Analytics cookies only (Google Analytics, PostHog). No ad networks, no cross-site tracking. See our privacy policy.
Self-host Spoolman
updated Nov 2026deployed by usprices checked · Jul 2026Install verified on Ubuntu 26.04 · Nov 2026 · how we test
We earn commissions when you shop through the links below. Full disclosure →
A web service for tracking 3D printer filament spools. It records manufacturers, filaments and individual spools with their remaining weight, colour, material, price, location and lot number, and has a label designer that prints QR codes for spools. Moonraker (Klipper) and OctoPrint integrations subtract used filament as prints run, a Home Assistant integration exposes spools as sensors, and there is a REST API and an optional Prometheus metrics endpoint. A community database of filaments can be imported. Data is kept in SQLite by default, with PostgreSQL, MySQL and CockroachDB as options. It has no authentication of its own.
Pick Spoolman when you run 3D printers and want to know what filament is left on each spool without weighing everything by hand. You add manufacturers, filaments and spools, print QR labels from the built-in label designer, and let Moonraker or OctoPrint subtract the filament each print uses. It is MIT-licensed, a Python and FastAPI service with a web client, and keeps its data in a SQLite file by default. Our snippet adapts the wiki's Docker install, pins `ghcr.io/donkie/spoolman:0.27.1` (released 2026-10-31) instead of `latest`, and stores the database in a named volume, which avoids the `chown 1000:1000` the wiki asks for on a bind mount. Look elsewhere if you want a general stock or parts inventory: Spoolman only models filament spools, and a general inventory tool fits better for screws, electronics or tools.
What you need
Any VPS with at least 256 MB of RAM
A domain you control — most self-hosted setups need HTTPS in front of them
Paid link — we earn a commission if you shop through it.
Install
Run these commands on your server:
# Spoolman — 3D printer filament inventory — adapted from the wiki's Docker install, pinned to ghcr.io/donkie/spoolman:0.27.1 with the SQLite database in a named volume (the wiki's bind mount needs a chown to uid 1000)cat > docker-compose.yml <<'EOF'services: spoolman: image: ghcr.io/donkie/spoolman:0.27.1 restart: unless-stopped ports: - "7912:8000" environment: TZ: UTC volumes: - spoolman-data:/home/app/.local/share/spoolmanvolumes: spoolman-data:EOFdocker compose up -dtimeout 180 bash -c 'until curl -fsSL -o /dev/null http://localhost:7912/api/v1/health; do sleep 3; done'# → http://SERVER_IP:7912 — the Spoolman web client with an empty spool library; there is no login, so keep the port on a trusted network or behind an authenticating reverse proxy
Spoolman's README and wiki flag these before you rely on it:
non-negotiableSpoolman has no authentication. The wiki's Security page says anyone who can reach it can read and change the whole inventory, and, if it is reachable from outside your network, tells you to put it behind a reverse proxy that handles authentication rather than forwarding a port to it. Keep port 7912 on a trusted network.
non-negotiableOut of the box Spoolman refuses writes and WebSocket connections that come from another website's origin. If a browser-based client such as Fluidd or Mainsail on another host needs to talk to it, you must add that origin through SPOOLMAN_CORS_ORIGIN; Moonraker, OctoPrint and Home Assistant need no setting.
non-negotiableThe data directory must be mounted, or the SQLite database is lost when the container is removed. The container logs a loud startup warning when it is not, and the wiki says your data will be lost every time you update. Our snippet mounts the spoolman-data volume at /home/app/.local/share/spoolman.
non-negotiableNightly SQLite backups are on by default and keep the last five, but the default backup directory is inside the data directory, so a lost volume takes the backups with it. Set SPOOLMAN_DIR_BACKUPS to another location or copy the backups off the machine.
non-negotiableThe 0.27.1 release notes say a tare of 0 is now kept as 0 instead of falling back to the filament's or manufacturer's empty spool weight, so check the remaining weight of any spool you entered with tare 0 after updating. The container starts as root and drops to uid 1000 through PUID/PGID; PUID=0 is refused.
non-negotiableOn start and hourly Spoolman syncs filament data from the external SpoolmanDB (EXTERNAL_DB_URL, default https://donkie.github.io/SpoolmanDB/), and on start and every 12 hours it checks GitHub for a newer release (SPOOLMAN_UPDATE_CHECK). The server therefore makes outbound requests; set EXTERNAL_DB_URL to an empty string and SPOOLMAN_UPDATE_CHECK=FALSE to stop them.
Spoolman used 105 MB of RAM at idle on GCP e2-standard-2 · Ubuntu 26.04 LTS · Docker 29.9.3 (first boot, measured Nov 2026). We list a minimum of 256 MB.
Is the Spoolman install snippet verified?
The Spoolman install snippet is marked verified on Ubuntu 26.04 (Nov 2026).