Subject: Re: large FFSv2 -> file system corruption
To: None <current-users@netbsd.org>
From: Jan Schaumann <jschauma@netmeister.org>
List: current-users
Date: 07/20/2004 23:04:31
--n/aVsWSeQ4JHkrmm
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Pavel Cahyna <pavel.cahyna@st.cuni.cz> wrote:
> On Tue, 20 Jul 2004 03:51:59 +0000, Jan Schaumann wrote:
>=20
> > I'm experiencing some serious problems on one of my machines with a
> > large FFSv2 system.

> A bit off-topic, but is there any reason for using UFS2 instead of FFSv1?

Well, aside from it being a whole lot faster to create and fsck, I
encountered problems with ffsv1, too.  ``Yay''.

After newfs'ing the file system using FFSv1, and trying to populate the
file system, I get a kernel panic:

/anon-root: bad dir ino 76362257 at offset 51632: mangled entry
panic: bad dir                                                       =20
Stopped in pid 366.1 (pax) at   netbsd:cpu_Debugger+0x1:        ret


Now doing a reboot from the kernel debugger doesn't even work (even
though I tried a 'reboot 0x4').  The machine will need a hard-reboot.
``yay'', again.

What the hell is going on here?

-Jan

--=20
Tradition is the illusion of permanence. -- Woody Allen

--n/aVsWSeQ4JHkrmm
Content-Type: application/pgp-signature
Content-Disposition: inline

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

iD8DBQFA/d0/fFtkr68iakwRAkzhAKC4LYmLkApltewR72qe/dn8vng0mgCg6B2W
FxfA/YBa4eTcWlfVTeGWaTs=
=D3fC
-----END PGP SIGNATURE-----

--n/aVsWSeQ4JHkrmm--