refactor: rename ScadaLink → ZB.MOM.WW.ScadaBridge (code + projects + namespaces)

Solution + 23 src projects + 26 test projects renamed; folders, csproj,
namespaces, and ScadaLinkDbContext/ScadaBridgeDbContext class updated.
ActorSystem "scadalink" → "scadabridge", Akka seed-node URLs migrated.
SQL roles/logins, LDAP domains, CLI command name, and CLI config dir
(~/.scadalink → ~/.scadabridge) also renamed.

Build green; 5 Host.Tests fail awaiting SQL login rename in next commit.
Pre-existing StaleTagMonitor timing flakes unchanged.

Rename script committed at tools/rename-to-scadabridge.sh.
This commit is contained in:
Joseph Doherty
2026-05-28 09:37:45 -04:00
parent 6d87ee3c3b
commit 7b0b9c7365
1531 changed files with 11180 additions and 11054 deletions
@@ -0,0 +1,79 @@
namespace ZB.MOM.WW.ScadaBridge.ClusterInfrastructure;
/// <summary>
/// Cluster configuration model, bound from the <c>ScadaBridge:Cluster</c> section
/// of <c>appsettings.json</c> via the Options pattern.
/// <para>
/// This project owns the cluster <em>configuration contract</em>. The actual
/// Akka.NET bootstrap — building the HOCON from these values, starting the
/// <c>ActorSystem</c>, configuring the split-brain resolver and wiring
/// <c>CoordinatedShutdown</c> — lives in <c>ZB.MOM.WW.ScadaBridge.Host</c>
/// (see <c>Component-ClusterInfrastructure.md</c> → "Implementation Note — Code Placement").
/// </para>
/// <para>
/// Node-identity settings (remoting hostname/port, cluster role, site identifier,
/// gRPC port) are deliberately <em>not</em> here — they are owned by
/// <c>ZB.MOM.WW.ScadaBridge.Host.NodeOptions</c> (<c>ScadaBridge:Node</c> section). Local SQLite
/// storage paths are owned by the database / store-and-forward options. This class
/// holds only the cluster-formation and failure-detection settings shared by every node.
/// </para>
/// </summary>
public class ClusterOptions
{
// ClusterInfra-011: the previous `public const string SectionName = "ScadaBridge:Cluster";`
// was documented as "single source of truth so binding sites do not hard-code the
// magic string" but no caller ever read it — the Host's SiteServiceRegistration and
// StartupValidator both hard-code the literal directly. Wiring those binding sites
// to reference the constant lives in the Host's edit scope (a separate code-review
// task); rather than carry a public constant whose guarantee the code does not
// deliver, the constant is removed and the literal stays in the Host until the
// Host-side wiring is done. If a future Host change wants the constant back, add it
// when the binding sites can be updated in the same commit.
/// <summary>
/// Akka.NET cluster seed nodes. Both nodes are seed nodes — each node lists
/// itself and its partner — so either can start first and form the cluster.
/// Must contain at least one entry.
/// </summary>
public List<string> SeedNodes { get; set; } = new();
/// <summary>
/// Split-brain resolver strategy. Must be <c>keep-oldest</c> for the two-node
/// clusters ScadaBridge uses: quorum strategies (<c>keep-majority</c>,
/// <c>static-quorum</c>) cannot distinguish a crash from a partition with only
/// two nodes and would shut down the whole cluster.
/// </summary>
public string SplitBrainResolverStrategy { get; set; } = "keep-oldest";
/// <summary>
/// Time the cluster membership must remain stable before the split-brain
/// resolver acts to down unreachable nodes. Must be positive. Default 15s.
/// </summary>
public TimeSpan StableAfter { get; set; } = TimeSpan.FromSeconds(15);
/// <summary>
/// Frequency of cluster failure-detector heartbeat messages between nodes.
/// Must be well below <see cref="FailureDetectionThreshold"/>. Default 2s.
