[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 2.8 read only file system
- To: misc_(_at_)_openbsd_(_dot_)_org
- Subject: Re: 2.8 read only file system
- From: Toni Mueller <openbsd-misc_(_at_)_oeko_(_dot_)_net>
- Date: Tue, 5 Dec 2000 13:53:48 +0100
- Reply-to: openbsd-misc_(_at_)_oeko_(_dot_)_net
On Mon, Dec 04, 2000 at 06:11:19PM -0800, Seth Arnold wrote:
> Jordan, since you are the first to post this, I am going to assume
> something is broken on your end, rather than in OpenBSD.
I'm not sure here, in general, see below.
> In order to remount your hard drive read/write, you can use mount(1) to
> do the job with -u and -w flags.
No, this doesn't work. Don't ask me why, but when fsck doesn't
complete on the root partition saying that the partition is
clean, even after it has declared to have repaired something,
the root partition won't be mounted r/w, and no update of file
system status happens. In every such case I had to use a spare
disk to mount the target root partition there and fsck it w/o
requiring to use it, then reboot from the target disk again
afterwards. This was also the case in 2.7 (for me, that is).
When the system boots from the just-repaired disk but w/o this
second-disk approach, then this is just a circle, with
mount -u -w / yielding "Device busy" etc., thus preventing
any attemted change to mtab from succeeding. It may be that
such a remount could succeed if one prevents the machine from
going into securelevel 1 at boot time, however, but I didn't
[ broken and cat for readability ]
> I have tried /sbin/mount -u /dev/sd0a /, and I get
> mount_ffs: /dev/sd0a on /:
> specified device does not match mounted device.
This fuels the idea of your CD still being in the drive.
> My fstab looks like this:
> 1Hdev/sd0a / ffs rw 1 1.
You might prefer a line reading
/dev/sd0a / ffs rw 1 1