The first one is talking about the quantum realm. So, absent of measurement or some kind of collapsing of the wave function, pure quantum states are 100% reversible. So the question is, where does the transition happen from the quantum world to the world we know. Whence Einstein's question about "when I stop looking at the moon does it disappear". So far, nobody knows. So far, we haven't observed any limits to how big a pure quantum system can get, and there's no "spontaneous" collapse if we don't measure something.
So the question is, could a pure quantum state model a meteor crashing into Earth? It doesn't seem like it. Friction, heat, and so on cannot be reversed classically. The question of Schrodinger's Cat even becomes moot because in an isolated system, the processes a body requires to sustain life (such as friction and heat) can't be modeled in the first place, so the cat will turn into a sloppy goop before anyone pulls the trigger. So, what's going on? Why does quantum mechanics seem to model everything that happens in the microscopic world, but in the macroscopic world we see things that it can't? Where does that crossover between microscopic and macroscopic occur? If measurement is what causes quantum wave collapse, then would none of this happen if nobody was there to measure it? And what created the measurers? 99 years since Schrodinger's equation, and we're not particularly any closer.
Black holes also seem to violate reversibility. They absorb information, but the radiation they give off is random. So, it's impossible to perform the process in reverse. This is spooky in a different way than the above because it's purely mathematical. For "regular" micro/macro quantum/classical paradoxes, we're generally talking about the classical realm of experiments. But for black holes, it's pure math, of our two fundamental theories of nature, and showing that they don't line up. Though, it's perhaps less spooky because the obvious answer is that one or both of the theories is slightly wrong, so we just need to fix it. The measurement problem, or understanding waveform collapse, I think is the more interesting problem, but that's just me.