[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Superblock unreadable with upgrade from 2.2.20 to 2.4.18-586tsc kernel on Woody



I am running Debian 3.0r1 (woody/stable).  Initially, I installed kernel-2.2.20.  I would like to upgrade to kernel-2.4.18-586tsc.

My system has two harddrives.  The second harddrive contains only a single extended partition divided into a logical linux swap partition and a logical ext2 partition.  This harddrive is a Western-Digital Cavalier 2340 (Capacity: 341MB, 3322 RPM, 13ms (Read), ECC (Reed Soloman), No CHS(?), No LBA, Settings: DMA).  Both the swap partition and ext2 partition operate with no problems with kernel-2.2.20 both before and after obtaining the fsck complaint noted below.

1) During the process of upgrading to kernel-2.4.18-586tsc, fsck complains when trying to read the superblock of the ext2 partition on the second harddrive.  See Installation message below.  After installation, executing "e2fsck -b 8193 /dev/hdb6", or "e2fsck -b 24577 /dev/hdb6", ... to correct the problem, repeats the message.

2) After installation and then booting with kernel-2.4.18-586tsc, the swap partition on my second harddrive is never added to swap space.

As I have alternated running the machine with kernel-2.2.20 'vanilla', kernel-2.4 'bf2.4', and kernel-2.4.18-586tsc, through a number of installations, reinstallations and upgrades, over the past month, and as the drive is alway readable with 2.2.20 and the partition table is always unreadable with either of the two 2.4 kernels, and as the drive had no problems when running Windows 98 a month ago, I question if something is going on in the 2.4 kernel or if something is wrong with the harddrive.

Might the 2.4 kernels (kernel-2.4.18-586tsc, bf2.4) not include a kernel option that is included in 'vanilla' which would solve my problem?  Say because this is an old disk that does not support LBA and CHS?

Are there particular reasons why the 2.4 kernels might be picking up something wrong with my harddrive that the 2.2.20 kernel is not?  Are the 2.4 kernels more sensitive?

Can anyone offer any other suggestions as to what I might try to do to eliminate the fsck and ignored swap partition problems?

Detailed messages and logs follow.

Thanks,

-- Mark
DrMarkAFriedman@earthlink.net

=======================================================================


Note: Installation of Woody on my machine with bf2.4 proceeds through creating partitions /dev/hdb6 and /dev/hdb6, however, both partitions are then unreadable and ignored through the remainder of the installation as their partition tables are immediately corrupt after they are written.


Note: Repartitioning the drive to contain two primary partitions using fdisk (as opposed to cfdisk did not change the problem.}


Installation messsages: 
======================
fsck 1.27 (8-Mar-2002)
fsck.ext2: Invalid argument while trying to open /dev/hdb6
/dev/hdb6:
The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2 
filesystem (and not swap or ufs or something else), then the superblock
is currupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <devide>

fsck failed.  Please repair manually.

CONTROL-D will exit from this shell and continue system startup.


Contents of /var/sys/messages during booting with kernel-2.4.18-586tsc:
======================================================================
Feb 25 21:52:40 braniac kernel: /dev/ide/host0/bus0/target1/lun0: [PTBL] [787/128/63] p1 p2 < p5 p6 > p3
Feb 25 21:52:40 braniac kernel: /dev/ide/host0/bus0/target1/lun0:hdb: lost interrupt
Feb 25 21:52:40 braniac kernel: hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 25 21:52:40 braniac kernel: hdb: drive not ready for command
Feb 25 21:52:40 braniac kernel:  unknown partition table
Feb 25 21:52:40 braniac kernel: Adding Swap: 644476k swap-space (priority -1)
Feb 25 21:52:40 braniac kernel: Real Time Clock Driver v1.10e
Feb 25 21:52:40 braniac kernel: hdb6: bad access: block=2, count=2
Feb 25 21:52:40 braniac kernel: end_request: I/O error, dev 03:46 (hdb), sector 2
Feb 25 21:52:40 braniac kernel: EXT2-fs: unable to read superblock


Output from initial initialization of this filesystem:
=====================================================
Select /dev/hdb6 to initialize as a Linux ext2 filesystem and run a bad-block scan
    mke2fs 1.27 (8-Mar-2002)
    Filesystem label=
    OStype: Linux
    Block size=1024 (log=0)
    Fragment size=1024 (log=0)
    60000 inodes, 239552 blocks
    11977 blocks (5.00% reserved for the super user
    First data block=1
    30 block groups
    8192 blocks per group, 8192 fragments per group
    2000 inodes per group
    Superblock backups stored on blocks:
            8193, 24577, 40961, 57345, 73729, 204801, 221185
Mount as user directory filesystem "/home"




Reply to: