raid1, cylinder, LBA and inexistent partitions

Hi all,

A customer of mine have a sistem w/ very simple raid1 configuration. Yesterday, the UPS has been manually switched off by mistake. The system woke up in degraded mode but in a strange way: one of the discs does not seem to be partitioned, according to devfs. So that I cannot use mdadm to assemble the partition.

More: fdisk, cfdisk and /proc do not agree on the unmber of cylinders. Only cfdisk sees it as remapped in LBA mode (9964) the others see 158816: does that mean anything usefull? Should I change the disk (3 weeks) or can I try something different?

Thanks in advance
sandro
*:-)

PS: not to mention that the disk on bus1 was partitioned exactly as the other one...

# tree /dev/ide
/dev/ide
`-- host0
    |-- bus0
    |   `-- target0
    |       `-- lun0
    |           |-- disc
    |           |-- part1
    |           |-- part2
    |           |-- part5
    |           |-- part6
    |           `-- part7
    `-- bus1
        |-- target0
        |   `-- lun0
        |       `-- disc
        `-- target1
            `-- lun0
                `-- cd

root@fw-voltimum root # cat /proc/mdstat Personalities : [raid1]
read_ahead 1024 sectors
md0 : active raid1 ide/host0/bus0/target0/lun0/part1[1]

      32000 blocks [2/1] [_U]

md1 : active raid1 ide/host0/bus0/target0/lun0/part5[1]

      995904 blocks [2/1] [_U]

md2 : active raid1 ide/host0/bus0/target0/lun0/part6[1]

      1951744 blocks [2/1] [_U]

md3 : active raid1 ide/host0/bus0/target0/lun0/part7[1]

      77055616 blocks [2/1] [_U]

Sandro Dentella  *:-)
e-mail: [email protected] 
http://www.tksql.org                    TkSQL Home page - My GPL work