0

Let's consider such an example:

I'm an author of application like Google Chrome. I would like to my application sends anonymous telemetry logs to my server. How do I prevent somebody to use my server API to pretend he is my application to insert some malicious / spammy logs?

User of that application might be anonymous. Would it be valuable to set some quota to my API that would prevent too many logs from the same source in some span of a time?

How would you approach that? How big-tech companies do that? How google do that with their Chrome application? Microsoft with Windows?

Also, would it be possible to access some API only by signed application?

1
  • "How do I prevent somebody to use my server API to pretend he is my application to insert some malicious / spammy logs?" Given a sufficiently clever and resourceful adversary, you can't. You need a more detailed threat model as to precisely which attacks you're trying to prevent. – Philip Kendall Mar 22 at 9:59
1

You can't stop them sending data, but you can identify anomalous data and remove it.

Say for example for every download of chrome you insert a unique id into the code and save that same id to your database.

Now when data is sent it can include that id. If the user hacks the code to send bad data they can either include the id, in which case you can see that one user is sending odd data and exclude it. Or replace with another id, in which case it will not match anything on your database and you can exclude it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.