How uniq is Python's id()?

tl;dr Does Python reuse ids? How likely it is that two objects with non overlapping lifetime will get the same id?

Background: I've been working on a complex project, written purely in Python 3. I've been seeing some issues in testing and spent a lot of time searching for a root cause. After some analysis, my suspicion was that when the testing is being run as a whole (it's orchestrated and being run by a dedicated dispatcher) it's reusing some mocked methods instead of instatiating new objects with their original methods. To check if the interpreter is reusing I used id().

Problem: id() usually works and shows the object identifier and lets me tell when my call is creating a new instance and not reusing. But what happens when ids if two objects are the same? Docs say:

id(object) Return the “identity” of an object. This is an integer which is guaranteed to be unique and constant for this object during its lifetime. Two objects with non-overlapping lifetimes may have the same id() value.

The questions:

1. When can the interpreter reuse id() value? Is it just when it randomly selects the same memory area? If it's just random, it seems extremely unlikely but it's still not guaranteed.

2. Is there any other method to check what object I am actually referencing? I encountered a situation where I had the object, it had a mocked method. The object was no longer used, garbage collector destroyed it. After that I create a new object of the same class, it got a new id() but the method got the same id as when it was mocked and it actually was just a mock.

3. Is there a way to force Python to destroy the given object instance? From the reading I did it appears that no and that it is up to a garbage collector when it sees no references no the object but I thought it's worth asking anyway.

from How uniq is Python's id()?