/// </summary>
public TimeSpan HeartbeatInterval { get; set; } = TimeSpan.FromSeconds(2);
/// <summary>
/// Time without a heartbeat before a node is considered unreachable
/// (Akka's <c>acceptable-heartbeat-pause</c>). Default 10s.
/// </summary>
public TimeSpan FailureDetectionThreshold { get; set; } = TimeSpan.FromSeconds(10);
/// <summary>
/// Akka's <c>min-nr-of-members</c>. Must be <c>1</c>: after failover only one
/// node runs, and a value of <c>2</c> blocks the cluster singleton (Site Runtime
/// Deployment Manager) — and therefore all data collection — indefinitely.
/// </summary>
public int MinNrOfMembers { get; set; } = 1;
/// <summary>
/// The keep-oldest resolver's <c>down-if-alone</c> flag. When <c>true</c> (the
/// design-doc requirement), the oldest node downs itself if it finds it has no
/// other reachable members, rather than running as an isolated single-node cluster.
/// </summary>
public bool DownIfAlone { get; set; } = true;
}
@@ -0,0 +1,95 @@
using Microsoft.Extensions.Options;
namespace ZB.MOM.WW.ScadaBridge.ClusterInfrastructure;
/// <summary>
/// CI-004: Validates <see cref="ClusterOptions"/> at startup. The values it
/// guards carry cluster-wide consequences — the design doc
/// (<c>Component-ClusterInfrastructure.md</c>) is emphatic that misconfiguring
/// them produces a total cluster shutdown or an indefinitely blocked singleton.
/// Registered with <c>ValidateOnStart()</c> so a bad <c>appsettings.json</c>
/// fails fast at boot rather than failing far from the cause.
/// </summary>
public sealed class ClusterOptionsValidator : IValidateOptions<ClusterOptions>
{
/// <summary>Split-brain resolver strategies safe for ScadaBridge's two-node clusters.</summary>
private static readonly HashSet<string> AllowedStrategies = new(StringComparer.OrdinalIgnoreCase)
{
"keep-oldest"
};
/// <summary>
/// Validates the cluster options, returning a failure result if any critical settings are misconfigured.
/// </summary>
/// <param name="name">Named options instance name (unused; all instances are validated identically).</param>
/// <param name="options">The cluster options to validate.</param>
public ValidateOptionsResult Validate(string? name, ClusterOptions options)
{
var failures = new List<string>();
if (options.SeedNodes is null || options.SeedNodes.Count < 2)
{
// CI-012: design doc states "both nodes are seed nodes — each node lists
// both itself and its partner" so a properly-configured deployment lists
// two. Accepting a single-seed configuration silently defeats the
// "no startup ordering dependency" guarantee called out by
// Component-ClusterInfrastructure.md (Node Configuration).
failures.Add(
"ClusterOptions.SeedNodes must contain at least 2 seed nodes "
+ "(Component-ClusterInfrastructure.md → Node Configuration: "
+ "both nodes are seed nodes); a single-seed configuration defeats "
+ "the no-startup-ordering-dependency guarantee.");
}
if (string.IsNullOrWhiteSpace(options.SplitBrainResolverStrategy)
|| !AllowedStrategies.Contains(options.SplitBrainResolverStrategy))
{
failures.Add(
$"ClusterOptions.SplitBrainResolverStrategy must be 'keep-oldest' for a two-node cluster; " +
$"'{options.SplitBrainResolverStrategy}' would risk a total cluster shutdown on a partition.");
}
if (options.MinNrOfMembers != 1)
{
failures.Add(
$"ClusterOptions.MinNrOfMembers must be 1 (was {options.MinNrOfMembers}); " +
"any other value blocks the cluster singleton after failover and halts all data collection.");
}
if (options.StableAfter <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.StableAfter must be a positive duration.");
}
if (options.HeartbeatInterval <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.HeartbeatInterval must be a positive duration.");
