Home

wranglarr wiki

wranglarr is a Docker Compose stack for a Postgres-backed *arr suite plus a web app that backs up, versions and restores every app in it.

Page	What's there
Architecture	the containers, networks, volumes and how traffic flows
Quickstart	start here: a first install step by step, with screenshots
Login	wranglarr's sign-in: password, two-factor (TOTP), passkeys, API keys, recovery
App Logins	the apps' logins: one for every app, AuthTunnel (sign in once, to wranglarr), or your own
Security	what the Docker socket and the web UI expose, how wranglarr runs without root, a service user for each OS
TrueNAS SCALE	running the stack on TrueNAS SCALE 24.10+: datasets, permissions, sudo, the Apps screen
Setup	the first-run wizard, init.sh, opting services in
Bringing in existing apps	moving a Radarr, Sonarr, Prowlarr, Bazarr or Seerr you already run into the stack, keeping its library and history: config, SQLite → Postgres, paths, connections
Environment Variables	every .env key, with defaults and which service reads it
Service: Postgres	the database server and the databases it creates
Service: Radarr, Sonarr, Lidarr	the config.xml-based arrs on Postgres
Service: Prowlarr	indexer manager (opt-in), registered in every arr automatically
Service: Bazarr	subtitles, Postgres via environment
Service: qBittorrent	torrent client
Service: SABnzbd	usenet client
Service: NZBGet	usenet client
Service: Seerr	requests (opt-in): your Plex / Jellyfin / Emby users ask, Radarr and Sonarr get it
Service: wranglarr	the backup/restore app: environment, wranglarr.toml, HTTP API
VPN: Shared Tunnel	one gluetun WireGuard/OpenVPN tunnel: every app, download clients and Prowlarr, or download clients only
VPN: Per App	hotio images, one tunnel per app
Remote Access	Tailscale, Cloudflare Tunnel, Cloudflare DNS: reaching the stack from outside, with or without the VPN
Reverse Proxy	config for Caddy, nginx, Traefik, caddy-docker-proxy, Cloudflare Tunnel, Nginx Proxy Manager; public, LAN only or not per app
Dashboard	posters for the media arrs with what's coming up and missing, cards for the rest with their queues and lists; where each app opens (IP and port, or a domain); each app's login
Notifications	Telegram, Discord, Slack, ntfy, Gotify, Pushover, Apprise, email, webhook: failed backups, apps offline, stuck downloads, blocked imports…
Connections	Plex, Jellyfin, Emby and Kodi into the arrs (library rescans); Seerr
Container Updates	registry checks, backup-first image updates with health checks, rollback, holds, auto-update
Release Notes	what each release adds, changes and fixes
Backups and Restore	what a version contains, retention, restore semantics, restore plans, running a version without wranglarr
Live status and API keys	the arrs tell wranglarr when something changes, so the pages update straight away; every app's API key in one place; a key changed in an app reconnects the apps using it
App Connections	download clients into every arr, Sonarr/Radarr into Bazarr, the arrs into Prowlarr; categories and folders
Indexers	the Indexers page: Torznab/Newznab feeds, Jackett, NZBHydra2 or an outside Prowlarr, pushed to the arrs you pick
Community configs	custom formats, quality profiles, sizes and release profiles for one job (x265 first at the smallest size, anime English dub…) as JSON: import with a preview and undo, test a release name, export your own
Shared Settings	define download clients / notifications once
HTTP API	every endpoint for scripts, Home Assistant and cron: API keys, jobs and runs, errors, bodies and answers
Troubleshooting	the gotchas, each with its fix
Development	how the code is organized; building, testing, adding a service, releases

Conventions used throughout: appdata/ is ${APPDATA} (default ./appdata); "the arrs" means Radarr, Sonarr and Lidarr (Prowlarr is an indexer manager: it is backed up like them but downloads nothing); a version is one backup set.