Cephalon.Observability.HuaweiCloud
Maturity:
M1· Ownership: cephalon-managed · Family:observability· See audit, matrix.
Cephalon.Observability.HuaweiCloud adds Huawei Cloud-hosted OTLP defaults for Cephalon hosts.
What it owns
Section titled “What it owns”- host-builder registration for OTLP logs, metrics, and traces over the shared
Engine:Observability:Telemetrycontract when a shared collector endpoint is already in place - explicit hosted Huawei Cloud resource defaults for ECS, CCE, and FunctionGraph deployments
- an opt-in Huawei Cloud APM managed trace-ingestion path over OTLP/gRPC plus the required
Authenticationheader - startup diagnostics that summarize the active Huawei Cloud export mode without logging secrets
Main surfaces
Section titled “Main surfaces”Configuration/HuaweiCloudTelemetryExportOptions.csHosting/HuaweiCloudHostApplicationBuilderExtensions.csHosting/HuaweiCloudSummaryHostedService.cs
Source structure
Section titled “Source structure”ConfigurationHosting
How it fits
Section titled “How it fits”This package stays outside Cephalon.Engine and Cephalon.Observability on purpose. The engine still owns diagnostics names and lifecycle signals, Cephalon.Observability still owns the shared telemetry contract, and this companion package layers Huawei Cloud-specific hosted defaults and managed APM trace-ingestion wiring on top of the same OTLP baseline.
The Huawei Cloud slice is intentionally explicit about its two modes. If a host already has a shared OTLP collector or gateway path, the package keeps using the shared Endpoint or UseSelfHostedDefaults flow and only adds hosted Huawei Cloud resource defaults. If the host wants Huawei Cloud APM direct trace ingestion instead, it can enable Engine:Observability:Telemetry:HuaweiCloud:UseApmManagedTraceIngestion, keep the protocol on otlp or otlp/grpc, and configure both ApmEndpoint and AuthenticationToken. That direct managed path is trace-only for now; logs and metrics stay on the collector or another runtime-specific route.