Re: 2.1.125 filesystem corruption under extreme load

Neil Conway (nconway.list@ukaea.org.uk)
Sun, 29 Nov 1998 15:28:24 +0000


Nicholas J. Leon wrote:
>
> On Sat, 28 Nov 1998, Neil Conway wrote:
>
> # > I just found that the 2.1.125 kernel corrupted my ext2 filesystem under
> # > extreme load.
> # > WHat I did:
> # > make -j modules
> #
> # This sounds exactly like what I did under 2.1.108, compiling 2.1.129.
>
> I just tried this on a 2.1.130 SMP ix86 kernel (192MB)
> (http://neko.binary9.net/scripts/sysinfo) and only got some
>
> vfork: Resource temporarily unavailable
>
> after my load hit about 55 or so. They'd come in groups of 2 or 3, then
> another gcc launched, then a few more. Other than that, it worked just
> fine.
>
> [dump question] what resource wasn't available? [/dump]

Hmm, guess this isn't an easy to reproduce problem or else it's gone in
the newest kernels. I presume the resource that was unavailable in your
case was spare process slots? Did you have lots of sleeping processes
already? Hard to be sure, as I would have thought the load would go way
higher than 55, but were you watching
the total process count with top or something?

In my case, I had almost nothing running, maybe a few dozen procs, and
then the make started about 400 processes, and then I interrupted it
before seeing any resource problems, but the damage was done.

Neil

-
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/