Subject: Re: other mem-related issue;was: overfilling mfs partitions
To: Rudi Ludwig <rudihl@gmx.de>
From: Bill Studenmund <wrstuden@netbsd.org>
List: port-macppc
Date: 11/27/2004 17:02:14
--8P1HSweYDcXXzwPJ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Nov 27, 2004 at 09:03:13PM +0100, Rudi Ludwig wrote:
> On Wed, 24 Nov 2004 16:15:22 -0800
> Bill Studenmund <wrstuden@netbsd.org> wrote:
>=20
> > .....
> > Check your resource limits. user limits are typically set for 128 MB
> > of data, so it sounds like your program got too big and was told it
> > couldn't grab more memory.
> >=20
> Could you elaborate on this please.

If you are using csh, run the "limit" command. If you are using sh, run=20
the "ulimit" command.

> I checked the kernel config-file, and found no parameter
> had a look at man(8) sysctl, also nothing that triggert me
> nor did /etc/defaults/rc.conf
>=20
> Someone to tell me where to change this per process memory resource
> limit.

For csh (or tcsh) add "limit datasize 262144" to .cshrc. For sh and=20
friends, add "ulimit -d 262144" to .profile.

> Not that I am that eager on running benchmark programms, but I would
> like to understand what happens there.

The default per-process limits weren't big enough to run the program. Or=20
at least that's what I think was wrong. I ran the test program you=20
referenced without issue.

Take care,

Bill

--8P1HSweYDcXXzwPJ
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFBqSOWWz+3JHUci9cRApjQAJ9gRXeAZsc1C2Z2C2UdF0FCxpWjqgCglVAa
9z3ab4aDLuGHjBEl2Y5TbzQ=
=lfml
-----END PGP SIGNATURE-----

--8P1HSweYDcXXzwPJ--