DISK_DSBL repeated on two drives, rebuilding fails

[jasperjames]

Unraid Version: 5.0-rc8a

 

Hi all,

 

I have been trying to fix my server for some weeks intermittently and now feeling a bit lost. I have done a lot of searching and head scratching but now would really  appreciate some support!

 

My array is currently four drives including parity. Recently disk 2 (a 2tb fairly new WD Red) became disabled (red icon) showing DISK_DSBL. Tried a rebuild but that failed, not having much time to investigate or dig further I plumped for ordering a new WD Green 2tb to replace and do a simple rebuild.

 

Once installed in the Red's palce, the rebuild was happily working away for 3 days on the WD Green until that failed too and now also displays DISK_DSBL and has left me scratching my head.

 

Logic would suggest to me this is not a coincidence, happening to both WD drives.

 

On attempting a 2nd rebuild on the Green today...

 

- I moved the hard drive further up the cage away from another drive, worrying about heat

 

- Replaced the SATA cable with new one, as suggested elsewhere

 

- Changed the power connector for another but on the same trunk (the only spare) of the Corsair PSU (which I assume would not make much difference if there is an issue with the PSU).

 

- Observed that drive 2 is on a dual SATA controller card along with another drive but not witnessing any errors with that drive.

 

The 2nd rebuild on the WD Green lasted less than an hour this time and failed again.

 

I would really appreciate if anyone could lend a hand and help figure this out. My hope is that both the Red and Green drives may be salvageable, and could expand my array, or at least save me forking out for a third drive?

 

I attach the most recent syslog and smart report on the Green WD below.

 

Many thanks,

 

Jasper

smartctl_sda.txt

[jasperjames]

Syslog was too big to attach so have cut in half!

syslog_pt1.txt.zip

[jasperjames]

Syslog part 2

syslog_pt2.txt.zip

[lime111]

This is not of immediate help, but if the disks are both bad, i would return them to where you got them for a refund or replacement, depending on the consumer protection rules on where you live of course.  I use crystaldiskinfo (and even provided its reports as proof of disk failure) to substantiate the failure to the retailer.  I've done this on disks less than 12 months old successfully. 

 

 

[itimpi]

Just a thought - it may be possible to recover data from the original failed 2TB WD Red drie by running Reiserfsck against it?    Do you have a spare SATA port where you can plug it in?  If so try running the command

reiserfsck --check /dev/sd?1

against it where ? corresponds to the device ID that has been allocated to it this boot. 

 

You should then report back with the results of that check for some guidance on where to go next.

 

I know it is not much help now, but if you get a drive disabled it is always a good idea if at all possible to remove the 'failed' drive rom the array and rebuild onto a different physical drive.  If the rebuild succeeds great, but if not then you have the original disk in the state it was at the point where it was disabled and one can normally recover virtually all the data if the drive has not physically failed.

[jasperjames]

This is not of immediate help, but if the disks are both bad, i would return them to where you got them for a refund or replacement, depending on the consumer protection rules on where you live of course.  I use crystaldiskinfo (and even provided its reports as proof of disk failure) to substantiate the failure to the retailer.  I've done this on disks less than 12 months old successfully.

 

Thanks Lime, I've used Western Digital's RMA system before for the Red originally, will go down that path again if needs be!

 

[jasperjames]

Just a thought - it may be possible to recover data from the original failed 2TB WD Red drie by running Reiserfsck against it?    Do you have a spare SATA port where you can plug it in?  If so try running the command

reiserfsck --check /dev/sd?1

against it where ? corresponds to the device ID that has been allocated to it this boot. 

 

You should then report back with the results of that check for some guidance on where to go next.

 

I know it is not much help now, but if you get a drive disabled it is always a good idea if at all possible to remove the 'failed' drive rom the array and rebuild onto a different physical drive.  If the rebuild succeeds great, but if not then you have the original disk in the state it was at the point where it was disabled and one can normally recover virtually all the data if the drive has not physically failed.

 

Thanks itimpi,

 

First thought, will running Reiserfsck work if I've already started a rebuild on the drive? As instructed by the support wiki, both the red and green have attempted a typical rebuild.

 

I will give that a go shortly and report back, thank you again.