.Threading.Tasks;
+using Shouldly;
+using Xunit;
+using ZB.MOM.WW.OtOpcUa.Core.Abstractions;
+
+namespace ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests.LiveStack;
+
+/// <summary>
+///     End-to-end smoke against the installed <c>OtOpcUaGalaxyHost</c> Windows service.
+///     Closes LMX follow-up #5 — exercises the full topology: <see cref="GalaxyProxyDriver"/>
+///     in-process → named-pipe IPC → <c>OtOpcUaGalaxyHost</c> service → <c>MxAccessGalaxyBackend</c> →
+///     live MXAccess runtime → real Galaxy objects + attributes.
+/// </summary>
+/// <remarks>
+///     <para>
+///         <b>Preconditions</b> (all checked by <see cref="LiveStackFixture"/>, surfaced via
+///         <c>Assert.Skip</c> when missing):
+///     </para>
+///     <list type="bullet">
+///         <item>AVEVA System Platform installed + Platform deployed.</item>
+///         <item><c>aaBootstrap</c> / <c>aaGR</c> / <c>NmxSvc</c> / <c>MSSQLSERVER</c> running.</item>
+///         <item>MXAccess COM server registered.</item>
+///         <item>ZB database exists with at least one deployed gobject.</item>
+///         <item><c>OtOpcUaGalaxyHost</c> service installed + running (named pipe accepting connections).</item>
+///         <item>Shared secret discoverable via <c>OTOPCUA_GALAXY_SECRET</c> env var or the
+///               service's registry Environment values (test host typically needs to be elevated
+///               to read the latter).</item>
+///         <item>Test process runs as the account listed in the service's pipe ACL
+///               (<c>OTOPCUA_ALLOWED_SID</c>, typically the service account per decision #76).</item>
+///     </list>
+///     <para>
+///         Tests here are deliberately read-only. Writes against live Galaxy attributes are a
+///         separate concern — they need a test-only UDA or an agreed scratch tag so they can't
+///         accidentally mutate a process-critical value. Adding a write test is a follow-up
+///         PR that reuses this fixture.
+///     </para>
+/// </remarks>
+[Trait("Category", "LiveGalaxy")]
+[Collection(LiveStackCollection.Name)]
+public sealed class LiveStackSmokeTests(LiveStackFixture fixture)
+{
+    [Fact]
+    public void Fixture_initialized_successfully()
+    {
+        fixture.SkipIfUnavailable();
+        // If the fixture init succeeded, Driver is non-null and InitializeAsync completed.
+        // This is the cheapest possible assertion that the IPC handshake worked end-to-end;
+        // every other test in this class depends on it.
+        fixture.Driver.ShouldNotBeNull();
+        fixture.Config.ShouldNotBeNull();
+        fixture.PrerequisiteReport.ShouldNotBeNull();
+        fixture.PrerequisiteReport!.IsLivetestReady.ShouldBeTrue(fixture.PrerequisiteReport.SkipReason);
+    }
+
+    [Fact]
+    public void Driver_reports_Healthy_after_IPC_handshake()
+    {
+        fixture.SkipIfUnavailable();
+        var health = fixture.Driver!.GetHealth();
+        health.State.ShouldBe(DriverState.Healthy,
+            $"Expected Healthy after successful IPC connect; Reason={health.LastError}");
+    }
+
+    [Fact]
+    public async Task DiscoverAsync_returns_at_least_one_variable_from_live_galaxy()
+    {
+        fixture.SkipIfUnavailable();
+        var builder = new CapturingAddressSpaceBuilder();
+        using var cts = new CancellationTokenSource(TimeSpan.FromSeconds(30));
+        await fixture.Driver!.DiscoverAsync(builder, cts.Token);
+
+        builder.Variables.Count.ShouldBeGreaterThan(0,
+            "Live Galaxy has > 0 deployed objects per the prereq check — at least one variable must be discovered. " +
+            "Zero usually means the Host couldn't read ZB (check OTOPCUA_GALAXY_ZB_CONN in the service Environment).");
+
+        // Every discovered attribute must carry a non-empty FullName so the OPC UA server can
+        // route reads/writes back. Regression guard — PR 19 normalized this across drivers.
