10.2-Beta i386..what's wrong..?

Holm Tiffe holm at freibergnet.de
Fri Jul 24 06:59:12 UTC 2015

Glen Barber wrote:

> On Thu, Jul 23, 2015 at 08:42:44PM -0400, Brandon Allbery wrote:
> > On Thu, Jul 23, 2015 at 8:40 PM, Mark Linimon <linimon at lonesome.com> wrote:
> > 
> > > zfs is a resource hog.  i386 is not able to handle the demand as well
> > > as amd64.
> > >
> > 
> > Even amd64 is no guarantee. I installed one of the Illumos spinoffs on a
> > 2GB amd64 netbook (they mostly force zfs). I think it lasted 2 days before
> > the kernel panics started.
> > 
> Even on amd64, you need to tune the system with less than 4GB RAM.
> Glen

..interrestingly people here seem to focus my problem to ZFS.. but my
problem was to build an raid over 4 disks on my old i386 machine and that
failed with 2 different approaches.

I'm accepting that ZFS is a too big thing for the i386 architecture
and I possibly should leave it alone on that machine.

But my 2nd try with gvinum failed also ...why?

In the meantime I've set up the first two disks to a geom_mirror and
installed 2x swap and a 66G ufs on the mirror, successfully installed
10.2-Beta, pulled the sources with svn and rebuild the entire world and a
kernel yesterday. That worked flawlessly.
So the hardware is out of the question here.

It seems that geom_vinum is broken and it is broken on 9.3 10.1 and
10.2-Beta. There isn't an other possibility anymore.
Can someone please confirm this with a similar machine?

Proably I could atach an IDE disk and move the new build system to it
to give ZFS a 2nd try on the 4 SCSI Disks to prove if something changed n
the meantime or if a kernel with KSTACK_PAGES=4 would fix the ZFS problem.

I had running gvinum and gmirror on that machine in the past up to
8.4-stable. It seems that we've lost this possibility now..


      Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, 
     Freiberger Straße 42, 09600 Oberschöna, USt-Id: DE253710583
  www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

More information about the freebsd-stable mailing list