Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 4 Oct 2004 09:04:53 -0700
From:      Brooks Davis <brooks@one-eyed-alien.net>
To:        Ion-Mihai Tetcu <itetcu@people.tecnik93.com>
Cc:        freebsd-current@freebsd.org
Subject:   Re: /dev/md0 keeps mounting on /var
Message-ID:  <20041004160453.GA7705@odin.ac.hmc.edu>
In-Reply-To: <20041004184708.248efdd8@it.buh.tecnik93.com>
References:  <20041004173545.3af254c6@it.buh.tecnik93.com> <20041004153257.GA30883@odin.ac.hmc.edu> <20041004184708.248efdd8@it.buh.tecnik93.com>

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

--Qxx1br4bt0+wmkIi
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Oct 04, 2004 at 06:47:08PM +0300, Ion-Mihai Tetcu wrote:
> On Mon, 4 Oct 2004 08:32:57 -0700
> Brooks Davis <brooks@one-eyed-alien.net> wrote:
>=20
> > On Mon, Oct 04, 2004 at 05:35:45PM +0300, Ion-Mihai Tetcu wrote:
> > > For some time I keep getting /dev/md0 mounting on /var as the last
> > > mount when going to multi-user. The obvious result is that all
> > > services that use /var fail to start/function properly. Since I
> > > don=B4t use it I suppose I=B4ve done something wrong with mergemaster.
> > >=20
> > > The only workaround I=B4ve found is to rename /boor/kernel/g_md.ko.
> > >=20
> > > And since I just can=B4t fix this since ~ beta4 could someone point me
> > > where to look ?
> >=20
> > This is proably happening because your /var is failing the writability
> > test in /etc/rc.d/var.  One situation I can of where this could happen
> > and you would otherwise have a valid /var would be if you have a
> > read-only NFS root and an NFS /var.
>=20
> No, it is not the case. Does the order in fstab meter ?=20
> /dev/ad0s3d is already mountrd on /var when md0 bumps in.
> >From memory it looks like this:
>=20
>  % mount
> /dev/ad0s3a on / (ufs, local)
> devfs on /dev (devfs, local)
> /dev/ad0s3f on /tmp (ufs, local, soft-updates)
> /dev/ad0s3d on /var (ufs, local, soft-updates)
> /dev/ad0s3e on /usr (ufs, local, soft-updates)
> /dev/ad0s2d on /home (ufs, local, soft-updates)
> /dev/ad0s4 on /vol/it/xp (msdosfs, local, read-only)
> /dev/ad1s3d on /vol/it/cur/h (ufs, local, soft-updates)
> /dev/ad1s2d on /vol/it/n (ufs, local, soft-updates)
> devfs on /var/named/dev (devfs, local)
> /dev/md0 on /var (ufs, local, soft-updates)

Weird.  Order shouldn't matter.  The code in the var script is really
simple.  It creates the md file system if you have varmfs=3D"YES" or if
"/bin/mkdir -p /var/.diskless" fails.  Hmm, I just noticed that ru never
MFC'd the fix to add -p to that command.  Do you have a /var/.diskless
on /dev/ad0s3d?  If so, that's probably the problem.

-- Brooks

--=20
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529  9BF0 5D8E 8BE9 F238 1AD4

--Qxx1br4bt0+wmkIi
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQFBYXSkXY6L6fI4GtQRAkKZAKDZOoKBpLH1SIlIk27e+OdEKFcHgACbBGoG
2IUpng1wVWYxCTau29sdVNY=
=rXIQ
-----END PGP SIGNATURE-----

--Qxx1br4bt0+wmkIi--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041004160453.GA7705>