tags 285257 - patch
thanks
[Petter Reinholdtsen]
> Can anyone test and verify that it solves the problem?
I got Sven Luther to test on a powerpc machine, without much success.
He was using test filesystem and used resize2fs to reduce it and then
tried to use ext2resize to reduce and extend it. The log messages are
from IRC, I hope I got it right.
Reducing the file systems using -v option gave this message:
ext2resize v1.1.19 - 2001/03/18 for EXT2FS 0.5b
ext2resize: 500000 is a bad size for an ext2 fs! rounding down to 499713
group 117 inode table has offset 2, not 7
group 118 inode table has offset 2, not 7
using 0 reserved group descriptor blocks
And then nothing happened. It had to be killed.
Extending the file system had more success, but gave a currupt file
system. These messages appeared during resize:
cache direct hits: 4152, indirect hits: 22, misses: 64
cache direct hits: 4152, indirect hits: 22, misses: 64
new filesystem size 971046
using 0 reserved group descriptor blocks
relocating blocks....
don't need relocation for group 0
don't need relocation for group 1
...
don't need relocation for group 62
...flushing buffer 63/block 507905
...
...flushing buffer 119/block 966657
cache direct hits: 4152, indirect hits: 22, misses: 64
fsck gave these errors after the resize:
Root inode is not a directory. Clear<y>?
Inode 8 has illegal block(s). Clear<y>?
Illegal block #0 (318898176) in inode 8. CLEARED.
Illegal block #1 (335675392) in inode 8. CLEARED.
Illegal block #2 (352452608) in inode 8. CLEARED.
Illegal block #3 (369229824) in inode 8. CLEARED.
Illegal block #4 (386007040) in inode 8. CLEARED.
Illegal block #5 (402784256) in inode 8. CLEARED.
Illegal block #6 (419561472) in inode 8. CLEARED.
Illegal block #7 (436338688) in inode 8. CLEARED.
Illegal block #8 (453115904) in inode 8. CLEARED.
Illegal block #9 (469893120) in inode 8. CLEARED.
Illegal block #10 (486670336) in inode 8. CLEARED.
Too many illegal blocks in inode 8.
Clear inode<y>?
Inode 11, i_blocks is 402653184, should be 0. Fix<y>?
I conclude that the patch is not enough to fix the problem on big
endian machines.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
