On this page

Subhosting Resources

To build Subhosting with Deno Deploy, it helps to understand some key resources within the system. These resources are also represented in the REST API.

overview of subhosting resources

Organizations Jump to heading

Organizations are a container for all data related to a subhosting implementation. Your organization will have a name and an ID. Each organization has an analytics endpoint which can be used to get metrics (such as request count and bandwidth used) from across the organization.

Other Deploy users can be invited to collaborate on an organization, and access tokens can give developers with organization access the ability to modify resources within the org via API. New organizations can be created in the Deploy dashboard.

Projects Jump to heading

Projects act as organizational containers for deployments. A project contains its deployments and the analytics and usage information for those deployments.

Projects are free and can be set up as required.

To track usage by individual users for billing there is an API endpoint that reports analytics (bandwidth usage, request count, etc), per project, with 15 minute granularity.

All deployments (whether within a same project or between different projects) share nothing by default. Projects are a way to organize your deployments and do not cost anything. However analytics are reported on a per-project basis, if you have multiple tenants we recommend setting up a project for each. Particularly if you expect to bill your users for their usage.

Deployments Jump to heading

Deployments: a deployment is a set of configuration, runnable code, and supporting static files that can run on an isolate in Deno Deploy. Deployments have an entry file that can launch a server, can have a Deno KV database associated with them, and can be set up to run on custom domains.

A deployment is an immutable object that consists of:

  • Source code to run
  • Static assets
  • Environment variables
  • Database bindings
  • Other settings

We provide endpoints for querying or streaming build logs and querying or streaming execution logs.

If you need to block or unblock a deployment you can do so by deleting the deployment that you need to block or by unassigning its domains. This will make the deployment unreachable.

The Subhosting system is built so that the behavior or load on one deployment does not affect other deployments. This also applies to different deployments within one organization. Capacity is auto-scaled on demand. If you want to limit resources to a particular deployment or application you can use the analytics API to provide you with detailed metrics (request count, bandwidth, etc) at project level granularity. You can use this to decide whether to shut off deployments and make them unreachable.

NB. Deployments are immutable, however, you can create a new deployment and then remap its domain to the new deployment. The redeploy endpoint can create a new deployment from an existing one with different settings.

Custom domains Jump to heading

Custom domains can be dynamically mapped to deployments, giving them a unique URL (eg mycompany.com).

Before a domain can be used you need to verify ownership and provision or upload TLS certificates.

If you are on the Builder tier you can use wildcard domains. Once you have a wildcard domain registered, you can use it in two ways:

  • Send all requests for *.mycompany.com to a specific deployment
  • (Coming soon) Assign different subdomains (e.g. foo.mycompany.com and bar.mycompany.com) to separate deployments.

Staging vs Production Environments Jump to heading

The Deno Deploy end-user platform automatically creates preview deployments when a developer opens a github pull request, and commits to the “main” branch are automatically turned into production deployments. Although subhosting does not provide github integration out of the box, it has all the primitives you need to define your own semantics for creating preview and production deployments.

Connecting a KV Database Jump to heading

A (KV) database stores key-value pairs You can make a database accessible to a deployment when you make the deployment. KV databases can be used by multiple deployments at the same time.

To use KV with Subhosting:

NB. Deno Cron and Queues do not currently work for Subhosting.

OpenAPI specification and tooling Jump to heading

The OpenAPI specification for the Deploy API can be found here:


This spec document can be used with a large number of OpenAPI-compatible tools. In addition to the documentation for the REST API maintained here, you can find auto-generated API documentation (including a browser-based testing tool) here.