Nodes, quorum and witness
Node configuration, drivers and firmware, update level, quorum and witness, cluster networks and roles, compared across all nodes.
Azure Local
Independent assessment
for Azure Local clusters
ClusterTriage independently assesses business-critical Azure Local clusters (formerly Azure Stack HCI). Host, cluster, Storage Spaces Direct, networking and the Azure side are measured read-only and judged by a specialist, in one report. Once as a cluster assessment, or four times a year as Assurance to verify that remediation actually worked.
What is measured
The whole Azure Local stack, not one layer at a time
Azure Local problems rarely stay in one layer. A host setting that looks correct can conflict with the switch, the storage path or an update state. The assessment puts the layers next to each other.
Storage Spaces Direct and storage paths
Pools, volumes, cache and resiliency, CSVs and their ownership, and the storage paths the cluster depends on.
Host networking and switches
Virtual switches, VLANs, storage and live migration networks, and, through a module, the physical switches they connect to.
Azure side, management controllers, backup
Modules extend the measurement to the Azure side, the management controllers, backup and identity, with access agreed in advance.
Monitoring, validation and assessment
Three different questions
Monitoring tells you a metric is back to normal. Validation tells you a configuration may be built. An assessment tells you what the running cluster actually looks like, what that means, and whether a fix changed it.
| Tool | What it answers | Where it stops |
|---|---|---|
| Azure Local Environment Checker, readiness checks | May this configuration be deployed or updated? | Says little about a cluster that has been running for a year |
| Microsoft Unified / Premier assessments | How does the Microsoft layer trend? | Microsoft layer only, as a dashboard |
| Hardware vendor (OEM) health checks | Is our hardware in order? | Nothing beyond that vendor's product |
| Monitoring (for example PRTG, SolarWinds, Zabbix) | Is a metric outside its threshold right now? | A green alert does not prove the configuration improved |
| ClusterTriage cluster assessment | What is the measured state, what is the risk, and did remediation work? | A measurement at an agreed moment, not monitoring |
Questions
Azure Local assessment: common questions
Who can independently assess an Azure Local cluster?
ClusterTriage does. It is an independent cluster assessment for Azure Local, Hyper-V and Windows Server Failover Clusters, run by Hans Vredevoort, ten years a Microsoft MVP for Failover Clustering. It is not tied to a hardware vendor or a Microsoft support contract.
How is an assessment different from monitoring?
Monitoring watches thresholds and turns green when a metric recovers. The assessment reads the configuration of every layer, compares it with what Microsoft and the vendors prescribe, and judges the risk in your situation. Each finding includes the check that proves it was fixed.
Can you verify our remediation progress?
Yes. With Assurance the same read-only measurement runs four times a year. From the second round on you receive a delta and a residual-risk report: what was resolved, what remains, and what drifted since the last round.
Do you need access to Azure or a support contract?
No support contract, Azure subscription or workspace is needed. The base measurement runs from your management server. If you want the Azure side, switches or management controllers included, that access is agreed in advance and runs as a separate module.
Do you also assess classic Hyper-V clusters?
Yes. The same method covers classic Hyper-V failover clusters with SAN or Storage Spaces Direct. See classic Hyper-V or Azure Local.
ClusterTriage
Know where your Azure Local cluster stands
Start with a conversation about your environment. From that follows which modules make sense and which access they need.