Jump to content

upgraded to version 6.1.9 and now disk 3 is missing from array


rob71399

Recommended Posts

Hello,

This evening I saw that version 6.1.9 was available and decided to upgrade.  Everything went smoothly until tower rebooted.  Disk 3 could not be added to the array.  I started the array without it.  As soon as i hit the start button I was thinking I made a mistake.  Now my 3rd data disk is no longer a part of the array and I cannot figure out how to get it back.  Any help would be appreciated.  I have attached my syslog.

 

thanks

r

tower-syslog-20160305-1937.zip

Link to comment

Post a screenshot of Main.

 

Syslog shows disk3 missing so it is probably being emulated. Can you read the files on the emulated disk?

 

Maybe a bad connection somewhere to disk3, or maybe it has died. Check SATA and power connections to disk3.

 

Do you have another disk you can rebuild to?

Link to comment

I checked the cables and everything looks good.  The disk is being emulated and I can see the files.  Unfortunately, I do not have a spare drive.  I will have to order one ASAP.  For the time being should I try and copy the emulated files to another location?  I have room on another data disk

 

Link to comment

SMART for all other drives looks OK. However, none of your drives are protected by parity now since all drives are required, so if there were another failure, you would loose both drives. If you copy the emulated drive's files to another array disk, you actually would be reading all the drives to get the emulated data, then writing the other drive and writing the parity drive so it will be a lot of activity by all disks and they are no longer protected.

 

I would probably stop the array and wait until I had tested a new disk with preclear and then do the rebuild. If you feel you cannot afford to be without your system that long, then you might just continue to use it normally. The other disks will read/write normally and only when you try to read/write the emulated drive will it be necessary to get all the disks involved.

 

Do you have backups? If there are any critical files on the emulated disk then I would copy them to another system rather than copying them to another disk on the server.

Link to comment

I know it seems like a waist but I would get 2 drives and have a cold pre-cleared spare on hand for the next time you have a drive problem.  I had the same problem myself and I ended up turning off my server until I got another drive and pre-cleared it (to test it) before I turned my server back on and replaced the emulated drive.  Since then I try to keep a cold spare on hand the size of my largest drive just in case.

Link to comment

if all you did was start the array with disk 3 missing. you can stop the array and create a new config and put back all the drives with parity in its place then start the array with parity already valid. everything should go back to normal.

 

btw im assuming parity drive is valid and all disks have no errors including disk 3.

 

Link to comment

if all you did was start the array with disk 3 missing. you can stop the array and create a new config and put back all the drives with parity in its place then start the array with parity already valid. everything should go back to normal.

 

btw im assuming parity drive is valid and all disks have no errors including disk 3.

 

That's not what has happened Furby....  Drive 3 is there but it has failed and is being emulated.  Generally not the best time to mess around with a new config unless you're really really sure you know what you're doing.

 

Personally I'd do what trurl and Bob have suggested.  And I'd also echo Bob's sentiments of keeping a cold spare.  (Precleared and ready to go)

Link to comment

That's not what has happened Furby....  Drive 3 is there but it has failed and is being emulated.  Generally not the best time to mess around with a new config unless you're really really sure you know what you're doing.

 

Personally I'd do what trurl and Bob have suggested.  And I'd also echo Bob's sentiments of keeping a cold spare.  (Precleared and ready to go)

 

im a bit reckless with my data hahaha

i had a failing psu before and would get red balls every other week

but i knew my drives were solid and in good working condition

so i kept doing new config for weeks until i was able to squeeze a new psu in the budget

havent had a problem since the new psu been installed

 

so i put that option out there just in case he likes to live on the edge :P

Link to comment

That's not what has happened Furby....  Drive 3 is there but it has failed and is being emulated.  Generally not the best time to mess around with a new config unless you're really really sure you know what you're doing.

 

Personally I'd do what trurl and Bob have suggested.  And I'd also echo Bob's sentiments of keeping a cold spare.  (Precleared and ready to go)

 

im a bit reckless with my data hahaha

 

i had a failing psu before and would get red balls every other week

but i knew my drives were solid and in good working condition

so i kept doing new config for weeks until i was able to squeeze a new psu in the budget

havent had a problem since the new psu been installed

 

so i put that option out there just in case he likes to live on the edge :P

 

That's your perogative Furby, but I don't think you should assume that position with someone else's data, and you've just said that you knew your drives were solid and the PSU was the cause, which isn't necessarily the case with the OP's problem either.

 

This is a great place to get advice from the community but as a rule whenever anyone's data is at risk, I personally will not reply to the thread unless I'm 110% sure that what I'm posting as advice for them is correct and will not worsen their chances of data recovery.

 

For all we know Rob may have all his most treasured photos of his family on that disk...

 

I know that important data should be backed up, and mine is, but can't assume everyone has followed this rule....

Link to comment

Thanks to everyone who has posted.  I ended up doing what trurl recommended last night.  I shut down tower and ordered a new drive.  It should be here on Monday.  I  know i should have ordered 2 but don't have the funds at this time to do that.  I do have backups of all of my important data and thought about trying what Furby mentioned but I decided to stay the course.  With my luck I would end up really screwing something up.  My plan is to put the new drive in and let it rebuild from parity (after doing a preclear).  Is there a way to test the faulty drive at this point? 

 

 

Link to comment

It is possible that the drive in question is actually fine!

 

When you have managed to get your system back to a good condition (presumably using the new drive) then you can check out the old drive:

  • Check the SMART attributes of the problem drive to look for obvious problems
  • If SMART looks OK put the drive through a preclear cycle to see if that looks OK
  • Assuming preclear went OK and SMART still looks good keep the drive as a precleared spare.

Link to comment

My new drive finally arrived yesterday.  When I powered up tower disk 3 was unassigned but this time I was able to chose it from the dropdown list.  I spent some time comparing my backup to the contents of disk 3, which was still being emulated.  All of my important content (i.e. photos, home video, etc) was backed up so I decided to add disk 3 back to the array.  The contents were rebuilt without error.  Everything still looks good this morning so I am going to leave it running.  I am going to preclear the recently purchased disk so I can be ready to swap out if any of my disks fail in the future.  I don't quite understand why this happened and why the disk appeared again but I will take the results.  And now I have a spare data disk :)

 

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...