Files
ScadaBridge/tests/ZB.MOM.WW.ScadaBridge.Host.Tests/SiteLocalDbWiringTests.cs
T
Joseph Doherty a2f0086fa1 fix(health): rows 80+73 — unconfigured replication reports null, reconciler presence surfaced
ROW 80 (observability correctness). An unreplicated site published
LocalDbReplicationConnected=false / LocalDbOplogBacklog=0, contradicting the
nullable semantics every doc and the proto already state ("null means no
reading, NOT disconnected with an empty backlog").

Root cause: the replication engine is registered unconditionally and idles
without a peer, so ISyncStatus answers false/0 on a node that was never meant to
replicate, and LocalDbReplicationStatusReporter passed that straight through.
Two consequences: an operator read connected:false on a healthy site as a fault,
and central's null branches — row 33's conditional KPI emission and
sustained-badge anchor exclusion — never received the null they exist for, so
rig site-b/site-c emitted a permanently misleading
localDbReplicationConnected=0 series.

Fix mirrors the SiteEventLogReplicationCheck precedent: the Host evaluates
SiteLocalDbSetup.ReplicationIsConfigured ONCE and passes it to
AddLocalDbReplicationHealthBridge as a third provider; unconfigured
short-circuits the probe to (null, null). HealthMonitoring never reads LocalDb
config itself, and the same captured value now also feeds
SiteEventLogReplicationCheck so the two surfaces cannot disagree.
ISiteHealthCollector.SetLocalDbReplicationStatus widens to bool? — the
non-nullable bool was the type-level root cause (null was inexpressible).
false now means exactly one thing: configured, peer session down = a real fault.

Wire/central/UI needed no change: the proto already carries BoolValue/Int64Value
wrappers and the aggregator + KPI source already branch on null. This makes code
match doc, so no doc claim was weakened.

Watch the trap that caused this: configured-ness is PeerAddress OR ApiKey (only
the dialling half sets a peer address). The pre-existing SiteLocalDbWiringTests
fixture sets ApiKey and IS a replicating node, so its false/0 assertion is
correct and is kept, with a comment saying so; the new
UnreplicatedSiteLocalDbHealthTests fixture is the genuinely-unreplicated one.

ROW 73 (silent loss of self-heal). AkkaHostedService skips
SiteReconciliationActor when no IDeploymentConfigFetcher is resolvable, and that
skip was a lone boot Warning — a node with no self-heal path at all went on
reporting itself perfectly healthy forever.

Adds tri-state SiteHealthReport.StartupReconcilerActive (null = not reported,
i.e. the central self-report or a pre-row-73 site build; true = running;
false = booted without it), latched once at boot by the Host in both branches.
Additive proto field 29 (BoolValue, vendored C# regenerated via
docker/regen-proto.sh centralcontrol; csproj restored, verified empty diff).
The skip branch also raises its log to Error — no configuration makes the
fetcher legitimately absent on a site node — and writes a best-effort
deployment/Error site event so the evidence outlives the boot log.

Central treatment: a "No self-heal" WARNING badge on the site tile, not a
failure. Config still reaches the node via LocalDb replication while it is up;
what it lost is recovery from a deploy that landed while it was down — degraded,
not offline. Only the explicit false is badged (badging null would light up the
whole fleet the moment central is upgraded ahead of the sites), and offline
sites are skipped, matching the replication-backlog badge.

Fail-before evidence: with the row-80 gate behaviourally reverted,
UnreplicatedSite_HealthReport_ReportsNullForBothLocalDbFields fails
"Expected: null / Actual: False" and 3 of the 8 new reporter unit tests fail.

Tests: HealthMonitoring 128, Host 518, CentralUI 1051, Communication 717,
Commons 715, SiteEventLogging 81 — all green.
2026-08-15 21:40:48 -04:00

