Lost etc, now what? Warning: Long winded story.
Hi all, sorry for the length of this: I had a dual boot setup, running suse 8.0 and windows 98se. I haven't booted into windows for almost a year now. It's on a box I built, asus a7v133 mb, athlon or duron 900, 2x256 mb cl2 name brand ram (forgot name), soundblaster live value pci card, hardware based US Robotics V90 modem (was removed to use in another computer, and mwave something was one of the error messages on every boot up, both prior to removing the modem, and afterward as well, one of the messages that it was advised to ignore, and I never used the modem under linux, so I don't know if it worked or not) adaptec ide raid card with one drive (other drive failed earlier, not related problem) in a raid 1 setup. This is the drive that boots both os's. a second drive is on the primary controller, and the secondary controller has a cdrom drive. The second drive is for data only. I partitioned the raid drive in the following manner: sda1: windows sda2 or 7: boot, ext2 sda4 or 5: extended partition sda8:reiserfs, / sda9: reiserfs, /home sda10: vfat sda11:vfat sda12:vfat I'm doing this from memory. The numbers may be off, but the basic layout is right. A couple of the missing numbers are swap (more than one swap partition was used), and perhaps etc, or etc is under / The box is/was my main workstation. It's on a dsl lan, behind a hardware device performing nat. It is completely invisible during port scans, with no open ports. I can't even ssh into the box from outside the nat device. I can't find the recovery floppy that I made when I first set this box up. Before the box failed, I tried making a rescue floppy, but it asked for the installation media, which I gave to someone else (we're on a tight budget and splitting the cost of the professional disks), so I couldn't make the rescue floppy again. After checking the logs recently, I was getting more errors: major modules, 108, 109, 111, 112, 114, and similar. Earlier, it had been confined to errors that I had looked up, and the advice was to ignore them, it was a bug in the release. Related errors were because I tried to install a cd burner, was never able to get it to work, so I removed it and installed elsewhere. But there were a few more lines in the logs that had me worried. Also, in the last few weeks, the box started freezing up, and the only way to correct it would be to reset the box. When this started happening, I decided to manually write down the partition table, in case something went wrong. I had used yast's online update to update the security patches once a week or more often. I don't run any servers on the box. So I logged on as root in X, planning on updating through online update, writing down the partition table through yast's partitioner, and a few other chores that are simply easier to do when logged on as root, as opposed to su'ing into the features. I had the partitioner open, when what can go wrong, does. The box crashed. I reset the system, and now nothing happens. I'm not sure if /etc disappeared right after the system crashed, or a few reboots afterward. It may have happened as a result of the above description of actions, or a hacker succeeded in accessing the box and wiped /etc. Some of the error messages are: hostname lookup failure Xauth: error while loading shared libraries: libXmuu.so.1 cannot open shared object file: No such file or directory Xinit: error while loading shared libraries: libXmuu.so.1 cannot open shared object file: No such file or directory # /bin/df: Warning: cannot read table of mounted filesystems: No such file or directory So I try to reboot, switch runlevels, etc, and: # init0 INIT:Switching to runlevel:0 INIT: No inittab found Can't open (/etc/ioctl.save 0_WRONLY) No such file or directory no more processes left in this runlevel The only thing I can do is log on as a single user, rescue, where I see the directories under / , but I can't go into them because I get error messages that tell me there is no /etc/fstab, or mtab. /etc doesn't appear in the list of directories under / . I was able to boot a knoppix disk, with a knoppix boot floppy, on the box. I can locate my data, and locate most of the linux directories (I haven't made a comprehensive list yet, but the boot partition is there, var, bin, sbin, usr, and most of the others are there. But /etc is missing. I don't recall if I originally made /etc a separate partition or not. It does not appear under / . Since knoppix gives you a convoluted directory structure (it has it's own / and other directories, plus it's own similar directories under the sub-directory of "KNOPPIX", I don't know if there is a missing partition that is not mounting, which may contain the files belonging to etc, but it doesn't appear to be the case. There is also another issue which you may be able to help me on. I seem to have trouble accessing some of my own data files and directories. The knoppix distro is reporting that some of the data directories on both