Environments

Envless.environments.unexpose

Stop the public feed for an environment

def unexpose(project, environment, product: nil, timeout: nil)

Takes an environment off its public URL by posting {exposed: false} to the same endpoint expose uses. This needs neither virtualEnvsEnabled nor a public product, so it is the safe kill switch: call it without checking any state first.

The public URL starts returning a uniform 404 the moment this lands, with no hint that the environment exists. Calling it on an environment that was never exposed is a no-op that still returns the full resource, so it is safe to run across a whole project during an incident.

Example

ruby
require "envless/sdk"

environment = Envless.environments.unexpose("api", "production")

puts environment[:exposed]

Parameters

ParameterTypeRequiredDescription
projectStringYesProject slug or uniqueId that owns the environment.
environmentStringYesSlug of the environment to unexpose; ids are not accepted.
product:String | SymbolNoProduct slug that disambiguates an ambiguous project slug.
timeout:Integer | Float | RationalNoSeconds each attempt of this call may take, redirects and the response body included, before it raises Envless::NetworkError with timeout? set. Overrides the client's timeout for this call, and 0 turns the limit off.

Returns

environment_resource with exposed false, plus updatedAt.

Worth knowing

  • Turning off the workspace virtualEnvsEnabled flag is the blunt instrument; this route is the per-environment one.
  • A link that was already handed out keeps the ciphertext its holder copied, so rotate the passphrase if the feed leaked.

Required scopes

environments:write

Underlying endpoint

POST /projects/{slug}/environments/{environmentSlug}/expose

Request and response schema, every error code, and a live playground.

In TypeScript

envless.environments.unexpose()

The same method in @goenvless/sdk, with the same request, retries and errors.