Regions
Geography, datacenters, zones, and fault isolation for sell-path nodes
What a region is
A region is what customers see when they pick geography on a plan (for example “Frankfurt”). Technically it maps to:
Country → City → Datacenter → ZoneYou select all four when creating a region. Marketplace region codes (for example a Frankfurt region) describe where you sell capacity. They are separate from the CLI --gateway-region value (AMS01 today), which selects the NodeRings agent gateway used for Liqo peering during nr cluster register.
Regions and zones
| Level | Meaning |
|---|---|
| Country / City | Customer-facing geography |
| Datacenter | Physical facility |
| Zone | Fault-isolation boundary inside that datacenter |
Zone A vs Zone B (and beyond)
Zones represent independent infrastructure within one datacenter, not “primary/secondary” UI toggles:
| Dimension | Often isolated per zone |
|---|---|
| Power feeds | Separate UPS / utility paths |
| Network | Distinct switches or fabric |
| Compute rows | Separate halls or racks |
Zones are a fault-isolation boundary. If you run Zone A and Zone B, a power loss in Zone A should not take Zone B offline. Customers buy a region; you make sure the nodes in it sit in the correct zone.
How to use
Create the region
Open /infrastructure/regions → Create region.
Select the geography
Pick country, city, datacenter, and the zone where your hypervisor nodes physically sit.
Save
The region becomes available to assign to nodes and plans.