Step 1: Navigate to Application Instance

Go to Applications → Select your application → Select an instance

Step 2: Click "Deploy"

The deploy dialog will open with configuration options.

Step 3: Select Target Clusters

Choose one or more clusters from the environment. You can deploy to multiple clusters simultaneously for high availability.

☑ production-us-east-1
☑ production-us-west-2
☐ production-eu-west-1

Step 4: Specify Version

• Git Source: Branch, commit hash, or tag (e.g., main, abc123, v1.2.3)
• Container: Image tag (e.g., latest, v1.2.3, sha-abc123)
• Helm Chart: Chart version (e.g., 1.2.3)

Step 5: Review and Deploy

Review configuration and click Deploy. Monitor progress in real-time.

Use Cases

• High Availability: Multiple availability zones within a region
• Disaster Recovery: Multiple regions for business continuity
• Global Performance: Deploy closer to users worldwide
• Load Distribution: Distribute traffic across clusters

Configuration

When creating a deployment, select multiple target clusters:

✓ us-east-1-prod (Primary)
✓ us-west-2-prod (Secondary)
✓ eu-west-1-prod (Secondary)

Traffic routing: Latency-based with health check failover

DNS Routing Policies

• Simple: Single cluster receives all traffic
• Latency-based: Route to nearest healthy cluster
• Failover: Primary cluster with automatic failover to secondary
• Weighted: Distribute traffic by percentage across clusters

Method 1: Mark Previous Deployment as Main

Fastest rollback - instantly switch traffic to a previous running deployment.

1. 1. Navigate to Deployments list
2. 2. Find the previous stable version (still running)
3. 3. Click "Mark as Main"
4. 4. Traffic switches immediately (no redeploy needed)

⚡ Rollback time: ~5 seconds

Method 2: Redeploy Previous Version

If the previous deployment was terminated, create a new deployment with the old version.

1. 1. Click "Deploy"
2. 2. Specify previous commit/tag/version
3. 3. Wait for deployment to complete
4. 4. Mark as main

⏱️ Rollback time: 2-5 minutes

List Deployments

GET /api/v1/deployments?instance_id=123

Response:
{
  "deployments": [
    {
      "id": "deploy-456",
      "instance_id": "123",
      "version": "v1.2.3",
      "status": "started",
      "main": true,
      "created_at": "2024-01-15T10:30:00Z",
      "clusters": ["cluster-789"],
      "url": "https://api.example.com"
    }
  ]
}

Create Deployment

POST /api/v1/deployments

{
  "instance_id": "123",
  "version": "v1.2.4",
  "cluster_ids": ["cluster-789", "cluster-790"]
}

Response: 201 Created
{
  "id": "deploy-457",
  "status": "queued"
}

Stop Deployment

POST /api/v1/deployments/deploy-456/stop

Response: 200 OK
{
  "status": "stopping"
}

Mark as Main

POST /api/v1/deployments/deploy-457/mark-main

Response: 200 OK
{
  "main": true,
  "previous_main": "deploy-456"
}

Get Deployment Logs

GET /api/v1/deployments/deploy-456/logs?type=container

Response: 200 OK (streaming)
2024-01-15T10:30:01Z [INFO] Server started on port 3000
2024-01-15T10:30:02Z [INFO] Connected to database
2024-01-15T10:30:03Z [INFO] Ready to accept requests