L
Deep sleep is a somewhat nebulous chapter; for one thing, the ESP32 manual doesn't have the power management chapter yet. Similarly, sleep mode is not yet implemented in Pycom's ESP32 port of MicroPython. I believe, when deep sleep actually does work, it cuts power to the RAM so a wakeup is a boot with the deep sleep reset cause, and the pins that can cause it to wake may be only the RTC ones. Most of that's guessing and conjecture.
PS. Well, considering that was your thread, I'm not sure why it was expected to work? I vaguely recall somewhere else mentioning sleep will be in place for 1.0, which was planned for sometime in December, but there's a lot of work still to do.