"Jim C. Nasby" <jim@nasby.net> writes:
> What would the failure mode be? Would we just keep going until the box
> ran out of memory? I think it'd be better to have some kind of hard
> limit so that a single backend can't grind a production server into a
> swap-storm. (Arguably, not having a limit is exposing a DoS
> vulnerability).
[ shrug... ] If we tried to guarantee such a thing we'd be putting
arbitrary limits into hundreds if not thousands of different bits of the
backend. I think the correct answer for an admin who is worried about
such a thing is to make sure that the process ulimit is a sufficiently
small fraction of the machine's available RAM. Only if we can't
gracefully handle running up against ulimit is it our problem (hence,
we have a stack-size overflow check, but not any such thing for data size).
regards, tom lane
