OpenADP is Open source Advanced Data Protection for everyone. It defends vs secret mass surveillance via software transparency and distributed trust. That distributed trust is built through OpenADP servers run by volunteers around the world.
OpenADP has servers in the US, but for resistance to any one government, volunteers in several countries are needed. There is a quick-start guide for running OpenADP on a Raspberry PI. If you've got a Raspberry PI to spare, and some free time, consider volunteering for OpenADP.
That said, I'm a bit confused about the use case. Wouldn't it be simpler and more secure to have all the encryption occur on the client side, and have the server be a dumb encrypted blob store?
Put another way, I think OpenADP tries to solve the problem "I don't trust hosting providers in any single sovereign nation" by splitting the trust between multiple nations; whereas it seems like it would be even better not to trust any of them.
First off, sometimes it shows the wrong number of attempts you still have (sometimes it shows 10 when it really is 5, sometimes it shows 0 when it really is 5 or 4 or just simply >0).
Secondly, the PIN provided does not always work, I used 1234 and decryption seemed unsuccessful. I set it up again and it worked.