>
> I can understand the problem there, but surely a better solution would
> be to have a limit such that no one process may use more than (say)
> 90% of non-kernel system memory, thus ensuring that there's always
> some memory free, and have that limit tweakable...
You might end up with 10% free memory being ~ 0 ..? If I understand you
correct something like this could happen:
1) Process A starts, allocates 90% of non-kernel system memory
(leaving 10%).
2) Process B starts, allocates 90% of non-kernel system memory
(leaving 1%).
3) Process C starts .....
>
> Also, just a thought, but would more swap space help?
It would. Problem is, we don't know how much ....! But our system is
designed to 'back off' if it gets a 'out of memory' which it can recover
from (it does on our HP-UX systems ...)
I just tried to emulate the HPUX way of doing things.
BTW, nobody have told me what they think ... ;o)
>
> Best wishes from Riley.
-
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/