Provision a Service
The Service Wizard guides you through provisioning a real Fabric service from a golden path template. The wizard collects your inputs across four steps and submits a provisioning request to Fabric on completion.
New to Nexus? Follow the Quick Start guide which includes service provisioning as part of the end-to-end flow.
Prerequisites
- You are signed in to Nexus with an active workspace — see Workspace setup
- At least one project exists in your workspace — see Create a project
- Your organization role includes
org.services.deploy(Owner, Admin, or Member)
Opening the wizard
Navigate to /create/service, or:
- From the Dashboard → Quick actions → Provision service
- From the Services page → Create from template
- From the Templates page → click Start flow on any template card
- Via the Command palette (
Cmd K→ "Provision service")
Step 1 — Template
Select the project and template for your service.
Choose a project
Select from the dropdown of live Atlas projects in your workspace. If no projects exist, create one first.
Choose a template
Browse the template cards loaded from the Prangana Registry. Each card shows:
- Template title and description
- Number of available versions
- Template type badge
Click a template card to select it. The selected card highlights in indigo.
Use the Command palette (Cmd K) to search templates by name, runtime, or type before opening the wizard.
Click Continue.
Step 2 — Runtime
Configure the template version and deployment parameters.
Version
Select the template version from the dropdown. The latest version is selected by default.
Only versions with hasSchema: true can be provisioned through the wizard. If the parameter form is empty, try a different version. See Templates for details.
Region
Enter the deployment region (e.g. centralindia, westeurope, uksouth). Default is centralindia.
Template parameters
Nexus renders a form from the selected template version's JSON Schema. Fill in all required fields. Common parameters:
| Parameter | Description |
|---|---|
serviceName | Name of the service — used as the repository name and Kubernetes resource name |
cloudProvider | Target cloud (azure, gcp, aws) |
namespace | Kubernetes namespace |
replicaCount | Number of pod replicas |
containerPort | Port the container listens on |
serviceName is required. If it is missing or empty, the wizard will show a validation error and block submission.
Parameter validation runs when you click Continue. Required fields that are empty will be highlighted with an error message.
Click Continue.
Step 3 — Delivery
Choose the rollout strategy for this service:
| Strategy | Description |
|---|---|
| Progressive rollout | Stage-based promotion with automated checks between stages |
| Blue/green swap | Parallel environment with traffic cutover on success |
Fabric currently accepts the provisioning request immediately. The delivery strategy is captured for operator visibility and will influence future rollout orchestration.
Click Continue.
Step 4 — Review
Review the full provisioning summary:
| Field | Value |
|---|---|
| Project | Selected project name |
| Service | Service name from parameters |
| Template | Selected template title |
| Version | Selected version |
| Cloud | Cloud provider from parameters |
| Region | Deployment region |
| Workspace | Active organization name |
Template parameter values are also shown below the summary.
Click Provision service to submit.
What happens after submission
- Nexus sends a
POST /api/fabric/servicesrequest with the full provisioning payload - Fabric accepts the request and returns a
serviceId - Nexus redirects you to the Service detail page at
/services/:serviceId - The service begins provisioning — you can track progress via the stage progress bar and status badges
Monitoring provisioning
On the Service detail page and Dashboard, you can see:
- Service status — Overall lifecycle state (Pending → Provisioning → Running)
- Execution status — Current pipeline run state
- Stage progress — Completed vs. total stages with a progress bar
- Current stage — The stage currently executing
- Last error — Error message if a stage failed
The Dashboard auto-refreshes every 8 seconds when active provisioning is detected.
Typically 3–8 minutes depending on the template and cloud region. The service moves to Running status when all stages complete successfully.
Accessing the service
Once the service reaches Running status, the access URL appears on the Service detail page. This is resolved from:
deploymentUrlset by Fabric- ACI FQDN (
aciFqdn) - ACI IP + port fallback
Troubleshooting
If provisioning fails or the service gets stuck, see Common issues — Services.