Hey guys

 

One of my drives keeps putting out errors that look like this:

Jun 22 13:43:37 Tower kernel: REISERFS error (device md3): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [91885 92057 0x0 SD]

 

So I figured I need to run reiserfsck on the drive to either fix the problem or at least find out what it is. Anyway, when I run resierfsck, it tell me that I need to rebuild the superblock. And when I tried doing that it asks me about a Journal device, and I have no idea what that means.

 

So if anyone could guide me through running reiserfsck (or has any other ideas at what to do about my errors), that'd be awesome.

 

Hey guys

 

One of my drives keeps putting out errors that look like this:

Jun 22 13:43:37 Tower kernel: REISERFS error (device md3): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [91885 92057 0x0 SD]

 

So I figured I need to run reiserfsck on the drive to either fix the problem or at least find out what it is. Anyway, when I run resierfsck, it tell me that I need to rebuild the superblock. And when I tried doing that it asks me about a Journal device, and I have no idea what that means.

 

So if anyone could guide me through running reiserfsck (or has any other ideas at what to do about my errors), that'd be awesome.

You are running it incorrectly, and on the wrong device.  Do not rebuild the superblock, as if you are using the wrong device that will only corrupt the disk and you WILL lose the data on it.  (it is roughly equivalent to re-formatting, but using the wrong partition will result in forcing the format where is guaranteed to clobber existing data)

 

See this in the wiki for instructions:

Based on your error, you need to run reiserfsck on /dev/md3

http://lime-technology.com/wiki/index.php/Check_Disk_Filesystems

 

Joe L.

Thanks man, I was trying to run it on the drive by going to it as  /dev/sdd rather than /dev/md3.

 

Cheers.

Thanks man, I was trying to run it on the drive by going to it as  /dev/sdd rather than /dev/md3.

 

Cheers.

that would be VERY bad if you had attempted any corrective action pointing there.  The reiserfsck command error message was correct, the location you gave it did not have the data structure it was looking for.  (the superblock)  You gave it the wrong location.

 

/dev/sdd = the entire physical disk, all blocks/sectors from 1 to the end.    This is the physical drive, anything done here would NOT be reflected in parity, so basically, parity would not be in sync.  No file-system operations ever are performed on the physical device.  (most of the time partition operations, or full disk clones/backups are performed using this device)

 

/dev/sdd1 = the first partition on the disk, typically from sector 63(or 64) to the end. if the disk is under 2.2TB in size.  It > 2.2TB, the partition start is handled by a GPT partition definition.  In any case, the file system is created on this first partition, not on the entire drive.

 

/dev/mdX = the unRAID protected partition of a physical /dev/sdX1 partition.  Fixes here are reflected correctly in parity.

 

99.9% of the time, file-system-repairs are performed on the /dev/mdX devices.  The only time they are not is if the /dev/mdX device was not available for some reason, and then you would need to specify the FIRST PARTITION on the disk, not the entire disk.  Afterwards, you would need to perform a full parity sync, since any repair would not be reflected properly on the parity disk.