Hi All,

 

I was testing my unRAID system this afternoon and one of my data drives went offline due to a read error.

 

My data drive and parity drive is 2TB. I have 4TB replacement drives on hand.

 

I obviously can't just replace the data drive with the 4TB since my parity drive is smaller. Do I upgrade my parity drive first, resync, then replace the failed data drive?

 

Thanks.

Do I upgrade my parity drive first, resync, then replace the failed data drive?

You can't do that without losing the data from the failed (or disabled) disk.

 

It may be there isn't anything wrong with the disk itself, in which case it would be simplest to rebuild to the same disk.

 

Go to Tools - Diagnostics and post the complete diagnostics zip. That will answer a lot of our questions before we decide how to proceed.

It may be there isn't anything wrong with the disk itself, in which case it would be simplest to rebuild to the same disk.

 

There were read/write errors - bad sectors on the disk; disk still works but is on it's last legs.

 

In this case, I would have to replace the failed data drive with an equivalent 2TB, rebuild, then upgrade the parity with a 4TB drive?

 

Thanks.

I was testing my unRAID system this afternoon and one of my data drives went offline due to a read error.

And unRAID won't disable a disk for a read error. Only for a write error. It is possible that a read error would cause it to try to write the correct data from the parity calculation back to the disk, and that may have failed, causing the disk to be disabled.

There were read/write errors - bad sectors on the disk.

Where did you look to see that it actually said there were bad sectors on the disk? Bad connections are more common than bad disks.

 

Please do what I asked.

Go to Tools - Diagnostics and post the complete diagnostics zip. That will answer a lot of our questions before we decide how to proceed.

There were read/write errors - bad sectors on the disk.

Where did you look to see that it actually said there were bad sectors on the disk? Bad connections are more common than bad disks.

 

Please do what I asked.

Go to Tools - Diagnostics and post the complete diagnostics zip. That will answer a lot of our questions before we decide how to proceed.

 

Sorry to be clear it was a write, not a read, as I was copying a large amount of files to the array when the drive died.

 

I already rebooted the machine, so I assume the diagnostics are lost. The data was not critical as I am evaluating unRAID.

 

There were pending sectors showing up on the drive. It's a 7 year old 1TB Green drive that I'm using to test out different scenarios with unRAID so I am not surprised there were bad sectors on it.

 

 

Hi All,

 

I was testing my unRAID system this afternoon and one of my data drives went offline due to a read error.

 

My data drive and parity drive is 2TB. I have 4TB replacement drives on hand.

 

I obviously can't just replace the data drive with the 4TB since my parity drive is smaller. Do I upgrade my parity drive first, resync, then replace the failed data drive?

 

Thanks.

 

I asked the same question and got a quick answer indicating no problem.

 

http://lime-technology.com/forum/index.php?topic=55422.0

I asked the same question and got a quick answer indicating no problem.

 

http://lime-technology.com/forum/index.php?topic=55422.0

 

 

Thanks for the link - this is my exact same scenario. I was in the process of trying that ... and for some reason, this step was not available:

 

Go to the Main -> Array Operation section
You should now have a Copy button, with a statement indicating "Copy will copy the parity information to the new parity disk".

 

I did not have a copy button - not sure why.

 

I rebuilt the array; unfortunately I can't try again - maybe in VM

It's the other way around, it only works with a failed drive, make sure you are following the instructions.

I would have pointed you to the parity swap but I was trying to make sure all your drives were good enough first. I've been dealing with a lot of threads lately where people had let themselves get to a place where they had multiple disks with problems.