As I already said, disabling overcommitment altogether is not *the* solution:
Say process A uses up all memory available and just sits there, trying to eat
up more memory as it becomes available. Since malloc simply fails, the process
(handling malloc's failure correctly) will not die. Such situation can and do
occur without even any bad-intentionned user.
But next thing you know, X (or init or whathaveyou) needs some memory and
fails, probably killing itself at the same time. Now is that a sane situation?
Basically, you end up killing the first process which *really* needs memory.
Stefan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/