4 days ago I come home and and my unraid was completely unresponsive so I had to restart the machine.  After starting the array, parity check kicked off and by the time it completed, it showed around 928k in corrected errors.  When it completed, all ran fine but I noticed when I tried to copy some files for backup, I was getting errors.  Last night I tried to watch a movie that I could not copy and it was corrupt.  Then I ran across couple others and I got worried about the array.  I started to see what was common and it all came down to one drive on which a lot of files were corrupt.  So I have stopped the array, started it in maintenance mode and ran reiserfsck --check /dev/md14.  When that finished, I got the following error:

Bad nodes were found, Semantic pass skipped

10 found corruptions can be fixed only when running with --rebuild-tree

 

So after that I ran  reiserfsck --rebuild-tree /dev/md14 and it stopped with the following error:

The problem has occurred looks like a hardware problem (perhaps memory). Send us the bug report only if the second run dies at the same place with the same block number.

 

mark_block_used: (149743376) used already

Aborted

 

So after that I ran this command reiserfsck --rebuild-tree -S /dev/md14  and got the following error after it ran for 14 hours:

bread: Cannot read the block (3304755201): (Invalid argument).

Segmentation fault

 

Also, This line showed up in syslog

May 19 22:26:26 Tower kernel: reiserfsck[21127]: segfault at c ip 0805bf21 sp bfc12370 error 4 in reiserfsck[8048000+4a000] (Errors)

 

Of course it has to be one of only two 2tb drives in my array that had around 1.7tb in data.  When I start the array it shows Unformatted.  My linux knowledge is almost nill and I am completely at loss in what should I do at this point  :'(  At this point i just wish I could retrieve list of files that were on the drive to see if I have their backups. 

 

Thanks.

Or, as reiserfsck suggested as a possibility, you might have a problem with your RAM.

 

Have you performed a memory test?  Is the clock speed, timing, and voltage set correctly for your specific memory strips ? (Some BIOS do it correctly, some do not, some just get it wrong)

 

Unless memory is correct, you'll have all kinds of issues.

 

Joe L.

unformatted simply indicates the drive could not be mounted.  DO NOT format the disk.  It would erase any files on it.

 

You must fix the file system corruption... but first, verify memory is correctly configured.  The segfault is NOT normal... it is one clue that memory might be an issue.

 

Joe L.

I ran memory test and it passed - no errors...  Should I try to remove the drive from this box and run reiserfck on another box? 

I ran memory test and it passed - no errors...  Should I try to remove the drive from this box and run reiserfck on another box?

You ran memtest86 from unRAID boot menu for 24hours and it passed every cycle?

It ran 4 full cycles.  Didn't run it for 24 hours.  I will kick it off tonight to run for that long.

I had to stop the memory check (feel safer with with array off with all these potential tornadoes in the area) at little bit over 21 hours with 22 passes 0 errors.  Any advice what I should do next?  I have another unraid box available and it has different controller - should I try to run it on that machine?  Thanks!

Not from me anyway.  I just noticed you didn't say how long you ran it for earlier so made my comment.

Post SMART reports.

Here is SMART report.  Hope its what you looking for.  Thanks.

smart.txt

Hi,

 

Is there any command switch that I could use to write reiserfsck log to a file on the flash drive to so that I could go through all files that are lost so that I know what I need to replace?  I assume that I will not be able to get the data back from the drive so i just need to assess damage done.  Thanks for your help.

There a several recovery programs available in Windows.