This service has been discontinued
The Public No Track Service was an experiment where a ready to go No Track service was made available from any internet connection.
No Track is a "DNS server which blocks Tracking websites from creating cookies or sending tracking pixels." [1] The Public No Track Service provided a ready to go No Track server available from any internet connection. This allowed you to:
- Use No Track without any having to set up your own No Track server on every network you need it on.
- Use No Track on networks which you don't control.
- Use No Track without any fuss.
The service has been discontinued for two reasons:
- Using No Track in this way is not secure as discussed in No Track Issue #210.
- The VPS provider that was being used has politely asked not to run an open, recursive, DNS resolver on their system.
As mentioned above, using No Track in the way that has been presented here is insecure. A discussion about this can be found in No Track Issue #210. Essentially, when requests are blocked and re-routed to the No Track server instead, the No Track server is actually pretending to be the original server. For good reasons, HTTPS is designed such that no server can pretend to be another server. This keeps you safe from entering in your bank credentials into a server pretending to be your bank. To get around this, when No Track returns alternative content for a blocked request, it must do so without a valid HTTPS connection. This leaves the user susceptible to "man in the middle" attacks where an attacker which controls some part of the network between the No Track server and the user can manipulate these blocked requests in any way they want without detection.
To mitigate against this vulnerability, No Track users need to be able to trust the route between themselves and the No Track server. This means that No Track is perfectly safe when used as it was intended, in your own local network. It could also be made safe by creating and using a virtual private network (VPN) where both the user and No Track server could safely communicate over an artificially secure network.
The service was originally hosted on a budget VPS (1 CPU, 512MB RAM, 10GB storage, 200GB transfer). This was more than enough to handle the small number of requests that it received very quickly.
2017-03-03 - The service has been discontinued due to the reasons listed in the README.
2017-02-06 - The Public No Track Service GitHub repo was created and Brisbane server announced to the public. Let's see how it goes!
Any feedback is welcome in the issues list. Private feedback is welcome via email at [email protected].
The configuration used to implement the service is documented in server-config.md. Any review or suggestion about the configuration is welcome. Discussion is welcome in the issues list.