Hey hoping I might be able to get some help here.  

 

I just came home to find my unraid server reading my disk 1 of the array was in a disabled and emulated state with 2500(approximate) read errors. 

I had also heard a number of clicking sounds coming from the compute like the harddrive location).   

 

As I was having to run out again and didn't know how long I was going to take I didn't want to leave my server on for things to get worse.  So I turned off the server before grabbing the diagnostic which I know now is probably the worst I could have done...).  I had grabbed from the Unraid Gui firefox the download smart. But don't know how to get it off before I turned the server off so don't know if it is lost now.   If anyone knows if this can be grabbed and can tell me how to get it I would be happy to provide it. 

 

When I turn the server back on I now get the disk as disabled and emulated.  When I try to run SMART on it it clicks a crapton so I stop it, if it is felt needed I again can let it run fully, and provide. 

 

I have attached my diagnostics to know if there is anything else I need to know or do with the server, or anything else I should be worried about/ address

 

I do have 2 parity drives. and the drive is under warrenty so I am thinking I might just replace it and rebuild array, but just want to check with the community any other thoughts. 

 

I appreciate the help 

rizznetunraid-diagnostics-20190513-1634.zip

4 minutes ago, Aceriz said:

so I am thinking I might just replace it and rebuild array

 

Sound like what I would do.  This sound like the old 'click of death' that has around ever since the first hard drives were being used.  Your only real decision is whether (1) to wait for the RMA replacement disk to arrive or (2) purchase a new disk to do the rebuilt immediately and use that RMA return as a spare for when the next disk fails.   (Or you need more capacity...)

3 hours ago, Frank1940 said:

 

Sound like what I would do.  This sound like the old 'click of death' that has around ever since the first hard drives were being used.  Your only real decision is whether (1) to wait for the RMA replacement disk to arrive or (2) purchase a new disk to do the rebuilt immediately and use that RMA return as a spare for when the next disk fails.   (Or you need more capacity...)

Ya my next challenge  is to figure out if there is a way to easily port my data from cache drives Two ssd's over to nvme drives to free up two more data slots for more drives if I can do that I might just get new drive  and when rma comes In i can just add to array.  

 

But hopefully  if anyone has any thoughts on diagnostic they can catch any other lurching  errors 

I would like to suggest another alternative.  Get a LSI SATA expansion card.  Cost would be less than $100US on E-bay.  You can find more details here:

 

      https://forums.unraid.net/topic/69018-sata-controller-replacement-question-and-advice/?tab=comments#comment-630097

 

 

If you use  LSI  IT Mode   as the search target, you will find numerous LSI cards.   Just be sure to vet the vendor as some of the new cards from China are counterfeit.  Used cards come servers removed from server farms that are being are upgraded with new servers.  

17 hours ago, Frank1940 said:

I would like to suggest another alternative.  Get a LSI SATA expansion card.  Cost would be less than $100US on E-bay.  You can find more details here:

 

      https://forums.unraid.net/topic/69018-sata-controller-replacement-question-and-advice/?tab=comments#comment-630097

 

 

If you use  LSI  IT Mode   as the search target, you will find numerous LSI cards.   Just be sure to vet the vendor as some of the new cards from China are counterfeit.  Used cards come servers removed from server farms that are being are upgraded with new servers.  

That is interesting option actually  noob question. Do you know can I keep the cards I ha e on my mother boards on board sata connectors and just add the lsi card  or similar  with new drives to it and have both working in array properly. 

Yes.  The card ports are in addition to the existing ports on your MB.  If you have six ports on your MB and you add an LSI card with eights ports on it, you will have a total of 14 SATA ports available.  (You do have to be careful adding SSD drives that plug into your MB as some of them will take away the use of some of the MB SATA ports.  Read your MB manual to see how the addition of MB mounted drives are handled.) 

Edited May 14, 20197 yr by Frank1940

Since you rebooted before getting diagnostics we don't know why the disk was disabled, but looking at SMART attributes for the disabled disk the only thing I notice is 16 reallocated, which might not be fatal. Are you sure it is that disk making the noise? I don't know if a "click of death" disk would give a SMART report or not, I haven't had one in many years.

