[clug] bebug puzzel

Scott Ferguson scott.ferguson.clug at gmail.com
Sat Feb 1 05:41:04 MST 2014


On 01/02/14 23:00, Wesley Bruce wrote:
> Got you scott, there was another line I missed when I cut and pasted the
> message from where I had written it. 
> Dropping to debug shell.
> 
> sh: can't access tty; job control turned off
> dracut:/#
> 
> 
> Sorry I missed that line. Does this help? 

It just confirms what I'd said earlier, part of the boot process runs
fsck on your /root slice, if it fails the slice is remounted readonly
and you'll get that precise error message. The reasons for failure are
numerous - they can be as minor as a timestamp in the future (common on
a dual boot box given Windoof's way of handling UTC, not) or sector
failures. Sector failures can be minor or catastrophic. It's unlikely
you have catastrophic sector failures if you have Windoof running
without complaint on the same drive. Generally you would have had errors
while the system was running if your disk was seriously failing - you
*would* have noticed them.

e2fsck -p won't break anything and the filesystem is mounted read only
anyway.

*Do the following if the slice is formatted with an ext* filesystem see
if e2fsck can fix the problem*

If you have xfs, jfs, reiser, or btrfs filesystem then the fix is
different. Depending on how old your Ubuntu installation is, and whether
you customised the filesystem it should be either ext3 or ext4. The
latter is easier to fix, but both should be simple to recover in most cases.

If the problem is simple *and* you have a journaled file system the fix
will be quick and simple. e2fsck will tell you if the problem is not simple.

from where you get that message:-
type and hit Enter (firmly):-
e2fsck -p `mount | grep "\/ " | grep \(ro`

NOTE the spaces and the use of the back tick(? next the the number 1 key)

explanation:-
mount | grep "\/ " | grep \(ro

looks for a root file system mounted as read only

When and if ex2fsck finishes successfully we can see if you have
problems with other slices - if not you should install and configure
smartd to check for a failing hard disk drive. Post back and I'll give
detailed instructions.

> 
> 
> 
> On Sat, Feb 1, 2014 at 10:45 PM, Wesley Bruce <wesleyianbruce at gmail.com
> <mailto:wesleyianbruce at gmail.com>> wrote:
> 
>     OK It has two drives each with two partitions with windows on C
>     partition 1 drive one (disk 0), Ubuntu appears to be nested in that
>     too. there's a 3 gig empty partition on that drive too. That was
>     where the Linux was supposed to go but it misbehaved (years ago on
>     first update ). The other drive is disk 1 and has E drive  781 gig;
>     Important stuff. and a second 150 gig unallocated partition. This
>     last partition is the one windows seems to have changed from healthy
>     drive to Unallocated and I think that's the only change that
>     occurred when the linux stopped working. I seem to remember that was
>     what it was using. How does this change the above command string? I
>     don't want to take any action, that may loose any further data,
>     until I understand that I'm hunting the right drives. C and E are
>     NTFS; will they be ignored automatically? Its a byzantine PC I even
>     still have a floppy drive on this Rube Goldburg machine. 

Thanks.

Windoof doesn't use intuitive slice naming so it's not clear which slice
you installed Linux to - I'm presuming/assuming you don't have a
separate /boot and /root. You don't say what your boot loader is or if
you change boot order using BIOS.

Windoof does have a nasty habit of prompting users to format Linux
partitions - but it doesn't sound like that is what has happened (even
so it's recoverable without much difficulty).

If you have a live Linux CD/DVD, ideally Knoppix microkernel you could
run it, mount the linux slices and copy or email /etc/fstab to the list
(or me) and we/I could give more precise information.
I'd just use the e2fsck command I gave a the top of this post - the
slice is mounted ro and e2fsck -p is gentle.

> 
> 
>     On Sat, Feb 1, 2014 at 7:12 PM, Scott Ferguson
>     <scott.ferguson.clug at gmail.com
>     <mailto:scott.ferguson.clug at gmail.com>> wrote:
> 
>         On 01/02/14 18:45, Wesley Bruce wrote:
>         > bebug, bebug, that shows how frazzled this heat's got me. I
>         mean debug. you
>         > all knew that I bet.
>         >
>         >
>         > On Sat, Feb 1, 2014 at 4:51 PM, Wesley Bruce
>         <wesleyianbruce at gmail.com <mailto:wesleyianbruce at gmail.com>>wrote:
>         >
>         >> I'm back. And I've blown up my computer again.lol.
>         >> Windows messed with a empty partition the linux side of my
>         dual boot uses
>         >> so I get a debug message when trying to start the linux. 

<snipped>
>         >>
>         >> I get this :
>         >> Dropping to debug shell.
>         >>
>         >> sh: can't access tty; job control turned off
>         >>
>         >> I've been told to run fsck

<snipped>


Kind regards


More information about the linux mailing list