website/.gitea/workflows/demo-wipe.yml

# The public demo's wipe (docs/website/DEMO_MODE.md §7). Every 2 hours on the
# hour, UTC (DM7), it puts demo.runicgateway.com and its Rust server back to the
# demo's starting point:
#
#   1. versions  the latest Rust bundle and the newest Module-Rust release on the
#                same protocol (§7.5, DM12)
#   2. the rig   FIRST: kill, write the new bundle if it changed, restore
#                demo-baseline.tar.gz, start (§7.2). A fresh site imports what is
#                on the rig, so the rig must be back before the site talks to it (§7.1)
#   3. the site  down -v, pull, up, seed:demo (§7.3)
#   4. smoke     the site answers, then the rig reports in (bounded)
#
# A redeploy is the same path: a push to main that changes deploy/demo/ runs it,
# and so does workflow_dispatch (§7.4).
#
# Runs on the `demo-site` runner, in host mode, on the demo host itself. The host
# needs Docker with Compose, and node for actions/checkout; everything else runs
# in containers. One-time host setup: deploy/demo/README.md.
#
# Secret:    PTERODACTYL_CLIENT_KEY  the demo subuser's client key (file + power)
# Variables: PTERODACTYL_URL, PTERODACTYL_SERVER (defaults below), DEMO_DIR

name: Demo wipe

on:
  schedule:
    - cron: '0 */2 * * *'
  workflow_dispatch: {}
  push:
    branches: [main]
    paths:
      - 'deploy/demo/**'
      - '.gitea/workflows/demo-wipe.yml'

# A wipe and a baseline capture must never overlap: both stop the rig.
concurrency:
  group: demo-rig
  cancel-in-progress: false

env:
  DEMO_DIR: ${{ vars.DEMO_DIR || '/opt/demo.runicgateway.com' }}
  PTERODACTYL_URL: ${{ vars.PTERODACTYL_URL || 'http://192.168.0.12' }}
  PTERODACTYL_SERVER: ${{ vars.PTERODACTYL_SERVER || '6eb9d621' }}
  NODE_IMAGE: node:22.23.4-alpine

jobs:
  wipe:
    runs-on: demo-site
    if: github.ref == 'refs/heads/main'
    steps:
      - name: Check out the deploy files
        uses: actions/checkout@v4

      - name: Install the deploy files
        # Everything but .env, which only the host holds, and state/, which the
        # wipe itself writes.
        run: |
          set -euo pipefail
          test -f "$DEMO_DIR/.env" || { echo "::error::$DEMO_DIR/.env is missing; see deploy/demo/README.md"; exit 1; }
          mkdir -p "$DEMO_DIR/state" "$DEMO_DIR/logs" "$DEMO_DIR/modules"
          # The app runs as uid 1000 and unpacks the declared module into ./modules.
          # The runner is not root, so a host chown fails; a container is root and
          # the runner may start one. A failure here stops the run: without it the
          # app comes up healthy with no module and the seed fails much later.
          docker run --rm -v "$DEMO_DIR:/demo" "$NODE_IMAGE" chown -R 1000:1000 /demo/modules /demo/logs
          cp deploy/demo/docker-compose.yml deploy/demo/seed.json deploy/demo/rig.mjs "$DEMO_DIR/"

      - name: Decide the versions
        run: |
          set -euo pipefail
          docker run --rm -v "$DEMO_DIR:/demo" -e STATE_DIR=/demo/state "$NODE_IMAGE" node /demo/rig.mjs plan

      - name: Restore the rig
        id: rig
        # The site is wiped even when this fails: a demo left un-wiped is worse
        # than one whose rig kept a visitor's changes for a cycle. The last step
        # fails the run so the failure is seen.
        continue-on-error: true
        env:
          PTERODACTYL_CLIENT_KEY: ${{ secrets.PTERODACTYL_CLIENT_KEY }}
        run: |
          set -euo pipefail
          docker run --rm -v "$DEMO_DIR:/demo" -e STATE_DIR=/demo/state \
            -e PTERODACTYL_URL -e PTERODACTYL_SERVER -e PTERODACTYL_CLIENT_KEY \
            "$NODE_IMAGE" node /demo/rig.mjs restore

      - name: Wipe and reseed the site
        run: |
          set -euo pipefail
          cd "$DEMO_DIR"
          docker compose pull
          docker compose down -v --remove-orphans
          docker compose up -d
          for attempt in $(seq 1 60); do
            STATUS="$(docker inspect -f '{{.State.Health.Status}}' rg-demo-app-1 2>/dev/null || echo starting)"
            [ "$STATUS" = healthy ] && break
            sleep 5
          done
          if [ "$STATUS" != healthy ]; then
            echo "::error::the app did not become healthy within 300s"
            docker compose logs --tail 100 app
            exit 1
          fi
          docker compose exec -T app npm run --silent --prefix server seed:demo -- /app/deploy/demo/seed.json
          docker compose ps

      - name: Smoke check
        run: |
          set -euo pipefail
          docker run --rm --network rg-demo_default -v "$DEMO_DIR:/demo" \
            -e SITE_URL=http://app:3000 "$NODE_IMAGE" node /demo/rig.mjs smoke

      - name: Report the rig restore
        if: steps.rig.outcome == 'failure'
        run: |
          echo "::error::the rig restore failed (see its step); the site was wiped and reseeded anyway"
          exit 1