Security
Protect your proxies from the panel: geo-restriction, rate and connection limiting, bot filtering and security headers.
Security
Each proxy has protections that you configure yourself from the panel, in the proxy settings. They apply to the traffic of that proxy without you having to touch the server.
In addition, the platform automatically blocks malicious IPs and scanners (probing for .env, .git, wp-login.php files, shells...) for all proxies. This protection is always active and requires no configuration on your part.
Geo-restriction
Geo-restriction allows or denies access based on the visitor's country of origin (determined by IP geolocation). This is useful when your service is intended only for certain countries, or when you notice abnormal traffic coming from regions you do not serve.
In the proxy settings, enable geo-restriction and choose the mode:
- Whitelist -- only the listed countries are allowed
- Blacklist -- the listed countries are blocked
Countries are designated by their two-letter ISO code (for example FR, BE, CH).
Rate limiting
Rate limiting caps the number of requests per second and per IP, with a burst allowance. It protects your backend against traffic spikes and brute-force attempts.
| Setting | Purpose |
|---|---|
| Rate limiting enabled | Enables limiting on this proxy |
| Requests per second | Rate allowed per IP |
| Burst | Number of requests tolerated beyond the rate |
Connection limiting
Connection limiting caps the number of simultaneous connections per IP. It is an effective protection against certain denial-of-service attacks.
Browser check
A proxy declared as a website (and not as an API) can check that its visitors use a real browser. The visitor's browser solves a short computation once, with no click and no captcha, in a second or two, then is not checked again for 24 hours. No outside service is called, and the cookie it sets is strictly necessary: no consent banner.
In the proxy settings, set the browser check:
- Off -- no visitor is checked (default)
- Suspicious visitors only -- requests that look like no browser: no user agent, a tool or a headless browser, or no language
- Every visitor -- each visitor goes through the check on their first visit
Search engine robots (Google, Bing) are never checked: your site stays indexed. Requests stopped by the check are not counted in the volume of your plan.
The automatic attack mode (on by default) checks every visitor for a few hours when the platform detects a traffic spike on the proxy, then goes back to your setting on its own. You receive an email when it starts.
An API proxy is never checked: an application, a connector or a synchronisation cannot solve the check. Visitors with JavaScript disabled cannot pass it.
Bot filtering
Bot filtering denies access to unwanted bots based on their signature. In the proxy settings:
- Enable filtering
- Select the bot categories to block
Signatures are maintained by the platform and sorted by category.
Security headers
You can have HTTP security headers added to the responses of your proxy: Strict-Transport-Security (HSTS), X-Frame-Options, X-Content-Type-Options, X-XSS-Protection and Referrer-Policy. They strengthen the protection of your visitors' browsers.
Automatic IP blocking
In addition to the permanent protection against scanners, you can enable automatic blocking of IPs that generate too many errors on your proxy. The settings:
| Setting | Purpose | Default value |
|---|---|---|
| Automatic blocking enabled | Enables blocking on this proxy | Disabled |
| Error threshold | Number of errors that triggers the block | 50 |
| Time window | Analysis period, in minutes | 60 |
| Block duration | Duration of the block, in minutes | 1440 (24 h) |
A blocked IP receives an error and can no longer reach your backend for the chosen duration; the block is lifted automatically when it expires.
Maintenance mode
Maintenance mode displays an unavailability page instead of your service, for the duration of an operation. It is enabled and disabled from the proxy settings.
Monitoring and alerts
The platform monitors the traffic of your proxies and alerts you by email in case of an anomaly (traffic spike, error spike, unusual geographic origin) or if the backend is unavailable. These alerts are informational and help you react. You choose which alerts you want to receive from your notification preferences.