14 hours ago, trurl said:

Since you rebooted before getting diagnostics we don't know why the disk was disabled, but looking at SMART attributes for the disabled disk the only thing I notice is 16 reallocated, which might not be fatal. Are you sure it is that disk making the noise? I don't know if a "click of death" disk would give a SMART report or not, I haven't had one in many years.

Ya I wish I had taken the diagnostics first.  Live and learn for the future... When I try to run the SMART report on the disk after now it pauses then makes a large amount of clicking high pitched sounds, more than just spinning and he read head moving. . 

 

14 hours ago, trurl said:

Since you rebooted before getting diagnostics we don't know why the disk was disabled, but looking at SMART attributes for the disabled disk the only thing I notice is 16 reallocated, which might not be fatal. Are you sure it is that disk making the noise? I don't know if a "click of death" disk would give a SMART report or not, I haven't had one in many years.

The Other problem I have is now when I boot up the drive does not read in terms of the main terminal page of unraid. It says device disabled, emulated. not reading or writing to it..  If anyone knows how to restart it I would be open to trying that. or other ideas. 

Just now, Aceriz said:

The Other problem I have is now when I boot up the drive does not read in terms of the main terminal page of unraid. It says device disabled, emulated. not reading or writing to it..  If anyone knows how to restart it I would be open to trying that. or other ideas. 

It is disabled. That is what Unraid always does when a write to a disk fails. After disabling a disk, all the rest of the array is used to emulate that disk from the parity calculation, for both reading and writing. The physical disk won't be used again until it is rebuilt or replaced.

 

Since you have dual parity, you are still protected until another disk goes down, but you don't want to leave it like this since all the disks are now working hard to emulate the disabled disk.

 

Replacing a disk is always the safer approach, even if you know the disk is good. That way you still have the original disk as a backup in case there is a problem rebuilding to a new disk. But of course, that approach does require a spare disk so many people elect to rebuild onto the same disk if they have reason to believe the disk is OK and something else, like a bad connection or cable, was the actual cause of the problem. And more often than not, a bad connection or cable is the actual cause of the problem.

 

But in your case, since you have reason to believe the disk is bad, replacement really is the best idea. Then when you get your array stable again you can decide whether or not that disk can still be used for anything or it should be trashed.

2 hours ago, trurl said:

It is disabled. That is what Unraid always does when a write to a disk fails. After disabling a disk, all the rest of the array is used to emulate that disk from the parity calculation, for both reading and writing. The physical disk won't be used again until it is rebuilt or replaced.

 

Since you have dual parity, you are still protected until another disk goes down, but you don't want to leave it like this since all the disks are now working hard to emulate the disabled disk.

 

Replacing a disk is always the safer approach, even if you know the disk is good. That way you still have the original disk as a backup in case there is a problem rebuilding to a new disk. But of course, that approach does require a spare disk so many people elect to rebuild onto the same disk if they have reason to believe the disk is OK and something else, like a bad connection or cable, was the actual cause of the problem. And more often than not, a bad connection or cable is the actual cause of the problem.

 

But in your case, since you have reason to believe the disk is bad, replacement really is the best idea. Then when you get your array stable again you can decide whether or not that disk can still be used for anything or it should be trashed.

Hey Thanks for explaining that to me. That actually really helps me understand whats going on.  So I have my RMA drive on route to me. Again I know this is a stupid question but obviously  I have not had to go through with this, but once I get my RMA drive back my plan was to preclear the new RMA drive to test for any defects in it.  Then to use the RMA drive to rebuild the array.  I order to rebuild to disk how exactly do I go about doing that. Is the option under settings?  I have my server just turned off right now till i get new drive back so that I can have less stress on the other drives as you mentioned. 

Just stop the array and assign the replacement disk to the same slot as the disk you are replacing, then start the array to begin rebuilding.

 

If for some reason it offers to format anything, DON'T.