src/ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient/OpcUaClientDriver.cs

@@ -59,6 +59,8 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
 
     private DriverHealth _health = new(DriverState.Unknown, null, null);
     private bool _disposed;
+    /// <summary>URL of the endpoint the driver actually connected to. Exposed via <see cref="HostName"/>.</summary>
+    private string? _connectedEndpointUrl;
 
     public string DriverInstanceId => driverInstanceId;
     public string DriverType => "OpcUaClient";
@@ -69,51 +71,38 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
         try
         {
             var appConfig = await BuildApplicationConfigurationAsync(cancellationToken).ConfigureAwait(false);
+            var candidates = ResolveEndpointCandidates(_options);
 
-            // Endpoint selection: let the stack pick the best matching endpoint for the
-            // requested security policy/mode so the driver doesn't have to hand-validate.
-            // UseSecurity=false when SecurityMode=None shortcuts around cert validation
-            // entirely and is the typical dev-bench configuration.
-            var selected = await SelectMatchingEndpointAsync(
-                appConfig, _options.EndpointUrl, _options.SecurityPolicy, _options.SecurityMode,
-                cancellationToken).ConfigureAwait(false);
-            var endpointConfig = EndpointConfiguration.Create(appConfig);
-            endpointConfig.OperationTimeout = (int)_options.Timeout.TotalMilliseconds;
-            var endpoint = new ConfiguredEndpoint(null, selected, endpointConfig);
+            var identity = BuildUserIdentity(_options);
 
-            var identity = _options.AuthType switch
+            // Failover sweep: try each endpoint in order, return the session from the first
+            // one that successfully connects. Per-endpoint failures are captured so the final
+            // aggregate exception names every URL that was tried and why — critical diag for
+            // operators debugging 'why did the failover pick #3?'.
+            var attemptErrors = new List<string>(candidates.Count);
+            ISession? session = null;
+            string? connectedUrl = null;
+            foreach (var url in candidates)
             {
-                OpcUaAuthType.Anonymous => new UserIdentity(new AnonymousIdentityToken()),
-                // The UserIdentity(string, string) overload was removed in favour of
-                // (string, byte[]) to make the password encoding explicit. UTF-8 is the
-                // overwhelmingly common choice for Basic256Sha256-secured sessions.
-                OpcUaAuthType.Username => new UserIdentity(
-                    _options.Username ?? string.Empty,
-                    System.Text.Encoding.UTF8.GetBytes(_options.Password ?? string.Empty)),
-                OpcUaAuthType.Certificate => BuildCertificateIdentity(_options),
-                _ => new UserIdentity(new AnonymousIdentityToken()),
-            };
+                try
+                {
+                    session = await OpenSessionOnEndpointAsync(
+                        appConfig, url, _options.SecurityPolicy, _options.SecurityMode,
+                        identity, cancellationToken).ConfigureAwait(false);
+                    connectedUrl = url;
+                    break;
+                }
+                catch (Exception ex)
+                {
+                    attemptErrors.Add($"{url} -> {ex.GetType().Name}: {ex.Message}");
+                }
+            }
 
-            // All Session.Create* static methods are marked [Obsolete] in SDK 1.5.378; the
-            // non-obsolete path is DefaultSessionFactory.Instance.CreateAsync (which is the
-            // 8-arg signature matching our driver config — ApplicationConfiguration +
-            // ConfiguredEndpoint, no transport-waiting-connection or reverse-connect-manager
-            // required for the standard opc.tcp direct-connect case).
-            // DefaultSessionFactory's parameterless ctor is also obsolete in 1.5.378; the
-            // current constructor requires an ITelemetryContext. Passing null is tolerated —
-            // the factory falls back to its internal default sink, same as the telemetry:null
-            // on SelectEndpointAsync above.
-            var session = await new DefaultSessionFactory(telemetry: null!).CreateAsync(
-                appConfig,
-                endpoint,
-                false,                                              // updateBeforeConnect
-                _options.SessionName,
-                (uint)_options.SessionTimeout.TotalMilliseconds,
-                identity,
-                null,                                               // preferredLocales
-                cancellationToken).ConfigureAwait(false);
-
-            session.KeepAliveInterval = (int)_options.KeepAliveInterval.TotalMilliseconds;
+            if (session is null)
+                throw new AggregateException(
+                    "OPC UA Client failed to connect to any of the configured endpoints. " +
+                    "Tried:\n  " + string.Join("\n  ", attemptErrors),
+                    attemptErrors.Select(e => new InvalidOperationException(e)));
 