346 lines
16 KiB
C#

using Microsoft.AspNetCore.Builder;
using Microsoft.Extensions.Configuration;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Hosting;
using Microsoft.Extensions.Logging.Abstractions;
using Microsoft.Extensions.Options;
using ZB.MOM.WW.ScadaBridge.HealthMonitoring;
using ZB.MOM.WW.Telemetry;
using ZB.MOM.WW.LocalDb;
using ZB.MOM.WW.LocalDb.Replication;
using ZB.MOM.WW.ScadaBridge.Host;
using ZB.MOM.WW.ScadaBridge.Host.Actors;
namespace ZB.MOM.WW.ScadaBridge.Host.Tests;
/// <summary>
/// LocalDb Phase 1 (Task 3) — locks the consolidated site database into the REAL site
/// composition root.
/// <para>
/// These tests build the actual container from
/// <see cref="SiteServiceRegistration.Configure"/> rather than asserting on a
/// hand-assembled <c>ServiceCollection</c>. That is deliberate and load-bearing: this
/// family has now shipped three wiring defects that only a container-built graph would
/// have caught — the Secrets Akka replicator that registered into a no-op sink (0.2.0),
/// the singleton cycle that deadlocked a real host (0.2.2), and Secrets.Ui's missing
/// <c>IAuditWriter</c> that 500'd only behind a login wall (ScadaBridge#22). A test that
/// merely checks "AddZbLocalDb was called" would have passed in every one of those cases.
/// </para>
/// </summary>
public class SiteLocalDbWiringTests : IDisposable
{
private readonly WebApplication _host;
private readonly string _tempDbPath;
private readonly string _tempSiteDbPath;
private readonly string _tempTrackingDbPath;
public SiteLocalDbWiringTests()
{
var stamp = Guid.NewGuid();
_tempDbPath = Path.Combine(Path.GetTempPath(), $"scadabridge_localdb_{stamp}.db");
_tempSiteDbPath = Path.Combine(Path.GetTempPath(), $"scadabridge_site_{stamp}.db");
_tempTrackingDbPath = Path.Combine(Path.GetTempPath(), $"scadabridge_track_{stamp}.db");
var builder = WebApplication.CreateBuilder();
builder.Configuration.Sources.Clear();
builder.Configuration.AddInMemoryCollection(new Dictionary<string, string?>
{
["ScadaBridge:Node:Role"] = "Site",
["ScadaBridge:Node:NodeName"] = "node-a",
["ScadaBridge:Node:NodeHostname"] = "test-site",
["ScadaBridge:Node:SiteId"] = "TestSite",
["ScadaBridge:Node:RemotingPort"] = "0",
["ScadaBridge:Node:GrpcPort"] = "0",
["ScadaBridge:Database:SiteDbPath"] = _tempSiteDbPath,
["ScadaBridge:OperationTracking:ConnectionString"] = $"Data Source={_tempTrackingDbPath}",
["ScadaBridge:Cluster:SeedNodes:0"] = "akka.tcp://scadabridge@localhost:2551",
["ScadaBridge:Cluster:SeedNodes:1"] = "akka.tcp://scadabridge@localhost:2552",
// The consolidated site database, configured exactly like the PASSIVE half of
// the rig's replicated pair (site-a node-b): an ApiKey and no PeerAddress.
//
// The key is what makes this a replicating node, and therefore what makes the
// CDC registration assertions below apply at all — capture is installed only
// when replication is configured. The absent PeerAddress keeps this fixture the
// default-OFF pin at the same time: nothing dials, so the engine must still
// resolve and idle rather than throw or report a connection.
["LocalDb:Path"] = _tempDbPath,
["LocalDb:Replication:ApiKey"] = "wiring-test-localdb-sync-key",
});
builder.Services.AddGrpc();
builder.Services.AddSingleton<ZB.MOM.WW.ScadaBridge.Communication.Grpc.SiteStreamGrpcServer>();
SiteServiceRegistration.Configure(builder.Services, builder.Configuration);
AkkaHostedServiceRemover.RemoveAkkaHostedServiceOnly(builder.Services);
_host = builder.Build();
}
public void Dispose()
{
(_host as IDisposable)?.Dispose();
foreach (var path in new[] { _tempDbPath, _tempSiteDbPath, _tempTrackingDbPath })
{
try { File.Delete(path); } catch { /* best effort */ }
}
GC.SuppressFinalize(this);
}
[Fact]
public void Site_Resolves_ILocalDb()
{
Assert.NotNull(_host.Services.GetService<ILocalDb>());
}
[Fact]
public void Site_LocalDb_RegistersBothReplicatedTables()
{
// The whole point of Phase 1. If onReady silently failed to register these,
// storage would still work and every other test would pass — the pair would
// just never replicate anything. Assert on the library's own view of what is
// registered, not on our belief that we called it.
var db = _host.Services.GetRequiredService<ILocalDb>();
Assert.Contains("OperationTracking", db.ReplicatedTables.Keys);
Assert.Contains("site_events", db.ReplicatedTables.Keys);
}
[Fact]
public void Site_LocalDb_ReplicatedTablesHaveExplicitPrimaryKeys()
{
// RegisterReplicated refuses a table with no explicit PK, so reaching this
// assertion at all proves the DDL ran before registration. The PK names are
// pinned because they are what last-writer-wins conflict resolution keys on.
