Hello.

I posted some of this previously and got one reply from bigrigdriver. I guess he went offline, but I still need some authoritative help. If anyone knows about these things, your assistance will be greatly appreciated.

I've been running Ubuntu 5.10 (Breezy) for a while with no special
problems. It's on a separate hard drive in a removable drawer. I'm not talking about USB or other removable media. A computer has room and cables for two hard drives (at least). The drawers, which I've been using for years, simply plug in to the cables. The frame of the drawer is permanently connected to the cables leading to the motherboard. The removable part simply plugs in with pins to the frame. That way, you can remove one hard drive and put in another without opening the machine. However, this is exactly the same as if you did open it and change the hard drives.

The Linux disk had been out of the machine for about 24 hours. It had been working fine and I had made no recent installations or other changes.

When I put in the disk and tried to boot into Ubuntu as usual, I got this:

Alert! /dev/hdc1 does not exist. Dropping to shell!

It just hung on that.

After going to Windows and coming back, I got the same thing several
times. Once, it continued, saying:

BusyBox v1.00-pre10 (Debian 20040623-1ubuntu 22) Built in shell (ash).
Enter 'help' for a list of built in commands.
/bin/sh: can't access tty: job control turned off
#_

Usually, it doesn't get that far, but just hangs on the first message.

Can anyone tell me how such a thing could happen and what - if
anything - I can do about it?

Bigrigdriver suggested this may have been caused by removing a hard disk before it was unmounted. I've never done that. I only change hard drives after going out of Linux (or Windows) in the proper manner and after the machine is turned off. However, I have some reason to suspect that another user on this machine is not quite so careful. It's not impossible that he hit "Shut Down" and immediately pulled the hard drive.

In any case, he said this may be fixable (I've spent a lot of time and effort setting up this system and it would be a shame to have to format and start again...especially since I can't even get in to copy out all the how-to's I had been compiling in case I needed to retrace my steps).

Under Windows, I would have a good idea where to start to attempt a
fix (although, if Windows breaks, it's often hopeless). At minimum,
I'd have a good chance of recovering data. But, I'm a Linux
newbie...at least as far as anything like this is concerned.

I do have an Ubuntu Live CD, but have never tried to use it (hope it
boots). Bigrigdriver mentioned fsck, but it has a lot of optional parameters and, if I understand correctly, I have to be sure that no part of the Linux system is mounted, or I could kill it for good.

I would think that, if the Linux hard drive were in the machine and I just booted from the Live CD, nothing would be mounted and I could proceed. Is this correct?

Searching around, I found that fsck won't run if superblock is corrupted. This is the quote:
------------------------------------------------------------------
"If the superblock is corrupted the file system still can be repaired using alternate superblock which are formed while making new file system .

the first alternate superblock number is 32 and others superblock numbers can be found using the following command :

newfs -N /dev/rdsk/c0t0d0s6

for example to run fsck using first alternate superblock following command is used

fsck -F ufs -o b=32 /dev/rdsk/c0t0d0s6"
------------------------------------------------------------------
How would I know if I needed to do this? Can I do it from Live CD?

What I need is a one-step-at-a-time walk-through, in baby talk. If
anyone knows how to go about this, I'd certainly appreciate the help.

Thanks very much.

Are both drives in removable bays?
If so have you changed the order to which they are installed?
Is one drive that was in there during the install no longer there?
I would make sure the system drives are plugged in the right bays and the drive is still set to be master or slave in the process during the install. If you had 2 drives plugged in at boot and linux install on the second then the bootloader on the master boot record is on the first drive. See if any of this helps.

Brian1

Thanks for the reply.

I have two drawers, one above the other. The bottom is usually Windows and the top one is whatever else I'm using at the moment. I could reverse that, but try not too because it gets confusing to know which boot sequence to use if you don't keep something constant.

If I have disks in both drawers, I change the boot sequence in CMOS. Both drives are masters.

Often, just because the computer runs faster with only one drive, I remove the bottom disk...then boot in without changing anything in CMOS. With nothing in the bottom drawer the machine boots to the top one. This is often how I run Linux and it is how I originally installed it.

I only have one disk in the machine when I'm installing an OS..to avoid crossovers where one system is actually partially installed on the other other disk, which would require both to be in the machine forever and booted in the right way (is that what you meant?). I remove the bottom drawer and put the disk I want to install to in the top...then boot from the installation CD. Thus all disks are always bootable masters.

My plan, if I need to use the Live CD, was to have Linux in the top drawer, as usual, with the bottom empty, and boot from the CD. This was the setup when I installed Ubuntu. Thus the CD can only act on that system.

I need to know, first of all, if that will allow me to try to run fsck from the Live CD with the certainty that nothing in Ubuntu is mounted. Sounds logical, but what do I know?

Thanks.