“Stack overflow is a programming error in which an attempt to write data to a particular block of memory fails because there is no space left in the block. This type of error happens in the same way as buffer overflows, heap overflows, and stack buffer overflows. The difference between these types of errors depends on the computer data structure being used, and the common feature is that an attempt occurs to write more data than there is available space to hold it.”
The point is, stack overflow is a programming error, and error made by the programmer, and occurs because the software attempted to do exactly what it was told to do. The failure was a human failure, by the programmer.
Now another matter: Hardware errors do occur. Not as often as some programmers want to think, but is does occur. At particular risk are those devices made at a very small integration scale. Temperature extremes can be a source of error in program execution. There can be places in memory where the instructions are not read into the instruction decoder/sequencer as originally established by the program loader. Hardware failures expresses itself as software failures.
Oops, I forgot. Overclocking can cause failures that are mistakes in program execution even when the software is without error.
Dang, I also got off track. This sort of failure might not be a software failure at all. It could be a problem with an auxiliary power supply, a solenoid or relay driver, a malfunction at a valve, or even a fault in wiring or the associated connectors. Oh, a hardware failure.
Ever seen a short-term software failure? A direct command was mistakenly implemented?
I've seen this sort of thing many times. Someone is blaming the software for a problem rather than confessing the real problem. The software people are down the food chain from the actual culprits.