var db = _host.Services.GetRequiredService<ILocalDb>();
Assert.Equal(
["TrackedOperationId"], db.ReplicatedTables["OperationTracking"].PkColumns);
Assert.Equal(["id"], db.ReplicatedTables["site_events"].PkColumns);
}
[Fact]
public void Site_LocalDb_IsASingleton()
{
// AddZbLocalDb is one-DB-per-process and first-registration-wins. Two
// instances would mean two SQLite handles and two trigger installations
// against the same file.
var first = _host.Services.GetRequiredService<ILocalDb>();
var second = _host.Services.GetRequiredService<ILocalDb>();
Assert.Same(first, second);
}
[Fact]
public void Site_Replication_IsRegistered_AndInertWithNoPeer()
{
// Task 7's default-OFF pin. This fixture configures NO
// LocalDb:Replication:PeerAddress, which is the shipping default, so the
// engine must resolve cleanly and sit idle rather than throw, retry, or
// report a connection. "Default-off" here means inert, NOT unregistered —
// the services are always in the graph.
var status = _host.Services.GetService<ISyncStatus>();
Assert.NotNull(status);
Assert.False(status!.Connected);
Assert.Null(status.PeerNodeId);
Assert.Null(status.LastSyncUtc);
}
[Fact]
public void Site_Replication_OplogBacklog_IsZero_NotNull_OnAHealthyIdleNode()
{
// OplogBacklog is deliberately nullable: null means "unknown" (no provider
// wired, or the poll failed) and must never be reported as a healthy 0. On a
// node whose local database is present and readable, a real 0 proves the
// backlog provider is actually wired to the oplog — a null here would mean
// Task 9's health signal is about to publish "unknown" forever.
var status = _host.Services.GetRequiredService<ISyncStatus>();
Assert.Equal(0L, status.OplogBacklog);
}
[Fact]
public void Site_Replication_HealthBridge_IsRegisteredAsAHostedService()
{
// Task 9. Registered via a factory lambda, so ImplementationType is null and the
// only honest assertion is on the resolved instance.
var hosted = _host.Services.GetServices<IHostedService>();
Assert.Contains(hosted, h => h is LocalDbReplicationStatusReporter);
}
[Fact]
public void Site_Replication_HealthBridge_PublishesStatusOntoTheHealthReport()
{
// End-to-end through the REAL collector: probe once, then collect. This is what
// catches a bridge that is registered but wired to the wrong provider — the
// registration test above would pass either way.
//
// This fixture IS a replicating node: it carries LocalDb:Replication:ApiKey (the
// passive half of the rig's site-a pair), and configured-ness is PeerAddress OR
// ApiKey. So false/0 is the correct, meaningful answer here — "replication is on,
// no session is up yet, and the oplog really is empty". Register row 80 is about
// the OTHER case, which is pinned by UnreplicatedSiteLocalDbHealthTests: a node
// with NEITHER key must report null/null, because the engine's inert false/0 there
// means "not applicable", not "disconnected".
//
// Do not weaken this assertion to null without checking which fixture you are in —
// conflating "no PeerAddress" with "replication unconfigured" is precisely the
// mistake that produced row 80.
var reporter = _host.Services.GetServices<IHostedService>()
.OfType<LocalDbReplicationStatusReporter>()
.Single();
var collector = _host.Services.GetRequiredService<ISiteHealthCollector>();
reporter.Probe();
var report = collector.CollectReport("TestSite");
Assert.False(report.LocalDbReplicationConnected);
Assert.Equal(0L, report.LocalDbOplogBacklog);
}
[Fact]
public void Site_HealthReport_LocalDbFields_AreNull_BeforeTheReporterHasRun()
{
// Null means "no data yet", and must be distinguishable from a real
// "disconnected, zero backlog". A collector that defaulted these to false/0 would
// report an unwired node as a healthy connected one.
var collector = new SiteHealthCollector();
var report = collector.CollectReport("TestSite");
Assert.Null(report.LocalDbReplicationConnected);
Assert.Null(report.LocalDbOplogBacklog);
}
[Fact]
public void Site_Telemetry_Allowlists_TheLocalDbReplicationMeter()
{
// ZbTelemetryOptions.Meters is an ALLOWLIST: an unlisted Meter's instruments are
// never observed, with no error and no missing-metric warning anywhere. The
// replication meter was silently absent from the rig's /metrics scrape until the
// live gate looked for it, so pin the allowlist rather than trusting the next
// reader to remember.
//
// This asserts on the allowlist constant, not on OTel's observed-meter set:
// AddZbTelemetry applies its options to a local instance and never registers
// IOptions, so there is nothing resolvable to interrogate. The end-to-end proof
