efecd26e31
The server-detail page gains a "Firewall-Regeln" panel — addressing "ich kann nur aktivieren, sonst nichts": list rules, add, and delete, with hard SSH-lockout guards. FirewallService: - status(): one privileged read — installed/active, default policies, and the rule list. ufw rules come from `ufw show added` (add-syntax, works active OR inactive), defaults from /etc/default/ufw; a CLUSEV_FWREAD_OK sentinel + a "Status:" check distinguish a real read from a failed/empty one. - addRule(): validated action/proto/port/from (proto before `from`, ufw grammar); refuses deny/reject on the SSH port or portless (would block all inbound). - deleteRule(): whitelists the spec against rules ufw actually reported (injection- proof — a forged spec can't match), deletes by spec (race-free, no rule number), guards SSH-port allow rules incl. ranges and trailing comments, distinguishes not_found from a real deletion so the audit only records true deletions. firewalld is READ-ONLY this release: status() reads the runtime state of every active zone (ports/services/rich rules, zone-attributed), rule mutation is refused with a German note. on/off + hardening (Phase A) still work. UI: x-panel with default-policy badges, "+ Regel" (modal modals.firewall-rule), per- rule delete via R5 ConfirmAction (audit deferred to the real outcome), graceful read-error / not-installed / inactive / firewalld-read-only states. Fix: the firewall row's tone variable was renamed $tone -> $ruleTone — the page-level $tone is a closure used by the gauges/Volumes panel; reusing the name clobbered it and 500'd the page below the firewall block. R12 now confirms the full loaded page. Default-policy EDITING was intentionally not exposed (highest lockout risk). Pint clean; Codex review clean; R12 — server-detail HTTP 200, 0 console errors, firewall panel + rules + volumes all render. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| app | ||
| bootstrap | ||
| config | ||
| database | ||
| docker | ||
| docs | ||
| public | ||
| resources | ||
| routes | ||
| storage | ||
| tests | ||
| .dockerignore | ||
| .editorconfig | ||
| .env.example | ||
| .gitattributes | ||
| .gitignore | ||
| CHANGELOG.md | ||
| CLAUDE.md | ||
| Dockerfile | ||
| README.md | ||
| artisan | ||
| composer.json | ||
| composer.lock | ||
| docker-compose.prod.yml | ||
| docker-compose.yml | ||
| handoff.md | ||
| install.sh | ||
| kickoff-prompt.md | ||
| package-lock.json | ||
| package.json | ||
| phpunit.xml | ||
| rules.md | ||
| vite.config.js | ||
README.md
Clusev
Self-hosted control panel to administer a fleet of Linux servers from one dashboard, agentless over SSH. Clusev is the control-plane (UI + orchestration); it talks to real servers with phpseclib (exec + SFTP) — it never reimplements daemons. Security-first: 2FA, encrypted SSH-credential vault, and a full audit log. Multi-server is free and never paywalled.
Status: v1 — dashboard/live metrics, systemd services, SFTP file manager, server details, auth + 2FA. UI copy is German; meta-docs are English.
Stack: Laravel 13 · Livewire 3 · Tailwind v4 · Reverb (realtime) · Redis · MariaDB · phpseclib3.
Everything runs in Docker. See CLAUDE.md for architecture and rules.md for the hard conventions.
Development
The host needs only Docker (no PHP/Composer/Node). The dev app container runs php-fpm + nginx +
Vite via supervisor; reverb, queue, mariadb, redis are their own services.
cp .env.example .env # then set DB_PASSWORD / DB_ROOT_PASSWORD
docker compose up -d --build # app on :80, Vite HMR on :5173
docker compose run --rm --no-deps -u "${HOST_UID:-1002}:${HOST_GID:-1002}" app php artisan key:generate
docker compose exec app php artisan migrate --seed
Run any tooling inside the container, e.g.
docker compose exec app php artisan make:livewire Servers/Show (class-based — never Volt).
Production install
One host, only Docker + a sudo user. Caddy is the single host-exposed service: auto-TLS when a
domain is set, plain HTTP on the bare IP otherwise. install.sh is idempotent — it generates
secrets once (never regenerates), brings up the prod stack, migrates, and creates the first admin
with a one-time random password printed only on the terminal.
git clone https://git.bave.dev/boban/clusev.git && cd clusev
./install.sh
Prompts (non-interactive: set CLUSEV_DOMAIN / CLUSEV_ADMIN_EMAIL in the environment):
Domain (empty = access by IP over HTTP): clusev.example.com
Admin e-mail (login + Let's Encrypt): admin@example.com
The closing banner shows the URL + the one-time admin password. On first login Clusev forces a password change and 2FA enrolment before the panel unlocks.
One knob: APP_DOMAIN
APP_DOMAIN |
proxy | URL | Reverb |
|---|---|---|---|
| (empty) | Caddy serves plain HTTP on APP_PORT |
http://<ip> |
ws://<ip>/app/* |
clusev.example.com |
Caddy gets a Let's Encrypt cert, forces HTTPS | https://… |
wss://…/app/* |
SITE_ADDRESS, APP_URL, REVERB_* are derived from it by the installer — nothing hardcoded.
Bare-IP mode serves 2FA/audit over cleartext HTTP; the installer warns loudly. Let's Encrypt needs
publicly reachable 80/443 — a private (RFC1918) target needs a DNS-01 Caddy build instead.
Manual deploy (for operators who don't curl | bash)
docker compose -f docker-compose.prod.yml build
docker compose -f docker-compose.prod.yml up -d
docker compose -f docker-compose.prod.yml exec -u app app php artisan migrate --force
docker compose -f docker-compose.prod.yml exec -u app app php artisan clusev:install --email=admin@example.com
In-dashboard updates (signed intent file + host-side updater, digest-pinned, cosign-verified, 2FA-gated, with backup + rollback) are designed in
docs/install-update-design.mdand land in v1.x — they are intentionally not shipped yet.
License
AGPL core + commercial Pro modules (open-core). Multi-server fleet management is always free.