fix(telemetry): export the Galaxy driver's spans and galaxy.events.* counters (#892)
ZbTelemetryOptions.Meters/ActivitySources are strict allowlists. The host named only ZB.MOM.WW.OtOpcUa and the LocalDb meter, so the in-process Galaxy driver's galaxy.* spans and its event pump's galaxy.events.* counters (both on "ZB.MOM.WW.OtOpcUa.Driver.Galaxy") were silently dropped from /metrics and OTLP. Found exercising OTLP end to end against a collector (scadaproj#13). - Galaxy driver: public GalaxyDriverTelemetryNames (MeterName / ActivitySourceName); GalaxyTelemetry and EventPump reference it. - Host: ObservabilityExtensions.ObservedMeters / ObservedActivitySources include it. - Host.IntegrationTests: ObservabilityAllowlistTests assert on the listener side (span sampled, instrument enabled) with an unregistered-name negative control; dropping the Galaxy entries fails 3 of 8. Claude-Session: https://claude.ai/code/session_01GT67ACmpMR7b9LGRPLijMi
This commit is contained in:
@@ -1,6 +1,7 @@
|
||||
using Microsoft.Extensions.Configuration;
|
||||
using ZB.MOM.WW.LocalDb.Replication;
|
||||
using ZB.MOM.WW.OtOpcUa.Commons.Observability;
|
||||
using ZB.MOM.WW.OtOpcUa.Driver.Galaxy;
|
||||
using ZB.MOM.WW.Telemetry;
|
||||
|
||||
namespace ZB.MOM.WW.OtOpcUa.Host.Observability;
|
||||
@@ -14,6 +15,31 @@ namespace ZB.MOM.WW.OtOpcUa.Host.Observability;
|
||||
/// </summary>
|
||||
public static class ObservabilityExtensions
|
||||
{
|
||||
/// <summary>
|
||||
/// Every meter the host exports. <c>ZbTelemetryOptions.Meters</c> is a STRICT allowlist (no
|
||||
/// wildcard): the LocalDb replication meter publishes its <c>localdb.sync.*</c> /
|
||||
/// <c>localdb.oplog.depth</c> series on driver nodes, and the in-process Galaxy driver its
|
||||
/// <c>galaxy.events.*</c> counters — each is silently absent from <c>/metrics</c> and OTLP unless
|
||||
/// named here. Adding a meter anywhere in the product means adding it here.
|
||||
/// </summary>
|
||||
public static readonly string[] ObservedMeters =
|
||||
[
|
||||
OtOpcUaTelemetry.MeterName,
|
||||
LocalDbMetrics.MeterName,
|
||||
GalaxyDriverTelemetryNames.MeterName,
|
||||
];
|
||||
|
||||
/// <summary>
|
||||
/// Every activity source the host exports over OTLP. Also a strict allowlist: the Galaxy
|
||||
/// driver's <c>galaxy.*</c> gateway-call spans were dropped until scadaproj#13 because its source
|
||||
/// was missing here.
|
||||
/// </summary>
|
||||
public static readonly string[] ObservedActivitySources =
|
||||
[
|
||||
OtOpcUaTelemetry.ActivitySourceName,
|
||||
GalaxyDriverTelemetryNames.ActivitySourceName,
|
||||
];
|
||||
|
||||
/// <summary>Adds OtOpcUa observability (metrics and tracing) to the service collection.</summary>
|
||||
/// <param name="services">The service collection to add observability services to.</param>
|
||||
/// <param name="configuration">
|
||||
@@ -28,12 +54,9 @@ public static class ObservabilityExtensions
|
||||
return services.AddZbTelemetry(o =>
|
||||
{
|
||||
o.ServiceName = "otopcua";
|
||||
// o.Meters is a STRICT allowlist — ZbTelemetry adds only the named meters, with no
|
||||
// wildcard. The LocalDb replication engine publishes under its own meter name, so
|
||||
// without this entry its localdb.sync.* / localdb.oplog.depth series are silently absent
|
||||
// from /metrics on driver nodes (the exact omission a ScadaBridge live gate caught).
|
||||
o.Meters = [OtOpcUaTelemetry.MeterName, LocalDbMetrics.MeterName];
|
||||
o.ActivitySources = [OtOpcUaTelemetry.ActivitySourceName];
|
||||
// Strict allowlists — see ObservedMeters / ObservedActivitySources.
|
||||
o.Meters = ObservedMeters;
|
||||
o.ActivitySources = ObservedActivitySources;
|
||||
if (Enum.TryParse<ZbExporter>(configuration["OtOpcUa:Telemetry:Exporter"], ignoreCase: true, out var exporter))
|
||||
o.Exporter = exporter;
|
||||
var otlp = configuration["OtOpcUa:Telemetry:OtlpEndpoint"];
|
||||
|
||||
Reference in New Issue
Block a user