Self-hosted PaaS in a single binary.

Tech stack

• 
  Rust

  Language
• 
  Axum

  HTTP + WebSocket
• 
  Tokio

  Async runtime
• 
  Bollard

  Docker Engine API
• 
  SQLite

  Embedded database (WAL)
• 
  MiniJinja

  Server-side templates
• 
  HTMX + Alpine.js

  Progressive frontend
• 
  Tailwind CSS

  Styling
• 
  Traefik

  Reverse proxy + Let's Encrypt
• 
  AES-256-GCM

  Secrets at rest

Frequently asked questions

What is Pier?

Pier is an open-source, self-hosted Platform-as-a-Service written in Rust. It packages container management, Docker Compose, reverse proxy with automatic HTTPS, Git-to-deploy, alerts and notifications, multi-server agents, scheduled S3 backups and 35+ one-click service templates into a single ~15 MB binary that uses 20–40 MB of RAM idle.

How is Pier different from Coolify?

Coolify is a Laravel/PHP application that runs 6+ support containers and needs 750 MB–1.2 GB RAM before you deploy anything. Pier delivers the same core features from a single Rust binary that uses 20–40 MB of RAM, making a $5/month VPS viable as a serious deployment target.

What are the minimum server requirements?

Pier runs on any Linux server with Docker installed. Minimum: 512 MB RAM and 1 vCPU. Recommended for active deployments: 1 GB RAM and 1 vCPU. Pier ships as a single static musl binary for x86_64 and aarch64.

Is Pier free?

Yes. Pier is licensed under AGPL-3.0 — free to self-host and modify. If you offer a modified version as a network service, you must share your modifications under the same license. Commercial licenses without AGPL obligations are available on request.

Where does Pier store my data?

Pier uses an embedded SQLite database with WAL mode stored alongside the binary. Environment variables and registry credentials are encrypted at rest with AES-256-GCM, with the encryption key stored outside the database. Daily backups of the database and key are written to the data directory.

Does Pier support multi-server / fleet management?

Yes. Pier installs lightweight agents on remote servers via a single curl | sh command. The Core dashboard tracks heartbeats, system metrics (CPU, RAM, disk, Docker version) and server location, and distributes deployments and load-balanced replicas across the fleet from one UI.

How do alerts and notifications work?

Pier ships 10 preset alert rules (high CPU, RAM, disk, container crashed, deploy failed, backup failed, agent offline, SSL expiring soon, etc.) with configurable thresholds and cooldown windows. Notifications are delivered to Telegram, Discord, Slack and Email channels — multiple channels per rule, with per-rule routing.

Can I deploy from a private Docker registry?

Yes. Pier supports per-project and global registry credentials for any Docker-compatible registry (Docker Hub, GHCR, GitLab Container Registry, AWS ECR, private Harbor, etc.). Credentials are encrypted with AES-256-GCM and injected at pull time without ever being exposed to the UI.

Does Pier support load balancing and replicas?

Yes. You can scale any service from 1 to N replicas with three load-balancing strategies — round-robin, weighted, and sticky-cookie session affinity. Replicas distribute across multiple servers when agents are connected, and Traefik routes requests transparently with health checks.

How does Pier compare to Dokku, Kamal and CapRover?

Dokku and Kamal are CLI-first and assume one box per project; CapRover is a heavier Docker Swarm UI. Pier sits between them: a single Rust binary with a full web UI, multi-server agents, alerts, encrypted secrets, registry auth, scheduled backups and a 35+ template catalog — without bringing along a Postgres, Redis or PHP runtime as a hard dependency.