NewObservability & Health Checks

Deterministic
deployments.

Automate your release cycle. SafeDeployer orchestrates zero-downtime updates, executes robust health checks, and guarantees immediate aborts on failure to protect production.

Deploy|Get started for free

The old way vs The SafeDeployer way

Stop relying on anxiety-inducing scripts and embrace deterministic deployments.

Without SafeDeployer

  • ✕Anxiety-inducing releases requiring manual verification and monitoring.
  • ✕Downtime during updates as old containers stop before new ones start.
  • ✕Complex custom bash scripts that are fragile and break easily.
  • ✕Broken production when a faulty container is accidentally deployed.

With SafeDeployer

  • ✓Deterministic deployments that just work seamlessly, every single time.
  • ✓True zero-downtime rollouts, completely transparent to your users.
  • ✓Drop-in integration with your existing Docker Compose configurations.
  • ✓Automatic aborts and rollbacks if pre-flight health checks fail.
SUPPORTED INFRASTRUCTURE
DockerDocker ComposeGitHub ActionsGitLab CINginxTraefikCaddy

Deploy with confidence in 3 steps.

A streamlined workflow designed for teams that need to ship fast without breaking things.

Step 1Configuration

Extend your compose file

Simply add an `x-safedeployer` block to your existing `docker-compose.yml` (or use a standalone `safedeployer.yml`) to define routing rules and health checks.

Step 2Execution

Execute the rollout

Run `sd-deploy up` in your CI/CD. The agent pulls your new container, runs it alongside the old one, and begins executing pre-flight health checks.

Step 3Monitoring

Traffic shift & Monitor

Once verified, the router instantly switches traffic to the new container. The dashboard provides live telemetry to ensure post-deployment stability.

End-to-end delivery pipeline.

From code merge to production monitoring, everything you need to ship software reliably.

Infrastructure as Code

The SafeDeployer CLI integrates seamlessly into your CI/CD pipelines. No more maintaining fragile custom shell scripts for routing.

$ sd-deploy up --config docker-compose.yml
✓ Parsed x-safedeployer configuration
✓ Provisioned Green environment (Port 39281)
↻ Running pre-flight health checks...
Stable (v1.0)90%
Canary (v1.1)10%

Canary Rollouts

Gradually shift traffic to new instances. Monitor latency and error rates before fully committing to the release.

Automated Aborts

If a new container fails to start or health checks degrade, the router instantly aborts the deployment and maintains traffic to the stable instance.

Health Check Failed (503)

Rolling back traffic to previous stable version. No user impact detected.

hooks:
before_deploy:
- "npm run db:migrate"
after_deploy:
- "npm run cache:sync"

Deployment Hooks

Run before_deploy and after_deploy scripts inside your container at any lifecycle stage — database migrations, cache warming, smoke tests. Available on all plans.

Deploy Now
Smart Traffic Shaping

Zero-downtime deployments, out of the box.

Traditional deployments require complex proxy reloads or accept minutes of dropped connections. SafeDeployer handles the complexity of blue/green state transitions natively.

  • Pre-flight Validation: New containers are fully booted and checked before any traffic is shifted.
  • Connection Draining: Old containers gracefully finish processing active requests before spinning down.
  • Native Integrations: Directly manipulates Nginx, Caddy, or Traefik configurations without external scripting.
api-server-v1.4.2
100% Traffic
api-server-v1.4.3 (New)
0% Traffic (Verifying)

Post-deployment observability.

Deployment isn't finished when the container starts. SafeDeployer continuously monitors system health, aggregating host system metrics (CPU, memory, disk I/O), Docker container stats (CPU, memory, network I/O), and application logs into a unified operations dashboard.

  • Host System Metrics: CPU, memory, and disk I/O for the entire server, not just the container.
  • Container Telemetry: Per-container CPU, memory, and network I/O streamed in real time.
  • Log Aggregation: Access stdout/stderr without SSH access.
  • Anomaly Detection: Set thresholds to trigger alerts or halt ongoing canary deployments.
CPU Usage12%
Memory480 MB
> tail -f /var/log/app.log
[2026-08-19 19:34:01] GET /api/v1/users 200 OK - 42ms
[2026-08-19 19:34:03] POST /api/v1/auth 201 Created - 110ms
[2026-08-19 19:34:05] GET /api/v1/metrics 200 OK - 15ms

Enterprise Compliance.

Engineered to meet strict regulatory constraints. SafeDeployer provides the controls required for mission-critical operations without compromising developer velocity.

Zero RCEStrict payload validation prevents arbitrary execution.
Air-Gapped ModeRun disconnected from external networks entirely.
Audit LoggingImmutable ledger of every configuration and state change.
RBACGranular permissions for environments and deployments.

Frequently Asked Questions

Everything you need to know about the product and architecture.

What permissions does the CLI require?

sd-deploy interfaces directly with your Docker daemon, requiring execution by a user in the docker group or as root. It also requires write access to your reverse proxy configuration files.

Does SafeDeployer have remote command execution (RCE) capabilities?

No. The agent listens only for strict, predefined telemetry events over WebSocket. It does not evaluate scripts, bash commands, or allow arbitrary remote execution.

Can cloud communication be disabled?

Yes. Telemetry is strictly opt-in. If the token is removed, the engine gracefully falls back to a 100% offline local orchestrator.

What reverse proxies are supported?

We natively support Nginx, Traefik, and Caddy. The orchestrator can integrate with these proxies whether they are running natively on the host OS or inside Docker containers.

Deploy with absolute confidence.

Join the engineering teams who have eliminated deployment anxiety and standardized their release process with SafeDeployer.

Deploy NowRead Documentation

No credit card required. Get started with our generous free tier.