ScadaBridge/src/ZB.MOM.WW.ScadaBridge.Communication/CommunicationOptions.cs

namespace ZB.MOM.WW.ScadaBridge.Communication;

/// <summary>
/// Configuration options for central-site communication, including per-pattern
/// timeouts and transport heartbeat settings.
/// </summary>
public class CommunicationOptions
{
    /// <summary>
    /// Central control-plane gRPC endpoints (preferred first), e.g.
    /// <c>["http://scadabridge-central-a:8083", "http://scadabridge-central-b:8083"]</c>. Dialled by
    /// <see cref="Grpc.CentralChannelProvider"/> with sticky failover/failback. Required on every site
    /// node — gRPC (<c>CentralControlService</c>) is the only site→central transport.
    /// </summary>
    /// <remarks>
    /// Sites reach central by container/host name, NOT via Traefik (which is HTTP/1 only; gRPC is
    /// h2c on the central node's dedicated <c>CentralGrpcPort</c>).
    /// </remarks>
    public List<string> CentralGrpcEndpoints { get; set; } = new();

    /// <summary>Timeout for deployment commands (typically longest due to apply logic).</summary>
    public TimeSpan DeploymentTimeout { get; set; } = TimeSpan.FromMinutes(2);

    /// <summary>Timeout for lifecycle commands (disable, enable, delete).</summary>
    public TimeSpan LifecycleTimeout { get; set; } = TimeSpan.FromSeconds(30);

    /// <summary>Timeout for artifact deployment commands.</summary>
    public TimeSpan ArtifactDeploymentTimeout { get; set; } = TimeSpan.FromMinutes(1);

    /// <summary>Timeout for remote query requests (event logs, parked messages).</summary>
    public TimeSpan QueryTimeout { get; set; } = TimeSpan.FromSeconds(30);

    /// <summary>Timeout for integration call routing.</summary>
    public TimeSpan IntegrationTimeout { get; set; } = TimeSpan.FromSeconds(30);

    /// <summary>Timeout for debug view subscribe/unsubscribe handshake.</summary>
    public TimeSpan DebugViewTimeout { get; set; } = TimeSpan.FromSeconds(10);

    /// <summary>Timeout for health report acknowledgement (fire-and-forget, but bounded).</summary>
    public TimeSpan HealthReportTimeout { get; set; } = TimeSpan.FromSeconds(10);

    /// <summary>
    /// Notification Outbox: timeout for forwarding a buffered notification to central
    /// and awaiting its <c>NotificationSubmitAck</c>. A timeout is treated as a
    /// transient failure — the Store-and-Forward engine keeps the message buffered
    /// and retries the forward at the fixed retry interval.
    /// </summary>
    public TimeSpan NotificationForwardTimeout { get; set; } = TimeSpan.FromSeconds(30);

    /// <summary>
    /// Audit Log: timeout for the SITE-side Ask that forwards one audit-telemetry batch
    /// (<c>IngestAuditEventsCommand</c> / <c>IngestCachedTelemetryCommand</c>) through the site's
    /// <c>SiteCommunicationActor</c> and awaits central's ack. Deliberately the LONGEST rung of the
    /// ingest timeout ladder.
    /// </summary>
    /// <remarks>
    /// <para>
    /// The ladder is strictly monotonic, outermost first:
    /// <c>AuditForwardTimeout</c> (35 s, this option) &gt;
    /// <c>SiteStreamGrpcServer.AuditIngestAskTimeout</c> (30 s — both the gRPC call deadline and
    /// central's own Ask of the ingest singleton) &gt;
    /// <c>AuditLogIngestActor.IngestBudget</c> (20 s) &gt;
    /// <c>AuditLogIngestActor.IngestSqlCommandTimeout</c> (15 s).
    /// </para>
    /// <para>
    /// Strictness matters: it used to reuse <see cref="NotificationForwardTimeout"/> (30 s), the
    /// same value as the rung below it, so a slow-but-succeeding central write could be acked to a
    /// caller whose Ask had already expired — the drain loop would treat the batch as unsent and
    /// re-ship it. Central dedups on <c>EventId</c>, so the duplicate was harmless but the retry
    /// traffic and the "stalled" telemetry signal were not. A timeout here is still transient: the
    /// rows stay <c>Pending</c> and drain on the next tick.
    /// </para>
    /// <para>
    /// <see cref="CommunicationOptionsValidator"/> enforces the outermost inequality — this value
    /// must be strictly greater than the 30 s gRPC/central Ask rung.
    /// </para>
    /// </remarks>
    public TimeSpan AuditForwardTimeout { get; set; } = TimeSpan.FromSeconds(35);