             // Wire the session's keep-alive channel into HostState. OPC UA keep-alives are
             // authoritative for session liveness: the SDK pings on KeepAliveInterval and sets
@@ -128,6 +117,7 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
             session.KeepAlive += _keepAliveHandler;
 
             Session = session;
+            _connectedEndpointUrl = connectedUrl;
             _health = new DriverHealth(DriverState.Healthy, DateTime.UtcNow, null);
             TransitionTo(HostState.Running);
         }
@@ -224,6 +214,71 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
         return config;
     }
 
+    /// <summary>
+    ///     Resolve the ordered failover candidate list. <c>EndpointUrls</c> wins when
+    ///     non-empty; otherwise fall back to <c>EndpointUrl</c> as a single-URL shortcut so
+    ///     existing single-endpoint configs keep working without migration.
+    /// </summary>
+    internal static IReadOnlyList<string> ResolveEndpointCandidates(OpcUaClientDriverOptions opts)
+    {
+        if (opts.EndpointUrls is { Count: > 0 }) return opts.EndpointUrls;
+        return [opts.EndpointUrl];
+    }
+
+    /// <summary>
+    ///     Build the user-identity token from the driver options. Split out of
+    ///     <see cref="InitializeAsync"/> so the failover sweep reuses one identity across
+    ///     every endpoint attempt — generating it N times would re-unlock the user cert's
+    ///     private key N times, wasteful + keeps the password in memory longer.
+    /// </summary>
+    internal static UserIdentity BuildUserIdentity(OpcUaClientDriverOptions options) =>
+        options.AuthType switch
+        {
+            OpcUaAuthType.Anonymous => new UserIdentity(new AnonymousIdentityToken()),
+            OpcUaAuthType.Username => new UserIdentity(
+                options.Username ?? string.Empty,
+                System.Text.Encoding.UTF8.GetBytes(options.Password ?? string.Empty)),
+            OpcUaAuthType.Certificate => BuildCertificateIdentity(options),
+            _ => new UserIdentity(new AnonymousIdentityToken()),
+        };
+
+    /// <summary>
+    ///     Open a session against a single endpoint URL. Bounded by
+    ///     <see cref="OpcUaClientDriverOptions.PerEndpointConnectTimeout"/> so the failover
+    ///     sweep doesn't spend its full budget on one dead server. Moved out of
+    ///     <see cref="InitializeAsync"/> so the failover loop body stays readable.
+    /// </summary>
+    private async Task<ISession> OpenSessionOnEndpointAsync(
+        ApplicationConfiguration appConfig,
+        string endpointUrl,
+        OpcUaSecurityPolicy policy,
+        OpcUaSecurityMode mode,
+        UserIdentity identity,
+        CancellationToken ct)
+    {
+        using var cts = CancellationTokenSource.CreateLinkedTokenSource(ct);
+        cts.CancelAfter(_options.PerEndpointConnectTimeout);
+
+        var selected = await SelectMatchingEndpointAsync(
+            appConfig, endpointUrl, policy, mode, cts.Token).ConfigureAwait(false);
+        var endpointConfig = EndpointConfiguration.Create(appConfig);
+        endpointConfig.OperationTimeout = (int)_options.Timeout.TotalMilliseconds;
+        var endpoint = new ConfiguredEndpoint(null, selected, endpointConfig);
+
+        var session = await new DefaultSessionFactory(telemetry: null!).CreateAsync(
+            appConfig,
+            endpoint,
+            false,                                              // updateBeforeConnect
+            _options.SessionName,
+            (uint)_options.SessionTimeout.TotalMilliseconds,
+            identity,
+            null,                                               // preferredLocales
+            cts.Token).ConfigureAwait(false);
+
+        session.KeepAliveInterval = (int)_options.KeepAliveInterval.TotalMilliseconds;
+        return session;
+    }
+
     /// <summary>
     ///     Select the remote endpoint matching both the requested <paramref name="policy"/>
     ///     and <paramref name="mode"/>. The SDK's <c>CoreClientUtils.SelectEndpointAsync</c>
@@ -347,6 +402,7 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
         catch { /* best-effort */ }
         try { Session?.Dispose(); } catch { }
         Session = null;
+        _connectedEndpointUrl = null;
 
