Messages in this thread | | | From | James Sutherland <> | Subject | Re: Avoiding OOM on overcommit...? | Date | Mon, 20 Mar 2000 13:44:58 +0000 |
| |
On 19 Mar 2000 18:00:42 -0800, you wrote:
>In article <linux.kernel.45hadsgku4f59qae3ouohgbk7k4p6lc5os@4ax.com>, >James Sutherland <jas88@cam.ac.uk> wrote: > >>Now we'll take a WWW server, with 100 processes forked, all sharing >>most of the image. You just blew 2Gb or so of my swap space, to >>achieve - nothing. > > Okay, I'm getting really curious here: what application do you > have that requires that you run 100 copies of a web server each > with 20mb of unique writable data?
I don't. The 20Mb is almost all shared across the processes - but it IS mostly writable. With overcommit, it's fine. Without it, I'd need to throw an extra Gb or so of swap into the machine, just to overcome the non-overcommit bug.
Disabling overcommit does NOT prevent ANY problems. It just changes the nature of a problem, making it much more frequent and a bit more predictable. A bit like compensating for an occasionally leaking seal on a submarine by removing the seal entirely.
> david parsons \bi/ I simply avoid the Linux overcommit bug by dropping > \/ half a gigabyte of ram into my workstations. You don't need to add RAM to overcome overcommit - you'd need to throw a Gb or so of swap at them to compensate for NOT having it.
James.
- 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/
| |