Regions and Nodes
Even in a debugging product, latency matters. Regions are the logical entry points, while nodes are the concrete relay instances that shape the tunnel experience.
What a region is
Logical locations like us-east or local that help reason about traffic flow.
In a mature deployment, users choose a region close to themselves or close to the systems generating traffic.
This affects response time, webhook timeout risk, and how predictable the development experience feels.
Nodes and Instances
| Concept | Meaning | MVP Status |
|---|---|---|
| Region | Logical entry point for traffic. | Defaults to a single 'local' region. |
| Node | Concrete relay instance serving tunnels. | Behaves like a single primary node. |
| Selection | How an operator chooses an entry point. | Planned for multi-region controls. |
Why latency matters
Tunnel products are network products. Slow regions make every request feel heavier and increase webhook retries for providers with tight timeouts.
Provider Timeouts
High latency can trigger automatic retries from webhook providers like Stripe or GitHub.
Debug Flow
A 500ms overhead on every hot-reload or request cycle breaks the developer‘s "state of flow.“
How to choose a region
• For self-hosted teams, start with one known-good region matching your engineers.
• Add more nodes only after you can observe health and route behavior clearly.
• If debugging webhooks, choose the region that best matches the provider‘s data center.
The repository is currently single-region. We explain this model now to help teams self-host with realistic expectations for future scaling.