I have a new version of a board (that worked fine) that won't boot from power up unless I short nRST physically. Then, about 5 seconds later, it will boot. Using the debugger, everything seems to work OK.
I have also noticed that there is about 2.5V on the NMI pin, though that is not coming from the circuit (I have removed a resistor so that it's just the pin that is driving that). I mention it because NMI can be involved in the boot sequence.
Once I short nRST and it boots, then the circuit works as expected. It goes to sleep in mode VLLS3 and wakes up as expected. But it won't boot on the initial power up.