v3.8.2 — Stable Release

Deployment Center

Ship to production securely. Configure infrastructure, manage pipelines, and monitor releases from a single command line or dashboard.

📦 Deployment Methods

                    
                    # Initialize deployment environment

                    divisions init --env production

                    # Validate configuration & run checks

                    divisions validate --strict

                    # Deploy to target region

                    divisions deploy --region us-east-1 --auto-scale

Install the CLI via npm i -g @divisions/cli or download from the releases page.

                    
                    curl -X POST https://api.divisions.io/v3/deploy \\
                      -H "Authorization: Bearer $DIV_TOKEN" \\
                      -H "Content-Type: application/json" \\
                      -d '{"target": "prod", "strategy": "rolling", "canary_pct": 10}'
                

Authenticate using service account tokens. See rate limits & pagination in the API docs.

                    
                    // Web UI Workflow

                    1. Navigate to Dashboard → Environments → Production

                    2. Select branch/tag under Source Mapping

                    3. Click "Deploy" → Configure traffic routing → Confirm

                    4. Monitor real-time rollout in Activity Log

UI deployments support preview environments, one-click rollbacks, and team approval gates.

                    
                    # terraform init && terraform plan

                    module "divisions" {
                      source = "divisions-io/core"
                      version = "~> 3.8"
                      environment = "production"
                      replicas = 3
                      auto_scaling = true
                    }
                

Infrastructure as Code modules are hosted in the official registry. State locking & remote backends supported.

🔄 Deployment Pipeline

1. Initialize & Configure

Define environment variables, secrets, and routing rules. The CLI validates schema against your project manifest.

2. Build & Test

Automated containerization, dependency resolution, and integration test suites run in isolated sandboxes.

3. Security Scan

SAST, SCA, and container image vulnerability checks block deployments that exceed defined risk thresholds.

4. Staging Rollout

Canary traffic is gradually shifted. Automated health checks and performance baselines validate stability.

5. Production Promotion

Full traffic migration with zero-downtime. Automatic rollback triggers on error rate spikes or latency degradation.

📟 Live Deployment Log

divisions deploy — us-east-1

[10:42:01] INFO Initializing deployment context...

[10:42:03] INFO Resolving dependencies (47 packages)

[10:42:05] PASS Security scan completed (0 critical, 2 low)

[10:42:08] INFO Building container image: divisions/core:3.8.2

[10:42:15] PASS Image pushed to registry (digest: sha256:7f3a...)

[10:42:18] INFO Updating load balancer rules...

[10:42:22] WARN High memory utilization on node-03 (82%)

[10:42:25] PASS Health checks passed (5/5 endpoints)

[10:42:27] PASS Deployment successful. Rollout ID: rl_89a2c1

[10:42:27]

⚙️ Requirements & Quotas

🌐 Network

Outbound HTTPS to api.divisions.io:443. Inbound ports 80/443 for public endpoints. VPC peering optional.

💾 Storage

Minimum 20GB ephemeral disk per node. Persistent volumes auto-provisioned based on dataset size.

🔑 Permissions

Service account requires `deploy:write`, `logs:read`, and `secrets:access` scopes. IAM roles supported.

📈 Quotas

10 concurrent deployments per org. 50GB log retention (default). Upgrade for custom limits & archival.

❓ Deployment FAQ

How do I rollback a failed deployment? ▼

Run `divisions rollback --latest` or specify a rollout ID. The system will restore the previous stable state, preserve logs, and notify your configured webhook endpoints.

Can I deploy to multiple regions simultaneously? ▼

Yes. Use `--multi-region` with a priority queue. Deployments fan out sequentially based on health thresholds to ensure global consistency.

What happens if a security scan fails? ▼

The pipeline halts automatically. You'll receive a detailed report via email/Slack. Override is possible with `--force` but requires admin approval in audit logs.

Is zero-downtime guaranteed? ▼

For rolling and blue-green strategies, yes. Traffic shifting is gradual and reversible. Downtime only occurs during maintenance windows or forced hard rollbacks.