3

I have a SaaS web application. One of my clients needs the app to be installed on his infrastructure (meaning physically on servers only they can access). The application stack is 2 docker images: one for the web app written in Python and the other for the database service.

  • How do I prevent the client from accessing the source code of my application? They can access the docker image and easily get the code which is not even compiled, the language being Python.
  • What are the choices of deploying the app within the client's server and having my code inaccessible or hardly accessible?
4

1 Answer 1

2

So doing this would be very difficult and unlikely to be a realistic prospect. However it's theoretically possible to make it difficult for someone to access the code, if you can control what hardware it's run on.

In theory, something like confidential containers can protect an application container's content from the host running the content.

It's aimed at cloud providers running containers, and not being able to see inside the container, but it could be done with any system that supported the specific features.

2
  • Is there currently any hardware out there that fulfils the "protected" criteria, let alone the "servers only they can access" criteria? Last time I checked both Intel and AMD were telling their marketing departments to claim such things could work, but real-world attempts all turned out to be regularly failing at their one job, impractical and/or unfinished.
    – anx
    Commented Jul 28, 2023 at 14:26
  • 1
    I've seen the claims but not practical demonstrations of them, thus "in theory" :) I'll be interested to see if it makes it into practical on-prem hardware. Commented Jul 28, 2023 at 17:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .