[smartmontools-support] bad blocks on crypt partition

[scar]

I'm trying to repair a disk with a bad sector that is waiting to be
re-written. I've run the extended offline test and it reports a bad LBA
at 1872148287. this is the same sector that libata reports in the
system log:
end_request: I/O error, dev sde, sector 1872148287

At the same time, the system log also reports some I/O errors on logical
blocks:
Buffer I/O error on device sde, logical block 234018535
Buffer I/O error on device sde, logical block 234018536
Buffer I/O error on device sde, logical block 234018537
Buffer I/O error on device sde, logical block 234018538
Buffer I/O error on device sde, logical block 234018539

The errors happen when i start the bitcoin client, so i think it's
related to the large dataset for the blockchain. However i tried
following the BadBlockHowto and the block number i came up with using
the formula was 234018279 for example. However this might not be
significant because all of these blocks are in the same inode anyways,
18095158. When i use "ncheck 18095158" with debugfs on the decrypted
filesystem (/dev/mapper/sde1_crypt) it tells me the file occupying this
inode (and it is not related to bitcoin).

I'm able to copy and calculate the md5sum of this file just fine, and
there aren't any errors in the syslog when i do this. So I guess there
is something different to figure out when the underlying partition is
encrypted. More details:

# fdisk -lu /dev/sde

Disk /dev/sde: 1000.2 GB, 1000204886016 bytes
81 heads, 63 sectors/track, 382818 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x0005fd19

Device Boot Start End Blocks Id System
/dev/sde1 2048 1953525167 976761560 83 Linux
#

/dev/sde1 is decrypted to /dev/mapper/sde1_crypt:
# tune2fs -l /dev/mapper/sde1_crypt | grep Block
Block count: 244189878
Block size: 4096
Blocks per group: 32768
#


I was thinking i would just try "hdparm --write-sector 1872148287
/dev/sde" instead but wanted to check here first

Thanks

[Carlos E. R.]

Post by scar
I'm trying to repair a disk with a bad sector that is waiting to be
re-written. I've run the extended offline test and it reports a bad LBA
at 1872148287. this is the same sector that libata reports in the

...

Post by scar
I was thinking i would just try "hdparm --write-sector 1872148287
/dev/sde" instead but wanted to check here first

Just a wild guess of mine: run a badblocks read test of the entire disk.