lmxopcua/tests/Server/ZB.MOM.WW.OtOpcUa.Server.Tests/PeerHttpProbeLoopTests.cs

using System.Net;
using System.Net.Http;
using Microsoft.EntityFrameworkCore;
using Microsoft.Extensions.Logging.Abstractions;
using Shouldly;
using Xunit;
using ZB.MOM.WW.OtOpcUa.Configuration;
using ZB.MOM.WW.OtOpcUa.Configuration.Entities;
using ZB.MOM.WW.OtOpcUa.Configuration.Enums;
using ZB.MOM.WW.OtOpcUa.Server.Hosting;
using ZB.MOM.WW.OtOpcUa.Server.Redundancy;

namespace ZB.MOM.WW.OtOpcUa.Server.Tests;

/// <summary>
///     Unit tests for <see cref="PeerHttpProbeLoop"/>. Drives <c>TickAsync</c> synchronously
///     via a <see cref="IHttpClientFactory"/> test double so we don't race the loop's
///     <c>Task.Delay</c>.
/// </summary>
[Trait("Category", "Unit")]
public sealed class PeerHttpProbeLoopTests : IDisposable
{
    private readonly OtOpcUaConfigDbContext _db;
    private readonly IDbContextFactory<OtOpcUaConfigDbContext> _dbFactory;

    public PeerHttpProbeLoopTests()
    {
        var opts = new DbContextOptionsBuilder<OtOpcUaConfigDbContext>()
            .UseInMemoryDatabase($"peer-http-{Guid.NewGuid():N}")
            .Options;
        _db = new OtOpcUaConfigDbContext(opts);
        _dbFactory = new DbContextFactory(opts);
    }

    public void Dispose() => _db.Dispose();

    [Fact]
    public async Task Tick_with_no_peers_is_a_no_op()
    {
        var tracker = new PeerReachabilityTracker();
        var coordinator = await SeedAndInitializeAsync("A", ("A", RedundancyRole.Primary, "urn:A"));

        var loop = new PeerHttpProbeLoop(coordinator, tracker,
            new StubHttpClientFactory(_ => new HttpResponseMessage(HttpStatusCode.OK)),
            NullLogger<PeerHttpProbeLoop>.Instance);

        await loop.TickAsync(CancellationToken.None);
        tracker.Get("B").ShouldBe(PeerReachability.Unknown);
    }

    [Fact]
    public async Task Tick_marks_peer_healthy_when_healthz_returns_200()
    {
        var coordinator = await SeedAndInitializeAsync("A",
            ("A", RedundancyRole.Primary, "urn:A"),
            ("B", RedundancyRole.Secondary, "urn:B"));
        var tracker = new PeerReachabilityTracker();
        var factory = new StubHttpClientFactory(req =>
        {
            req.RequestUri!.AbsolutePath.ShouldBe("/healthz");
            return new HttpResponseMessage(HttpStatusCode.OK);
        });

        var loop = new PeerHttpProbeLoop(coordinator, tracker, factory, NullLogger<PeerHttpProbeLoop>.Instance);
        await loop.TickAsync(CancellationToken.None);

        tracker.Get("B").HttpHealthy.ShouldBeTrue();
    }

    [Fact]
    public async Task Tick_marks_peer_unhealthy_when_healthz_throws()
    {
        var coordinator = await SeedAndInitializeAsync("A",
            ("A", RedundancyRole.Primary, "urn:A"),
            ("B", RedundancyRole.Secondary, "urn:B"));
        var tracker = new PeerReachabilityTracker();
        var factory = new StubHttpClientFactory(_ => throw new HttpRequestException("no route to host"));

        var loop = new PeerHttpProbeLoop(coordinator, tracker, factory, NullLogger<PeerHttpProbeLoop>.Instance);
        await loop.TickAsync(CancellationToken.None);

        tracker.Get("B").HttpHealthy.ShouldBeFalse();
    }

    [Fact]
    public async Task Tick_preserves_UaHealthy_bit_when_flipping_HttpHealthy()
    {
        var coordinator = await SeedAndInitializeAsync("A",
            ("A", RedundancyRole.Primary, "urn:A"),
            ("B", RedundancyRole.Secondary, "urn:B"));
        var tracker = new PeerReachabilityTracker();
        tracker.Update("B", new PeerReachability(HttpHealthy: false, UaHealthy: true));

        var factory = new StubHttpClientFactory(_ => new HttpResponseMessage(HttpStatusCode.OK));
        var loop = new PeerHttpProbeLoop(coordinator, tracker, factory, NullLogger<PeerHttpProbeLoop>.Instance);
        await loop.TickAsync(CancellationToken.None);

        var current = tracker.Get("B");
        current.HttpHealthy.ShouldBeTrue();
        current.UaHealthy.ShouldBeTrue("UA bit must not be clobbered by the HTTP probe");
    }

