Hi,
Thanks for the quick reply! I'll try to give some answers to the questions.
Jari Ruusu wrote:
Any kernel error messages? Use 'dmesg' command to display them.
This is what is logged in dmesg when I tried to chmod the corrupt file
(actually I did a chmod -R to see which files were corrupt)
ReiserFS: warning: is_tree_node: node level 35571 does not match to the
expected one 1
ReiserFS: loop4: warning: vs-5150: search_by_key: invalid format found
in block 3014659. Fsck?
ReiserFS: loop4: warning: vs-13070: reiserfs_read_locked_inode: i/o
failure occurred trying to find stat data of [1311 1314 0x0 SD]
Can you rule out RAM problems?
Does memtest86 run ok?
Is this on brand new computer?
I have not tried to run memtest for a while, but the server has been in
use for quite some time as a firewall. I just connected the disk to it
to test loop-AES. But, I would say for 99% that the memory is okey. I
tested it when I got the server. It is an old IBM with two PII/333MHz.
Nothing special. Anyway, I wouldn't think that this is the problem,
because I have had the same problem in another machine, a PIII.
I can supply you with additional information if you need, because I'm not
sure what might be useful to know.
Kernel version?
2.6.8 SMP enabled (if this might be of interest)
gcc version?
3.3.4
ATA chipset info? Some Silicon Image SATA controllers are really bad.
This is from dmesg:
PDC20262: IDE controller at PCI slot 0000:00:0d.0
PDC20262: chipset revision 1
PDC20262: 100% native mode on irq 16
PDC20262: (U)DMA Burst Bit ENABLED Primary PCI Mode Secondary PCI Mode.
ide2: BM-DMA at 0x6000-0x6007, BIOS settings: hde:DMA, hdf:pio
ide3: BM-DMA at 0x6008-0x600f, BIOS settings: hdg:DMA, hdh:DMA
Is this what you wanted to know?
I just tried to do some more tests with the partition with corrupt files
and I was really surprised that a file, which was corrupt (I couldn't
delete it as root or move/rename it) was okey and I could delete it. Now
another file was corrupt instead. What is going on here?! I have not
even touched the files since I wrote the mail to the mailing list, so
I'd suppose the same file, which was corrupt in the previous fsck would
be corrupt now as well. Any ideas or comments? Could it be a problem
with Reiserfs?
Best regards,
Tommy
-
Linux-crypto: cryptography in and on the Linux system
Archive: http://mail.nl.linux.org/linux-crypto/
