Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 7 Oct 2004 10:02:50 -0700
From:      Kris Kennaway <kris@obsecurity.org>
To:        Chris Elsworth <chris@shagged.org>
Cc:        Kris Kennaway <kris@obsecurity.org>
Subject:   Re: dump/mksnap_ffs & fsck failure
Message-ID:  <20041007170250.GA4053@xor.obsecurity.org>
In-Reply-To: <20041007165520.GA22942@shagged.org>
References:  <20041007110802.GA4960@shagged.org> <20041007133415.GA93146@xor.obsecurity.org> <20041007165520.GA22942@shagged.org>

next in thread | previous in thread | raw e-mail | index | archive | help

--pWyiEgJYm5f9v55/
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, Oct 07, 2004 at 05:55:20PM +0100, Chris Elsworth wrote:

> > > # /sbin/dump -0 -LuaC32 -f /backup/root.dump /
> > > mksnap_ffs: Cannot create //.snap/dump_snapshot: Input/output error
> > >=20
> > > # fsck_ufs /dev/mirror/gma
> > > ** /dev/mirror/gma (NO WRITE)
> > > ** Last Mounted on /
> > > ** Root file system
> > > ** Phase 1 - Check Blocks and Sizes
> > > fsck_ufs: cannot alloc 4294967292 bytes for inoinfo
> >=20
> > Why can't it write to the device?  Is it mounted read-write?  Drop it
> > to read-only and retry.  Vice versa for the root filesystem; is it
> > mounted read-only?
>=20
> Ah, sorry, should have said - yes, the filesystem was mounted
> read-write (/dev/mirror/gma and / are the same thing), but it does the
> same thing in read-only:
>=20
> # init 1
> ..
> # umount -a
> # mount -o ro /
> # mount
> /dev/mirror/gma on / (ufs, local, read-only)
> devfs on /dev (devfs, local)
> # fsck -y /
> ** /dev/mirror/gma
> ** Last Mounted on /
> ** Root file system
> ** Phase 1 - Check Blocks and Sizes
> fsck_ufs: cannot alloc 4294967292 bytes for inoinfo
>=20
>=20
> Surely fsck doesn't really need 4GB of physical mem? There is 1GB of
> physical and 4GB of swap configured in this.

Perhaps it's trying to allocate a negative amount of memory (that
value is 2^32-4, or (unsigned int)(-4)), perhaps because it's failing
to check for an error condition and using the resulting negative
value.

Kris

--pWyiEgJYm5f9v55/
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (FreeBSD)

iD8DBQFBZXa6Wry0BWjoQKURAvCNAJ0TXbSHQB5HAQuA3AsYeNhFQd/00gCg/pw/
PbE1NsA2tzzIQlc4HXSELoU=
=BQcK
-----END PGP SIGNATURE-----

--pWyiEgJYm5f9v55/--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041007170250.GA4053>