         TransitionTo(HostState.Unknown);
         _health = new DriverHealth(DriverState.Unknown, _health.LastSuccessfulRead, null);
@@ -718,8 +774,16 @@ public sealed class OpcUaClientDriver(OpcUaClientDriverOptions options, string d
 
     // ---- IHostConnectivityProbe ----
 
-    /// <summary>Endpoint-URL-keyed host identity for the Admin /hosts dashboard.</summary>
-    public string HostName => _options.EndpointUrl;
+    /// <summary>
+    ///     Endpoint-URL-keyed host identity for the Admin /hosts dashboard. Reflects the
+    ///     endpoint the driver actually connected to after the failover sweep — not the
+    ///     first URL in the candidate list — so operators see which of the configured
+    ///     endpoints is currently serving traffic. Falls back to the first configured URL
+    ///     pre-init so the dashboard has something to render before the first connect.
+    /// </summary>
+    public string HostName => _connectedEndpointUrl
+        ?? ResolveEndpointCandidates(_options).FirstOrDefault()
+        ?? _options.EndpointUrl;
 
     public IReadOnlyList<HostConnectivityStatus> GetHostStatuses()
     {

src/ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient/OpcUaClientDriverOptions.cs

@@ -13,9 +13,32 @@ namespace ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient;
 /// </remarks>
 public sealed class OpcUaClientDriverOptions
 {
-    /// <summary>Remote OPC UA endpoint URL, e.g. <c>opc.tcp://plc.internal:4840</c>.</summary>
+    /// <summary>
+    ///     Remote OPC UA endpoint URL, e.g. <c>opc.tcp://plc.internal:4840</c>. Convenience
+    ///     shortcut for a single-endpoint deployment — equivalent to setting
+    ///     <see cref="EndpointUrls"/> to a list with this one URL. When both are provided,
+    ///     the list wins and <see cref="EndpointUrl"/> is ignored.
+    /// </summary>
     public string EndpointUrl { get; init; } = "opc.tcp://localhost:4840";
 
+    /// <summary>
+    ///     Ordered list of candidate endpoint URLs for failover. The driver tries each in
+    ///     order at <see cref="OpcUaClientDriver.InitializeAsync"/> and on session drop;
+    ///     the first URL that successfully connects wins. Typical use-case: an OPC UA server
+    ///     pair running in hot-standby (primary 4840 + backup 4841) where either can serve
+    ///     the same address space. Leave unset (or empty) to use <see cref="EndpointUrl"/>
+    ///     as a single-URL shortcut.
+    /// </summary>
+    public IReadOnlyList<string> EndpointUrls { get; init; } = [];
+
+    /// <summary>
+    ///     Per-endpoint connect-attempt timeout during the failover sweep. Short enough that
+    ///     cycling through several dead servers doesn't blow the overall init budget, long
+    ///     enough to tolerate a slow TLS handshake on a healthy server. Applied independently
+    ///     of <see cref="Timeout"/> which governs steady-state operations.
+    /// </summary>
+    public TimeSpan PerEndpointConnectTimeout { get; init; } = TimeSpan.FromSeconds(3);
+
     /// <summary>
     ///     Security policy to require when selecting an endpoint. Either a
     ///     <see cref="OpcUaSecurityPolicy"/> enum constant or a free-form string (for

tests/ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient.Tests/OpcUaClientFailoverTests.cs