+        builder.Variables.ShouldAllBe(v => !string.IsNullOrEmpty(v.AttributeInfo.FullName));
+    }
+
+    [Fact]
+    public void GetHostStatuses_reports_at_least_one_platform()
+    {
+        fixture.SkipIfUnavailable();
+        var statuses = fixture.Driver!.GetHostStatuses();
+        statuses.Count.ShouldBeGreaterThan(0,
+            "Live Galaxy must report at least one Platform/AppEngine host via IHostConnectivityProbe. " +
+            "Zero means the Host's probe loop hasn't completed its first tick or the Platform isn't deployed locally.");
+
+        // Host names are driver-opaque to the Core but non-empty by contract.
+        statuses.ShouldAllBe(h => !string.IsNullOrEmpty(h.HostName));
+    }
+
+    [Fact]
+    public async Task Can_read_a_discovered_variable_from_live_galaxy()
+    {
+        fixture.SkipIfUnavailable();
+        var builder = new CapturingAddressSpaceBuilder();
+        using var cts = new CancellationTokenSource(TimeSpan.FromSeconds(30));
+        await fixture.Driver!.DiscoverAsync(builder, cts.Token);
+        builder.Variables.Count.ShouldBeGreaterThan(0);
+
+        // Pick the first discovered variable. Read-only smoke — we don't assert on Value,
+        // only that a ReadAsync round-trip through Proxy → Host pipe → MXAccess → back
+        // returns a snapshot with a non-BadInternalError status. Galaxy attributes default to
+        // Uncertain quality until the Engine's first scan publishes them, which is fine here.
+        var full = builder.Variables[0].AttributeInfo.FullName;
+        var snapshots = await fixture.Driver!.ReadAsync([full], cts.Token);
+
+        snapshots.Count.ShouldBe(1);
+        var snap = snapshots[0];
+        snap.StatusCode.ShouldNotBe(0x80020000u,
+            $"Read returned BadInternalError for {full} — the Host couldn't fulfil the request. " +
+            $"Investigate: the Host service's logs at {System.Environment.GetFolderPath(System.Environment.SpecialFolder.CommonApplicationData)}\\OtOpcUa\\Galaxy\\logs.");
+    }
+
+    /// <summary>
+    ///     Minimal <see cref="IAddressSpaceBuilder"/> implementation that captures every
+    ///     Variable() call into a flat list so tests can inspect what discovery produced
+    ///     without running the full OPC UA node-manager stack.
+    /// </summary>
+    private sealed class CapturingAddressSpaceBuilder : IAddressSpaceBuilder
+    {
+        public List<(string BrowseName, DriverAttributeInfo AttributeInfo)> Variables { get; } = [];
+
+        public IAddressSpaceBuilder Folder(string browseName, string displayName) => this;
+        public IVariableHandle Variable(string browseName, string displayName, DriverAttributeInfo attributeInfo)
+        {
+            Variables.Add((browseName, attributeInfo));
+            return new NoopHandle(attributeInfo.FullName);
+        }
+        public void AddProperty(string browseName, DriverDataType dataType, object? value) { }
+
+        private sealed class NoopHandle(string fullReference) : IVariableHandle
+        {
+            public string FullReference { get; } = fullReference;
+            public IAlarmConditionSink MarkAsAlarmCondition(AlarmConditionInfo info) => new NoopSink();
+            private sealed class NoopSink : IAlarmConditionSink
+            {
+                public void OnTransition(AlarmEventArgs args) { }
+            }
+        }
+    }
+}

tests/ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests/ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.Tests.csproj

@@ -22,6 +22,7 @@
     <ItemGroup>
         <ProjectReference Include="..\..\src\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Proxy.csproj"/>
         <ProjectReference Include="..\..\src\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Host\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.Host.csproj"/>
+        <ProjectReference Include="..\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.TestSupport\ZB.MOM.WW.OtOpcUa.Driver.Galaxy.TestSupport.csproj"/>
     </ItemGroup>
 
     <ItemGroup>