Map cluster fault domains to the actual rack and site layout

Source facts

Microsoft defines a fault domain as hardware sharing a single failure point; tolerance at a level requires multiple domains at that level. The documented hierarchy includes site, rack, chassis, and node. Nodes are discovered automatically, while additional levels are optional. Storage Spaces uses fault-domain information to place redundant copies on separate failure domains. Microsoft documentation.

Applicability

Use this review when documenting a cluster’s physical placement. Identify real racks, chassis, sites, and shared dependencies. Include only hierarchy levels that describe the deployment rather than inventing unused physical distinctions.

DSE recommendation

Have the facilities and server owners reconcile the logical hierarchy with the equipment inventory. Record each node’s actual placement and the failure event each level represents. Review recent equipment moves for stale mappings. Keep data-copy placement objectives separate from quorum-witness placement, and obtain storage-owner review before changing the configured hierarchy.

Verification

Compare the reported fault-domain hierarchy with a checked physical inventory. Inspect the selected storage configuration through the supported management tools and record the relevant placement evidence. Exercise only the approved failure scenario and retain the observed workload result. Update both the map and its ownership record after accepted hardware moves or site changes.

Official references

Microsoft Learn: Fault domain awareness. Source reviewed September 8, 2026.