From: Manuel Reimer on
Hello,

I tried to restore data from a broken NAS device. This small box
contained a embedded linux system and two 1TB hard drives.

After some tries, I managed to connect both drives via mdadm and was
able to mount this raid as XFS.

Now I connected a brand new eSATA drive to this PC and started to copy
the data (about 700 Gigabytes). Target disk was 1TB, so enough space to
hold the source data.

With my first try (ext2 on target disk) I got the error "Cannot read
inode bitmap". The kernel reported this error for the *target* device
(the brand new SATA hard drive). I did a second try and reformated the
target drive to ext3 and, again, I got the same error. So I installed
"smartmontools" into my running live system and checked the drive. The
new drive has no "reallocated sectors" but several read errors (about 500).

The worst thing about this is, that a diff between source and target
showed many files differing, so I didn't just get errors, but I also
have bad data on the target disk. This is a pretty *unstable* disc
behaviour...

The copy program, I used, was "cp" from busybox.

What should I check to find the reason for this errors? Is it really
possible that a brand new disc, which doesn't have conspicuous smart
values, is the reason for this problem? May the eSATA disc enclosure
(has a small circuit to also offer USB access) be the reason?

The log entries, I could get via dmesg, are available here:
http://pastebin.org/423211

Thank you very much in advance

Yours

Manuel Reimer

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/