    /// <summary>
    /// Preshared key authenticating this node's gRPC control plane — the site↔central
    /// boundary. On a site node this is the key its inbound gate
    /// (<c>ControlPlaneAuthInterceptor</c>) expects on every <c>SiteStreamService</c> call, and
    /// which central must present; central resolves the matching value per site from its own
    /// secret store under the name <c>SB-GRPC-PSK-{siteId}</c>.
    /// </summary>
    /// <remarks>
    /// <para>
    /// In production this is supplied as <c>${secret:SB-GRPC-PSK-&lt;siteId&gt;}</c> and expanded
    /// out of the secrets store before the host is built, so the plaintext never sits in
    /// appsettings. Development rigs set a literal, mirroring the LocalDb replication key.
    /// </para>
    /// <para>
    /// <b>Empty means closed, not open.</b> With no key set the interceptor rejects every gated
    /// call. This is not optional configuration: a node that ships without a key serves no
    /// streams and no audit pulls.
    /// </para>
    /// <para>
    /// Distinct from <c>LocalDb:Replication:ApiKey</c>, which authenticates the pair partner for
    /// database replication over the same listener. The two are never shared.
    /// </para>
    /// </remarks>
    public string GrpcPsk { get; set; } = "";

    /// <summary>
    /// Central-side per-site gRPC preshared keys, keyed by site identifier — the mirror image
    /// of <see cref="GrpcPsk"/>, which is the single key a site node expects on its own inbound
    /// gate. An entry here takes precedence over the secret store for that site.
    /// </summary>
    /// <remarks>
    /// <para>
    /// <b>Why both a config map and a secret store.</b> The store is the primary source and the
    /// only one that works for the real case: sites are added at runtime from the Central UI, so
    /// their keys cannot be enumerated in configuration at boot, and <c>SitePskProvider</c>
    /// resolves <c>SB-GRPC-PSK-{siteId}</c> on demand. This map covers the cases the store
    /// cannot or should not: a development rig that runs with no master key and injects every
    /// credential as an environment override, and an operator pinning one site's key without
    /// touching the store. Values may themselves be <c>${secret:…}</c> references, since a map
    /// declared in configuration IS enumerable at boot.
    /// </para>
    /// <para>
    /// Absence is not a fallback to "unauthenticated" in either source — a site with no key in
    /// the map and none in the store cannot be dialed at all.
    /// </para>
    /// </remarks>
    public Dictionary<string, string> SitePsks { get; set; } = new();

    /// <summary>gRPC keepalive ping interval for streaming connections.</summary>
    public TimeSpan GrpcKeepAlivePingDelay { get; set; } = TimeSpan.FromSeconds(15);

    /// <summary>gRPC keepalive ping timeout — stream is considered dead if no response within this period.</summary>
    public TimeSpan GrpcKeepAlivePingTimeout { get; set; } = TimeSpan.FromSeconds(10);

    /// <summary>Maximum lifetime for a single gRPC stream before the server forces re-establishment.</summary>
    public TimeSpan GrpcMaxStreamLifetime { get; set; } = TimeSpan.FromHours(4);

    /// <summary>Maximum number of concurrent gRPC streaming subscriptions per site node.</summary>
    public int GrpcMaxConcurrentStreams { get; set; } = 100;

    /// <summary>
    /// Send-channel capacity for a per-instance Debug View stream (<c>SubscribeInstance</c>).
    /// Lossy by design: the Debug View is a diagnostic surface and drops its oldest events
    /// under backpressure rather than stalling the site's event hub.
    /// </summary>
    public int GrpcInstanceStreamChannelCapacity { get; set; } = 1000;

    /// <summary>
    /// Send-channel capacity for the site-wide alarm stream (<c>SubscribeSite</c>). Larger
    /// than the Debug View's (WP2.3): that feed backs the operator Alarm Summary, where a
    /// silently dropped transition is a missed alarm rather than a missed diagnostic frame,
    /// and an alarm burst arriving during a WAN stall must survive the stall.
    /// </summary>
    public int GrpcSiteAlarmStreamChannelCapacity { get; set; } = 20_000;

    /// <summary>
    /// Maximum events coalesced into one site→central stream frame (R2 — gRPC event
    /// batching). At target scale a site emits ~37.5k events/s and every one of them used to
    /// cost its own gRPC message; batching amortises that framing overhead. Set to 1 to
    /// disable batching on this node without a wire change (every frame then carries exactly
    /// one event, which is also what an un-negotiated subscription gets).
    /// </summary>
    public int GrpcStreamBatchMaxEvents { get; set; } = 100;

    /// <summary>
    /// Maximum time the site lingers accumulating a stream batch once a backlog has been
    /// observed. Deliberately small: the target-scale load test measures end-to-end event
    /// latency against a 250 ms P99 threshold (measured P99 4.57 ms), and this window is the
    /// only latency batching can add — so it is validated strictly below that threshold.
    /// A lone event on a quiet stream is never delayed by it (see
    /// <c>SiteStreamEventBatcher</c>); the window applies only after a backlog is proven.
    /// <see cref="TimeSpan.Zero"/> means "pack only what is already queued, never wait".
    /// </summary>
    public TimeSpan GrpcStreamBatchWindow { get; set; } = TimeSpan.FromMilliseconds(25);