}
if (options.FailureDetectionThreshold <= TimeSpan.Zero)
{
failures.Add("ClusterOptions.FailureDetectionThreshold must be a positive duration.");
}
if (options.HeartbeatInterval >= options.FailureDetectionThreshold)
{
failures.Add(
$"ClusterOptions.HeartbeatInterval ({options.HeartbeatInterval}) must be well below " +
$"FailureDetectionThreshold ({options.FailureDetectionThreshold}); otherwise nodes are " +
"declared unreachable before a heartbeat can arrive.");
}
if (!options.DownIfAlone)
{
failures.Add(
"ClusterOptions.DownIfAlone must be true for the keep-oldest resolver "
+ "(Component-ClusterInfrastructure.md → Split-Brain Resolution); with it false the "
+ "oldest node can run as an isolated single-node cluster during a partition while the "
+ "younger node forms its own, producing two live clusters.");
}
return failures.Count > 0
? ValidateOptionsResult.Fail(failures)
: ValidateOptionsResult.Success;
}
}
@@ -0,0 +1,46 @@
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.DependencyInjection.Extensions;
using Microsoft.Extensions.Options;
namespace ZB.MOM.WW.ScadaBridge.ClusterInfrastructure;
/// <summary>
/// DI registration for the Cluster Infrastructure component.
/// </summary>
public static class ServiceCollectionExtensions
{
/// <summary>
/// Registers the Cluster Infrastructure services. This component owns the
/// cluster <em>configuration contract</em> (<see cref="ClusterOptions"/>); the
/// Akka.NET bootstrap itself lives in <c>ZB.MOM.WW.ScadaBridge.Host</c>
/// (see <c>Component-ClusterInfrastructure.md</c>).
/// <para>
/// Registering the <see cref="ClusterOptionsValidator"/> means a misconfigured
/// <c>ScadaBridge:Cluster</c> section (e.g. <c>MinNrOfMembers: 2</c> or a quorum
/// split-brain strategy) throws an <see cref="OptionsValidationException"/> the
/// first time <see cref="IOptions{TOptions}"/> is resolved, rather than booting
/// into a broken cluster.
/// </para>
/// </summary>
/// <param name="services">The service collection to register into.</param>
public static IServiceCollection AddClusterInfrastructure(this IServiceCollection services)
{
services.TryAddEnumerable(
ServiceDescriptor.Singleton<IValidateOptions<ClusterOptions>, ClusterOptionsValidator>());
return services;
}
// ClusterInfra-014: the previous `AddClusterInfrastructureActors` extension
// was dead surface — its XML doc told callers "do not call", its body
// unconditionally threw `NotImplementedException`, and no production caller
// existed anywhere in the solution (verified by grep). The CI-002
// "throw loudly" decision was made while CI-001's ownership question was
// still open; that question is now permanently settled by the
// "Implementation Note — Code Placement" section of
// Component-ClusterInfrastructure.md, which records that all actor wiring
// lives in ZB.MOM.WW.ScadaBridge.Host (AkkaHostedService). Keeping a public extension
// method that exists only to throw was API-surface noise that an IDE would
// still suggest via auto-complete, so the method and its companion
// `AddClusterInfrastructureActors_ThrowsRatherThanSilentlySucceeding` test
// were both removed.
}
@@ -0,0 +1,19 @@
<Project Sdk="Microsoft.NET.Sdk">
<PropertyGroup>
<TargetFramework>net10.0</TargetFramework>
<ImplicitUsings>enable</ImplicitUsings>
<Nullable>enable</Nullable>
<TreatWarningsAsErrors>true</TreatWarningsAsErrors>
</PropertyGroup>
<ItemGroup>
<PackageReference Include="Microsoft.Extensions.DependencyInjection.Abstractions" />
<PackageReference Include="Microsoft.Extensions.Options" />
</ItemGroup>
<ItemGroup>
<ProjectReference Include="../ZB.MOM.WW.ScadaBridge.Commons/ZB.MOM.WW.ScadaBridge.Commons.csproj" />
</ItemGroup>
</Project>