Automation recipes: gluing the platform together

You've met the pieces one at a time — Schedules, Git deploy, webhooks, templates. This is where they stop being separate features and start being a server that looks after itself. Every recipe below is built only from things the panel actually does; each names the plan it needs. Copy the ones that fit, and hands-off becomes the default.

The building blocks

Two things to hold in mind throughout: schedule tasks run top to bottom in the order you list them, and the fastest way to test any schedule is its Run now button — fire it by hand, watch the console, then check the schedule's Logs.

Recipe 1 — Nightly backup, then restart

A fresh restart clears memory creep; a nightly backup means you always have yesterday to fall back to. Do both in one schedule.

• Trigger: Daily, e.g. 4:00 AM (times use the server owner's timezone).
• Tasks, in order:
  1. (optional) Command — say Nightly maintenance in 60s to warn players.
  2. Backup — name it "nightly".
  3. Power action → Restart.

Because tasks run in order, the backup finishes before the restart begins. Plan note: the Backup and Power-action tasks are premium; on free, the Command task runs while the server is online.

Recipe 2 — Deploy on push, with a build

Push to your repo and let the server pull, build, and restart itself.

1. On the Git page, connect the repo via GitHub / GitLab / Bitbucket (OAuth) and enable auto-deploy — a push now triggers a deploy via webhook.
2. Add post-deploy commands for the build — npm run build, tsc, or go build -o app — and set the post-deploy action to Restart.

Now: push → Falix pulls the branch into /home/container → your build runs → the server restarts on the new code. Nothing to click. See Auto-deploy and Build steps.

🎯 Good to know: Linked a public repo by URL instead of connecting an account? Those don't get push webhooks — but they support scheduled pulls every 1 / 6 / 12 / 24 hours, plus manual deploys. Same hands-off result, on a timer instead of on push.

Recipe 3 — A Discord ping when a backup completes

You don't want to watch the Logs to know a backup ran — you want to be told, especially when one fails.

1. Set up the nightly backup schedule from Recipe 1.
2. Attach a webhook to that schedule: paste a Discord webhook URL (or any HTTPS endpoint) and choose which events notify you — schedule started, schedule completed, task completed, or task failed.
3. Press Test to confirm the message lands in the right channel.

Now a completed backup pings your channel, and — more importantly — a failed one does too, so a quietly broken backup can't hide from you.

💡 Tip: The webhook URL must be HTTPS. Of the four events, task failed is the one most people actually want — it's how you hear about the backup or restart that broke without babysitting the panel. Separately, Settings → General has a Discord webhook for crash notifications; the two work well together.

Recipe 4 — Auto-restart on crash (premium)

Keep the server up even when it falls over at 3 AM.

• Trigger: Event → server crashed.
• Task: Power action → Start.

A crash is now followed immediately by a boot instead of hours of downtime. This one needs a power task and a just-stopped server, so it's premium in practice; premium also has dedicated crash-detection settings on the Settings page. Pair it with Recipe 1 and the server mostly runs itself. See Keeping apps online.

Recipe 5 — Scheduled housekeeping

Servers accumulate cruft — temp files, old logs, stale data. Sweep it on a timer.

• Trigger: Daily or an Interval preset (5/15/30 min, 1/2/6/12 h).
• Task: a Command that runs your app's cleanup, or a Delete files task pointed at a temp folder.

Keep the paths specific — Delete files does exactly what it says, with no undo — and take a backup before trusting a destructive schedule.

The recipe table

Templates in the mix

A one-click template writes starter files onto a server (overwriting only same-named files) — a great starting point that pairs naturally with these recipes: deploy a template to scaffold the project, then wire up Git auto-deploy (Recipe 2) for everything after. See One-click templates.

Troubleshooting

• A timed schedule fired at the wrong hour — clock times follow the server owner's timezone (Settings → Environment).
• The schedule never ran — on free it only runs while the server is online, and only the Command task is available. Check both.
• The webhook never posts — the URL must be HTTPS; test it with the Test button.
• Git sync did nothing — the repo has to be linked on the Git page first. See Deploy failed.

Next steps

• Automate your server with Schedules
• Auto-deploy on every push
• A /health route and schedules that watch it