About¶
azurelocal-scom-mp is a community project under the AzureLocal GitHub organization that delivers production-grade health monitoring for Azure Local (HCI) infrastructure in two parallel tracks:
| Track | Format | Purpose |
|---|---|---|
| SCOM Management Pack | Sealed .mp + override .xml | Classic SCOM authoring — unit/aggregate/dependency/Distributed Application rollup |
| Azure Monitor Health Models | ARM/Bicep + Service Group | Native cloud-side health propagation using Azure Monitor's preview Health Models feature |
Both tracks share the same conceptual health model — they monitor the same components, agree on what "healthy" means, and propagate health the same way. They differ only in implementation surface.
Project Goals¶
- Cover the full Azure Local infrastructure footprint — the on-prem cluster plus every Azure-side resource that the deployment provisions (managed identity, Key Vault, Storage Account, Custom Location, Arc-enabled servers, Resource Bridge, etc.). See the Roadmap for the complete entity inventory.
- Ship parity across SCOM and Azure Monitor — every signal that exists in one track has an equivalent in the other. Customers migrating from SCOM to Azure Monitor see the same health story.
- Customizable without forking — every threshold, alert severity, and behavior is parameterized. Customers ship their own override pack (SCOM) or
*.bicepparamfile (Azure Monitor) and customizations survive upgrades. See Customization. - Document the health model first, the implementation second — the design (this site) is the product; the XML and Bicep are downstream artifacts.
- Conform to the AzureLocal/platform standards — same docs stack, same CI workflows, same release tooling as the rest of the org.
Scope¶
In scope (Phase 1–6):
- Azure Local cluster, nodes, storage pool, volumes, network ATC / network intent
- Cluster-resident platform services: Arc Resource Bridge, MOC, AKS Arc platform
- Azure-side resources deployed by Azure Local: HCI cluster resource, Arc-enabled servers, Custom Locations, Managed Identities, Service Principals, Key Vault, Storage Account, storage containers, logical networks, Update Manager
Out of scope (Phase 1–6) — but on the Roadmap as future MPs:
- Workloads running on Azure Local (VMs, AKS Arc pods/containers, application-level monitoring)
- These would be future Management Packs that take a dependency on this infrastructure health model — e.g. "VM workload X is impacted because the cluster is unhealthy."
Project Structure¶
azurelocal-scom-mp/
├── decisions/ # ADRs (architectural decision records) for the health model design
├── diagrams/ # Mermaid + draw.io source diagrams
├── docs/ # MkDocs Material site (this site)
├── src/ # SCOM MP authoring (VSAE) + Azure Monitor ARM/Bicep + KQL
├── PLAN.md # Full implementation roadmap
├── REFERENCES.md # Annotated link library (~100 references)
└── STANDARDS.md # Pointer to AzureLocal/platform standards
Maintainers¶
See CODEOWNERS.