Skip to content

Environment Requests

Environment Requests provide a self-service workflow for provisioning deployment environments. Developers request the environment they need, administrators review and approve, and Farm tracks the full lifecycle from creation through expiry.

Core Concepts

Environment Request

An environment request is a formal submission to create a new deployment environment. It captures the desired configuration (type, tier, TTL) and moves through an approval workflow before provisioning begins.

Ephemeral vs Persistent

Ephemeral environments are short-lived and include a time-to-live (TTL) after which they are automatically torn down. Use ephemeral environments for feature previews, pull request testing, and time-boxed experiments. Persistent environments remain available until manually removed and are suited for shared staging or long-running integration testing.

Environment Tier

Tiers control the resource allocation for the provisioned environment:

Tier CPU Memory Use Case
Small 0.5 512 MB Lightweight smoke tests
Medium 2 2 GB Integration and functional testing
Large 4 8 GB Performance and load testing

Time to Live (TTL)

TTL defines how long an ephemeral environment remains active after provisioning completes. When the TTL expires, the environment transitions to expired and its resources are reclaimed. Persistent environments do not require a TTL.

Requesting a New Environment

Step 1: Open the Request Form

Navigate to Environment Requests in the sidebar and click New Request.

Step 2: Fill in Request Details

Field Description
Name A descriptive name (e.g., feature-xyz-preview)
Description Justification for why the environment is needed
Type ephemeral for short-lived or persistent for long-lived
Tier Resource tier: small, medium, or large
TTL (hours) How long the environment should remain active (ephemeral only)
Component Optional link to the catalog component this environment serves

Step 3: Submit

Click Submit Request. The request enters pending status and is visible to administrators for review.

Approval Workflow

Environment requests follow a defined state machine:

pending --> approved --> provisioning --> active --> expired
  |
  +--> rejected

Status Descriptions

Status Description
Pending Request submitted and awaiting administrator review
Approved Administrator approved the request; provisioning will begin
Rejected Administrator rejected the request with a reason
Provisioning Environment resources are being created
Active Environment is running and available for use
Expired Environment has been torn down (TTL elapsed or manual expiry)

For Administrators

  • Navigate to the pending requests queue to review incoming requests.
  • Click Approve to authorize provisioning. Add an optional comment.
  • Click Reject to deny the request. Provide a reason so the requester understands why.
  • For active environments that need early termination, click Expire to reclaim resources.

Monitoring Environment Status

The Environment Requests dashboard shows all requests grouped by status. Each request card displays:

  • Environment name and description
  • Type and tier badges
  • Current status with timestamp
  • TTL countdown for active ephemeral environments
  • Linked component (if any)
  • Reviewer name and comment (after review)

Click any request to view full details, including the status history and provisioning logs.

Updating a Pending Request

While a request is still in pending status, you can update the name, description, and TTL. Navigate to the request detail page and click Edit. Once a request has been approved or rejected, it can no longer be modified.

Best Practices

  • Use ephemeral environments for feature work. Set a TTL that matches your development cycle (24-72 hours) to avoid resource waste.
  • Choose the smallest tier that meets your needs. Start with small for smoke tests and scale up only when required for integration or load testing.
  • Write clear descriptions. Help administrators understand why the environment is needed so they can approve quickly.
  • Link requests to components. Associating a request with a catalog component provides context and improves traceability.
  • Monitor TTL expiry. Plan your testing around the TTL window. If you need more time, update the request before it expires.
  • Clean up persistent environments. Since persistent environments do not auto-expire, coordinate with your team to remove them when no longer needed.