// that localdb_* actually reaches /metrics is the Task 12 live gate; this test
// exists to stop the entry being dropped again.
Assert.Contains(LocalDbMetrics.MeterName, SiteServiceRegistration.ObservedMeters);
}
[Fact]
public void Site_LocalDb_CreatesThePhase2Tables_InTheConsolidatedFile()
{
// Phase 2 (Task 7). The nine configuration tables and the store-and-forward buffer
// now live in the consolidated file rather than in scadabridge.db and
// store-and-forward.db. Asserting through the REAL composition root is what proves
// OnReady applies both schemas — the stores themselves also call Apply, so a test
// that went through a store would pass even if OnReady never touched them.
var db = _host.Services.GetRequiredService<ILocalDb>();
var tables = TableNames(db);
// One from each Phase 1 schema, then every Phase 2 table.
Assert.Contains("OperationTracking", tables);
Assert.Contains("site_events", tables);
Assert.Contains("sf_messages", tables);
foreach (var table in new[]
{
"deployed_configurations", "static_attribute_overrides", "shared_scripts",
"external_systems", "database_connections", "notification_lists",
"data_connection_definitions", "smtp_configurations", "native_alarm_state",
})
{
Assert.Contains(table, tables);
}
}
[Fact]
public void Site_LocalDb_RegistersExactlyTheTenReplicatedTables()
{
// The Phase 2 cutover. This is an EXACTLY check rather than a Contains in both
// directions, and both directions are load-bearing.
//
// Too few: the table silently stops replicating. Storage still works, every other
// test still passes, and the pair just quietly diverges — the failure mode Phase 1
// existed to end.
//
// Too many: notification_lists and smtp_configurations must NOT be here. They are
// permanently empty by design, and registering them would open a standing
// replication channel whose only historical payload was plaintext SMTP passwords.
// A Contains-based test would never catch that.
var db = _host.Services.GetRequiredService<ILocalDb>();
Assert.Equal(
[
// Ordinal order: '_' (0x5F) sorts before 'b' (0x62), so
// data_connection_definitions precedes database_connections.
"OperationTracking", "data_connection_definitions", "database_connections",
"deployed_configurations", "external_systems", "native_alarm_state",
"sf_messages", "shared_scripts", "site_events", "static_attribute_overrides",
],
db.ReplicatedTables.Keys.OrderBy(k => k, StringComparer.Ordinal).ToArray());
}
[Fact]
public void Site_LocalDb_DoesNotReplicateTheCentralOnlyNotificationTables()
{
// Stated separately from the exact-set test above because this one is a security
// property, not a wiring property, and deserves to fail with its own name. The
// tables exist (SiteStorageSchema creates them) — they simply must never be
// captured. See SiteLocalDbSetup.OnReady for the rationale.
var db = _host.Services.GetRequiredService<ILocalDb>();
Assert.DoesNotContain("notification_lists", db.ReplicatedTables.Keys);
Assert.DoesNotContain("smtp_configurations", db.ReplicatedTables.Keys);
}
[Fact]
public void Site_LocalDb_ReplicatedPhase2TablesHaveTheirExpectedPrimaryKeys()
{
// RegisterReplicated refuses a table with no explicit PK, so reaching these
// assertions proves the DDL ran before registration. The composite keys are the
// interesting ones: LWW conflict resolution keys on the FULL PK, so a wrong or
// truncated key set would silently collapse distinct rows into one.
var db = _host.Services.GetRequiredService<ILocalDb>();
Assert.Equal(["id"], db.ReplicatedTables["sf_messages"].PkColumns);
Assert.Equal(
["instance_unique_name"], db.ReplicatedTables["deployed_configurations"].PkColumns);
Assert.Equal(
["instance_unique_name", "attribute_name"],
db.ReplicatedTables["static_attribute_overrides"].PkColumns);
Assert.Equal(
["instance_unique_name", "source_canonical_name", "source_reference"],
db.ReplicatedTables["native_alarm_state"].PkColumns);
}
private static HashSet<string> TableNames(ILocalDb db)
{
using var connection = db.CreateConnection();
using var cmd = connection.CreateCommand();
cmd.CommandText = "SELECT name FROM sqlite_master WHERE type = 'table'";
using var reader = cmd.ExecuteReader();
var names = new HashSet<string>(StringComparer.Ordinal);
while (reader.Read()) names.Add(reader.GetString(0));
return names;
}
[Fact]
public void Site_LocalDb_CreatesTheConfiguredFile()
{
// Resolving is what triggers onReady; the file should exist afterwards at the
// configured path and not somewhere CWD-relative.
_ = _host.Services.GetRequiredService<ILocalDb>();
Assert.True(File.Exists(_tempDbPath),
$"Expected the consolidated site database at '{_tempDbPath}'.");
}
}