New drive started failing during another new drives rebuild.

[Paravane]

So I got three 3TB Hard drives to replace some older and smaller drives in my unRAID.

 

I put one in and had it added to the array and rebuilt it took roughly 20 hours. Everything was successful, 8 green dots! and all my shares protected.

Cool!

So I do the same process to replace another drive with another 3TB.

 

Said it would take another 20 hours so I get it started then go to bed. Woke up went to work. Came home and then log into unRAID expecting to start the third and last drive but I had errors everywhere and the first 3TB is throwing SMART test failed and the expected build to completion is at this time sitting at 160 Days.

 

With 1 drive being built in as the new one and another failing unRAID won't let me just pull the bad one and put in the other new one. I'm not sure what my options are. There's only 6GB of data on the bad one and it's running at 114KB/s for it's build. I'm not sure what my options are but I would like to swap out the bad drive for the other 3TB and go back to Microcenter and get a replacement drive but unRAID won't let me do that.

 

 

Let me know if you need anymore information from me. I have remote access to my unRAID but not physical at this time. I can get whatever screen shots or logs you need.

[trurl]

8 minutes ago, Paravane said:

Let me know if you need anymore information from me.

Of course we do. Tools - Diagnostics, post complete zip.

 

Most likely you have disturbed a connection. You should always check and recheck all connections any time you go into the case.

 

[JorgeB]

Please post your diagnostics, but if you still have the old disks and they are OK, best course would probably to do a new config and return the array to how it was, if no data was changed you can trust parity but still should do a parity check.

[Paravane]

5 minutes ago, trurl said:

Of course we do. Tools - Diagnostics, post complete zip.

 

Most likely you have disturbed a connection. You should always check and recheck all connections any time you go into the case.

 

I'm downloading the Diagnostics now. It'll probably take a while, but as for the connections it's a case with a back plane. I don't have to open it up to swap drives. The original drive is completely dead it was spun down and being emulated so putting it back wouldn't really do anything. I originally thought it was weird I had 2 failed drives in the same socket so I hooked the new one up to the motherboard directly outside the case and it still shows the same errors it did in the bad socket. So I don't think it's the socket. I'll post the diagnostics as soon as I can.

[Paravane]

Here's the Diag. Thanks for your assistance.

 

lodgecloset-diagnostics-20180911-1016.zip

[JorgeB]

Your new disk6 is failing, same for the parity disk, which also has read errors, if the old disks were also failing there's going to be some data loss.

[Paravane]

1 minute ago, johnnie.black said:

Your new disk6 is failing, same for the parity disk, which also has read errors, if the old disks were also failing there's going to be some data loss.

The parity was next to be replaced since it's still passing SMART tests. Disk 6 is the new disk that is apparently a lemon and it only has roughly 5.5GB of data and I know what that data is and I'm ok losing it. The problem I'm having is that unRAID will not let me start the array without Disk6 since Disk2 is currently building. Idealy I'd like to just delete disk6 and then take the drive back to the store and replace it.

[JorgeB]

You'll need to do a new config with the good disks only and a new parity disk, then let it sync.

 

P.S. Disk7 is also failing.

[trurl]

So which disk is it that you meant by this?

24 minutes ago, Paravane said:

The original drive is completely dead

 

2 minutes ago, johnnie.black said:

You'll need to do a new config with the good disks only and a new parity disk, then let it sync.

 

P.S. Disk7 is also failing.

Which would mean no rebuild of anything so you might have to recover from backups or try to read the original disks.

[Paravane]

I'm aware. I was replacing the worst drives first.  I know my situation is dire. Really didn't expect a new drive to crap itself mid setup.

[JorgeB]

Really didn't expect a new drive to crap itself mid setup.

Even if that disk didn't fail you'd still have data corruption from the read errors on parity disk, and likely disk7, you only have a single parity disk, you can only have one bad disk at a time, not 3, make sure you have notifications enable and you need to act as soon as a single disk is showing issues.

 

 

[trurl]

So which disk is truly dead?

 

Do you have Notifications setup? If you had multiple disks with SMART issues I am guessing that you didn't. unRAID will notify you by email or another agent if any disk has SMART issues.

[trurl]

Also have to ask, do you have backups of any important and irreplaceable files? unRAID parity is not a substitute for backups.

[Paravane]

None of the data is irreplaceable. It'd just be a PITA to re-download all of it. I will definitely have notifications setup by the end of tonight. 

 

What I think I'm going to do it get a 5-6tb external and transfer everything over that I can as the shares are still up and functioning. The only data that should be on the dead disk is some steam cache content. Then just nuke and rebuild the whole thing from the ground up so I can set it up correctly this time. Unless yall have a better idea.