    /// <summary>Akka.Remote transport heartbeat interval.</summary>
    public TimeSpan TransportHeartbeatInterval { get; set; } = TimeSpan.FromSeconds(5);

    /// <summary>
    /// Application-level site→central heartbeat cadence. Distinct from
    /// <see cref="TransportHeartbeatInterval"/> (the Akka.Remote failure-detector
    /// setting) — tuning the transport FD must not silently retune the health
    /// heartbeat. Default equals the old effective value (5s) — no behavior change.
    /// </summary>
    public TimeSpan ApplicationHeartbeatInterval { get; set; } = TimeSpan.FromSeconds(5);

    /// <summary>Akka.Remote transport failure detection threshold.</summary>
    public TimeSpan TransportFailureThreshold { get; set; } = TimeSpan.FromSeconds(15);

    /// <summary>
    /// Base URL (Traefik/LB) the SITE uses to fetch deploy configs from central,
    /// e.g. "https://central.example:9000". Carried in RefreshDeploymentCommand so
    /// sites need no new standing config. Empty disables notify-and-fetch fallback.
    /// </summary>
    public string CentralFetchBaseUrl { get; set; } = "";

    /// <summary>
    /// How long a staged PendingDeployment (and its fetch token) stays valid. Must
    /// comfortably cover both site nodes' fetches within one deploy window.
    /// </summary>
    public TimeSpan PendingDeploymentTtl { get; set; } = TimeSpan.FromMinutes(5);

    /// <summary>
    /// How often the central <c>PendingDeploymentPurgeActor</c> singleton reclaims
    /// expired (TTL-elapsed) PendingDeployment staging rows. Best-effort hygiene only:
    /// supersession bounds pending rows to ≤1 per instance and the config-fetch endpoint
    /// already enforces the TTL, so this purge merely sweeps rows left behind by instances
    /// that are deployed once and never re-deployed. Default 1 hour ≫ <see cref="PendingDeploymentTtl"/>.
    /// </summary>
    public TimeSpan PendingDeploymentPurgeInterval { get; set; } = TimeSpan.FromHours(1);

    // ── Aggregated live alarm cache (plan #10, Task 4) ───────────────────────────
    // Introduced by Task 4 with sane defaults; Task 6 formalizes eager validation
    // (positive interval/linger, positive concurrency/subscriber cap) and telemetry.

    /// <summary>
    /// How long the shared per-site live alarm aggregator keeps its gRPC stream +
    /// in-memory cache warm after its LAST Alarm Summary viewer leaves, before it is
    /// torn down. A short linger avoids re-seed thrash when an operator navigates away
    /// and straight back. Default 30s.
    /// </summary>
    public TimeSpan LiveAlarmCacheLinger { get; set; } = TimeSpan.FromSeconds(30);

    /// <summary>
    /// Cadence of the per-site aggregator's periodic reconcile snapshot fan-out. The
    /// backstop that corrects instance-set drift (enable/disable/deploy/delete) and any
    /// alarm delta missed by the live stream, since the aggregated stream never carries
    /// an explicit "removed" event. Default 60s.
    /// </summary>
    public TimeSpan LiveAlarmCacheReconcileInterval { get; set; } = TimeSpan.FromSeconds(60);

    /// <summary>
    /// Max concurrent per-instance snapshot fetches during a live-cache seed/reconcile
    /// fan-out — the aggregated analogue of the Alarm Summary poll's
    /// <c>MaxConcurrentFetches</c>. Default 8.
    /// </summary>
    public int LiveAlarmCacheSeedConcurrency { get; set; } = 8;

    /// <summary>
    /// Upper bound on concurrent Alarm Summary viewers sharing one site's aggregator.
    /// A defensive cap only — one gRPC stream per site is shared regardless of viewer
    /// count; this just bounds the subscriber list. Default 200.
    /// </summary>
    public int LiveAlarmCacheMaxSubscribersPerSite { get; set; } = 200;

    /// <summary>
    /// Publish-coalescing window for live alarm deltas: an applied delta marks the cache
    /// dirty and one publish (fresh snapshot + per-viewer onChanged fan-out) fires after
    /// this window, batching an alarm storm into ~4 publishes/second instead of one per
    /// transition (review 02 round 2, N6). Zero = publish per delta (legacy). Seed and
    /// reconcile publishes are always immediate. Default 250 ms.
    /// </summary>
    public TimeSpan LiveAlarmCachePublishCoalesce { get; set; } = TimeSpan.FromMilliseconds(250);

    /// <summary>
    /// Random jitter added to each per-site reconcile tick, as a fraction of
    /// <see cref="LiveAlarmCacheReconcileInterval"/> (WP2.3). Without it every aggregator
    /// started by one central failover fans its whole-site snapshot out on the same 60s
    /// boundary forever. Zero disables jitter.
    /// </summary>
    public double LiveAlarmCacheReconcileJitterFraction { get; set; } = 0.2;
}