@@ -0,0 +1,81 @@
+using Shouldly;
+using Xunit;
+using ZB.MOM.WW.OtOpcUa.Core.Abstractions;
+
+namespace ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient.Tests;
+
+[Trait("Category", "Unit")]
+public sealed class OpcUaClientFailoverTests
+{
+    [Fact]
+    public void ResolveEndpointCandidates_prefers_EndpointUrls_when_provided()
+    {
+        var opts = new OpcUaClientDriverOptions
+        {
+            EndpointUrl = "opc.tcp://fallback:4840",
+            EndpointUrls = ["opc.tcp://primary:4840", "opc.tcp://backup:4841"],
+        };
+        var list = OpcUaClientDriver.ResolveEndpointCandidates(opts);
+        list.Count.ShouldBe(2);
+        list[0].ShouldBe("opc.tcp://primary:4840");
+        list[1].ShouldBe("opc.tcp://backup:4841");
+    }
+
+    [Fact]
+    public void ResolveEndpointCandidates_falls_back_to_single_EndpointUrl_when_list_empty()
+    {
+        var opts = new OpcUaClientDriverOptions { EndpointUrl = "opc.tcp://only:4840" };
+        var list = OpcUaClientDriver.ResolveEndpointCandidates(opts);
+        list.Count.ShouldBe(1);
+        list[0].ShouldBe("opc.tcp://only:4840");
+    }
+
+    [Fact]
+    public void ResolveEndpointCandidates_empty_list_treated_as_fallback_to_EndpointUrl()
+    {
+        // Explicit empty list should still fall back to the single-URL shortcut rather than
+        // producing a zero-candidate sweep that would immediately throw with no URLs tried.
+        var opts = new OpcUaClientDriverOptions
+        {
+            EndpointUrl = "opc.tcp://single:4840",
+            EndpointUrls = [],
+        };
+        OpcUaClientDriver.ResolveEndpointCandidates(opts).Count.ShouldBe(1);
+    }
+
+    [Fact]
+    public void HostName_uses_first_candidate_before_connect()
+    {
+        var opts = new OpcUaClientDriverOptions
+        {
+            EndpointUrls = ["opc.tcp://primary:4840", "opc.tcp://backup:4841"],
+        };
+        using var drv = new OpcUaClientDriver(opts, "opcua-host");
+        drv.HostName.ShouldBe("opc.tcp://primary:4840",
+            "pre-connect the dashboard should show the first candidate URL so operators can link back");
+    }
+
+    [Fact]
+    public async Task Initialize_against_all_unreachable_endpoints_throws_AggregateException_listing_each()
+    {
+        // Port 1 + port 2 + port 3 on loopback are all guaranteed closed (TCP RST immediate).
+        // Failover sweep should attempt all three and throw AggregateException naming each URL
+        // so operators see exactly which candidates were tried.
+        var opts = new OpcUaClientDriverOptions
+        {
+            EndpointUrls = ["opc.tcp://127.0.0.1:1", "opc.tcp://127.0.0.1:2", "opc.tcp://127.0.0.1:3"],
+            PerEndpointConnectTimeout = TimeSpan.FromMilliseconds(500),
+            Timeout = TimeSpan.FromMilliseconds(500),
+            AutoAcceptCertificates = true,
+        };
+        using var drv = new OpcUaClientDriver(opts, "opcua-failover");
+
+        var ex = await Should.ThrowAsync<AggregateException>(async () =>
+            await drv.InitializeAsync("{}", TestContext.Current.CancellationToken));
+
+        ex.Message.ShouldContain("127.0.0.1:1");
+        ex.Message.ShouldContain("127.0.0.1:2");
+        ex.Message.ShouldContain("127.0.0.1:3");
+        drv.GetHealth().State.ShouldBe(DriverState.Faulted);
+    }
+}