    [Fact]
    public async Task Tick_marks_peer_unhealthy_on_non_2xx_response()
    {
        var coordinator = await SeedAndInitializeAsync("A",
            ("A", RedundancyRole.Primary, "urn:A"),
            ("B", RedundancyRole.Secondary, "urn:B"));
        var tracker = new PeerReachabilityTracker();
        var factory = new StubHttpClientFactory(_ => new HttpResponseMessage(HttpStatusCode.ServiceUnavailable));

        var loop = new PeerHttpProbeLoop(coordinator, tracker, factory, NullLogger<PeerHttpProbeLoop>.Instance);
        await loop.TickAsync(CancellationToken.None);

        tracker.Get("B").HttpHealthy.ShouldBeFalse();
    }

    [Fact]
    public async Task Tick_does_not_mutate_factory_vended_client_Timeout()
    {
        // Server-012: timeouts belong on the named-client registration or a per-request CTS,
        // NOT on a factory-vended HttpClient (which IHttpClientFactory may pool/recycle).
        // Mutating client.Timeout per tick is at minimum a bad smell and races with
        // IHttpClientFactory's lifecycle expectations.
        var coordinator = await SeedAndInitializeAsync("A",
            ("A", RedundancyRole.Primary, "urn:A"),
            ("B", RedundancyRole.Secondary, "urn:B"));
        var tracker = new PeerReachabilityTracker();
        var factoryInitialTimeout = TimeSpan.FromMinutes(2);
        var factory = new RecordingHttpClientFactory(
            _ => new HttpResponseMessage(HttpStatusCode.OK),
            factoryInitialTimeout);

        var loop = new PeerHttpProbeLoop(coordinator, tracker, factory, NullLogger<PeerHttpProbeLoop>.Instance,
            options: new PeerProbeOptions { HttpProbeTimeout = TimeSpan.FromSeconds(3) });

        await loop.TickAsync(CancellationToken.None);

        factory.LastCreatedClient.ShouldNotBeNull();
        factory.LastCreatedClient.Timeout.ShouldBe(factoryInitialTimeout,
            "the probe loop must not mutate the factory-vended HttpClient's Timeout — "
            + "per-call timeout should be enforced via a CancellationToken or via "
            + "AddHttpClient.ConfigureHttpClient on the named registration.");
    }

    // ---- fixture helpers ---------------------------------------------------

    private async Task<RedundancyCoordinator> SeedAndInitializeAsync(string selfNodeId, params (string id, RedundancyRole role, string appUri)[] nodes)
    {
        _db.ServerClusters.Add(new ServerCluster
        {
            ClusterId = "c1", Name = "Warsaw", Enterprise = "zb", Site = "warsaw",
            RedundancyMode = nodes.Length == 1 ? RedundancyMode.None : RedundancyMode.Warm,
            CreatedBy = "test",
        });
        foreach (var (id, role, appUri) in nodes)
        {
            _db.ClusterNodes.Add(new ClusterNode
            {
                NodeId = id, ClusterId = "c1",
                RedundancyRole = role, Host = id.ToLowerInvariant(),
                ApplicationUri = appUri, CreatedBy = "test",
            });
        }
        await _db.SaveChangesAsync();

        var coordinator = new RedundancyCoordinator(_dbFactory, NullLogger<RedundancyCoordinator>.Instance, selfNodeId, "c1");
        await coordinator.InitializeAsync(CancellationToken.None);
        return coordinator;
    }

    private sealed class DbContextFactory(DbContextOptions<OtOpcUaConfigDbContext> options)
        : IDbContextFactory<OtOpcUaConfigDbContext>
    {
        public OtOpcUaConfigDbContext CreateDbContext() => new(options);
    }

    private sealed class StubHttpClientFactory(Func<HttpRequestMessage, HttpResponseMessage> respond) : IHttpClientFactory
    {
        public HttpClient CreateClient(string name) =>
            new(new StubHandler(respond), disposeHandler: true) { Timeout = TimeSpan.FromSeconds(1) };

        private sealed class StubHandler(Func<HttpRequestMessage, HttpResponseMessage> respond) : HttpMessageHandler
        {
            protected override Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, CancellationToken cancellationToken)
                => Task.FromResult(respond(request));
        }
    }

    /// <summary>
    ///     Server-012 — captures the most-recently-vended <see cref="HttpClient"/> so the
    ///     test can assert the probe loop didn't mutate its <see cref="HttpClient.Timeout"/>.
    /// </summary>
    private sealed class RecordingHttpClientFactory(
        Func<HttpRequestMessage, HttpResponseMessage> respond,
        TimeSpan initialTimeout) : IHttpClientFactory
    {
        public HttpClient? LastCreatedClient { get; private set; }
        public HttpClient CreateClient(string name)
        {
            var client = new HttpClient(new RecordingHandler(respond), disposeHandler: true)
            {
                Timeout = initialTimeout,
            };
            LastCreatedClient = client;
            return client;
        }

        private sealed class RecordingHandler(Func<HttpRequestMessage, HttpResponseMessage> respond) : HttpMessageHandler
        {
            protected override Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, CancellationToken cancellationToken)
                => Task.FromResult(respond(request));
        }
    }
}