Before you install
Nothing to prepare — no API keys or tokens. You’ll set up your admin account in the browser after install, and add credentials for individual integrations from inside n8n as you build workflows.n8n 2.x dropped HTTP basic-auth in favour of an interactive owner-account setup on first run. There is no generated password and no credentials file — the first person to open the URL creates the owner account, so set yours up immediately after install (see below).
Install
Dashboard → Apps → n8n → Install:| Field | Required | Notes |
|---|---|---|
| VM | yes | Pick a VM with free capacity, or create one. |
| Subdomain | no | Suggestion is n8n. The app is reached at https://<subdomain>.suji.fr. |
| Timezone | no | Defaults to UTC. Sets GENERIC_TIMEZONE — used for schedule/cron nodes. |
running in ~1 minute. You’ll get a public URL like https://n8n-<random>.suji.fr — that’s the n8n editor, served over HTTPS through the tunnel.
First connection — create your owner account
- Open the install URL (
https://n8n-<random>.suji.fr). - n8n shows a Set up owner account screen on first run.
- Enter your email, name, and a password. This becomes your admin login.
- You’re taken straight into the workflow editor.
Build your first workflow
Inside the editor:- Click Add first step and pick a trigger (Manual, Schedule, or Webhook).
- Add action nodes and connect them by dragging between the node dots.
- For nodes that talk to an external service (Slack, Google, an HTTP API…), n8n prompts you to add a credential — these are stored encrypted on the install’s volume.
- Click Execute workflow to test, then toggle Active (top right) to run it on its trigger.
Webhooks & external triggers
Webhook and “Wait for webhook” nodes generate a callback URL using the install’s public address, e.g.https://n8n-<random>.suji.fr/webhook/.... Give that URL to the external service.
- Test vs Production URLs — the editor shows a test URL that’s only live while the canvas is open, and a production URL that works whenever the workflow is Active. Use the production URL for real integrations.
- OAuth-based credentials (Google, etc.) use the same public host for their redirect URL — register that redirect in the provider’s console.
Day-to-day management
| Want to… | Where |
|---|---|
| Build / edit / run workflows | The n8n editor (https://n8n-<random>.suji.fr) |
| Manage stored credentials | Editor → Credentials |
| See past runs / debug failures | Editor → Executions |
| View container logs | Dashboard → Logs (pick n8n) |
| Open a shell inside the container | Dashboard → Terminal (pick n8n) |
| Browse the data volume on disk | Dashboard → Files (pick the n8n volume) |
| Restart the app | Install detail page → Restart |
| Upgrade to a newer version | Install detail page → Upgrade (when available) |
| Remove the install + its data | Install detail page → Uninstall |
Troubleshooting
I forgot the owner password
I forgot the owner password
Reset it from the dashboard’s per-app Terminal (pick n8n):This resets the owner account so you can set a new password on the next visit. (Exact subcommand can vary by n8n version — run
n8n --help to confirm.)The login page won't accept my credentials / cookie error
The login page won't accept my credentials / cookie error
A webhook URL shows the wrong host
A webhook URL shows the wrong host
n8n derives webhook URLs from the request host, which is your
*.suji.fr subdomain — so they should be correct out of the box. If you’ve put another proxy in front, or the URL still looks wrong, set an explicit WEBHOOK_URL via Edit Install → environment variables and redeploy.Bad gateway (Cloudflare 502)
Bad gateway (Cloudflare 502)
n8n isn’t reachable through the tunnel yet. Usually it’s still starting — wait ~30 s and retry, especially right after an upgrade. If it persists, the Logs tab shows why the container isn’t up.
Where things live
| What | Inside container | Named volume |
|---|---|---|
| SQLite database (workflows, executions) | /home/node/.n8n/database.sqlite | n8n-data |
| Credential encryption key | /home/node/.n8n/config | n8n-data |
| Custom nodes / binary data | /home/node/.n8n/nodes, /home/node/.n8n/storage | n8n-data |
Recommended size
- Small is plenty. n8n needs only ~512 MB to run and the compose caps it at 1 GB.
- Memory/CPU pressure comes from what your workflows do — heavy data transforms, large HTTP payloads, or many concurrent executions. If you run big batch workflows, size up; for typical automation a Small VM is comfortable.
- If you run other apps on the same VM, remember the VM’s total CPU/memory is shared.
Reporting issues
| Class | Where |
|---|---|
| n8n bug (node behavior, editor, workflow engine) | n8n-io/n8n issues |
| Marketplace packaging bug (compose / manifest / install form) | suji-hq/suji-templates issues |
| Suji platform bug (dashboard, billing, network) | Support ticket from the dashboard |