fsck problem after fdisk

Erika Pacholleck pchllck at nexgo.de
Mon Oct 1 13:12:33 PDT 2001

I had to change my partition layout a bit, and after some nice
words on irc I felt couragous enough to try it (quite an adventure
without a backup media big enough to make things safer).
I am working with linux fdisk which never gave me problems, no
matter what I did - well, there is a first time for everything ;)

previous layout:
hdc1      1 - 32775 extended containing hdc5 - hdc21
hdc2  32776 - 39560 primary

new layout:
hdc1      1 - 26629 extended containing hdc5 - hdc19
hdc3  26630 - 28678 primary previous ext hdc20
hdc4  28679 - 32775 primary previous ext hdc21
hdc2  32776 - 39560 primary

I had to do the fdisk in single mode from within hdc2's lfs, so I
switched all tasks off and mounted hdc2 as ro.

I did the whole thing reverse again from new to previous layout.
This happened as a comparision:
1. verify gave different unallocated numbers 
2. write the table both times said
- callin ioctl() to re-read partition table
- Re-read partition table failed with error 16: Device or resource
- busy. Reboot to be sure the table is updated.
3. I rebooted the reversed action and all was ok - like before
3. I rebootet the new layout version and all except now hdc4 was ok
   Calling fdisk here now notes: 
   partition table entries are not in disk order
   But despite of that hdc2 works fine (no wonder - it's an lfs)

And here fsck complains about superblock corrupt blahblah...
My 1gb and 3gb partitions have both Blocks 4096, so I assume the
hdc4 2gb is the same (I did not specify any special things) and
therefore the alternate superblocks should also be the same numbers
as the others. I tried 3 other alternates, but without success.
Someone said try fsck version 1.25 instead of now 1.19. - no change.

Does anyone know what is causing this behaviour? And if not, what
are your ideas of it?

SUSE-6.4 on hdc14  
MDK-8.0  on hdc15
LFS-2.4x on hdc2 so lilo update is no problem.
2gb Data on hdc21 could be transferred wherever is space
So I am ready to veri/falsify assumptions and find the anwer.

Erika Pacholleck <pchllck at nexgo.de> @@ -18,9 +18,12 @@ -p+pa -h+ho -c+ec
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe blfs-support' in the subject header of the message

More information about the blfs-support mailing list