the raid disk, and the data disk are owned by root, and the group is root. I believe that it is likely this is an error, that the owner and group should be a different owner, with group "users". Anyway, I am logged on as root, not knoppix user, when using the knoppix disk to look at the filesystems and data. I still can't access the data. Is this because my knoppix root user is a different user id than the suse root user? And my knoppix user id, and "users" group id are different than the suse group id for "users"? If so, where are the id numbers kept? /etc? Or is this a problem because /etc is missing? I have looked up solutions for a missing partition table, and rebuilding fstab. I have a suse 7.3 desktop that I can look at as an example for fstab, but what do I do when the entire /etc directory is gone? Should I attempt to install 8.2 over this 8.0 installation? There is too much data on several of the partitions to back them up. /home is simply too large, along with a few other partitions (three of the vfat partitions that I can think of right now), plus the separate data hard disk. In addition, when I did try to backup some of the data to another computer, the backup (dd) failed, because it said that some filenames were too long, and it simply stopped the backup right where it ran into the problem. I don't have the space anymore to backup the data elsewhere. And since I never was able to get the cd burner working, I can't backup to multiple cds (which was the original reason for buying the burner, backups), and couldn't backup the copies of the suse 8.0 disks. I do have suse 7.3 professional disks that I can use for getting files. This will take some time to sort the issues out, as I don't have regular 'net access anymore due to the failure of the above workstation. TIA for any help/suggestions. Bing. -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ All spam received is reported to http://spamcop.net Any spam designed to avoid spamcop processing is processed and reported manually to all the isps in the headers. Don't send it. I don't respond to it.
lists1 wrote:
sda1: windows sda2 or 7: boot, ext2 sda4 or 5: extended partition sda8:reiserfs, / sda9: reiserfs, /home sda10: vfat sda11:vfat sda12:vfat
Maybe reiserfsck is what you're looking for. You should can find the man page online easily enough. Then boot into single-user mode and try to get reiserfsck to rebuild the damaged partitions. -- JDL Non enim propter gloriam, diuicias aut honores pugnamus set propter libertatem solummodo quam Nemo bonus nisi simul cum vita amittit.
On Wed, 25 Jun 2003 06:33:24 -0400 lists1 <lists1@pilosoft.net> wrote:
Hi all, sorry for the length of this: You can boot from the SuSE distribution CD. I don't recall if 8.0 had a rescue option, but if not, you can begin the install, then use ctrl-alt-fn to get a shell. I don't recall exactly which virtual terminal is available. Once you do this, then take a look at your partition table. My thoughts is that somehow your root file system is probably not /dev/sda8.
-- Jerry Feldman <gaf@blu.org> Boston Linux and Unix user group http://www.blu.org PGP key id:C5061EA9 PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9
The 03.06.25 at 06:33, lists1 wrote:
installation media, which I gave to someone else (we're on a tight budget and splitting the cost of the professional disks), so I couldn't make the rescue floppy again. After checking the logs recently, I was getting more errors:
You should have copied, at least, the 1st CD, for the rescue system.
/etc doesn't appear in the list of directories under / .
From a rescue system (I think yo said you have SuSE 7.3 CD available) you can use fdisk to see all the existing partitions. Then, manually mount each one on /mnt, check what they are, what type, etc), and see if any one appears to be /etc - no, that can not be. I don't think "/etc" can be a separate partition, because no partition can be mounted without reading /etc/fstab first.
Any way, check them: it's the way to discover what fstab should have. Run reiserfsck something to check and repair filesystem, I think you said you have a reiser root filesytem. If not, fsck whatever - from a rescue system, even from another SuSE version.
I have looked up solutions for a missing partition table, and rebuilding fstab. I have a suse 7.3 desktop that I can look at as an example for fstab, but what do I do when the entire /etc directory is gone?
Rebuild from backup ;-)
stopped the backup right where it ran into the problem. I don't have the space anymore to backup the data elsewhere. And since I never was able to get the cd burner working, I can't backup to multiple cds (which was the
Too bad. I would try to free a relatively small partition, install SuSE on it, from scratch, and attach to it the needed data partitions from the old install, which would be deleted on the end. I wouldn't like to be on your skin, though :-) -- Cheers, Carlos Robinson
participants (4)
-
Carlos E. R.
-
Jerry Feldman
-
John Lamb
-
lists1