ah ok gotcha! that would make sense because the movie share was working before I did the new config but it was missing all the data from disk1, once I did the new config and got back all the data from disk1 the movie share stopped working again!

Once I copy all the data to an external drive how do I format a drive in unraid? I tried to look for instructions in the wiki but couldn't find anything

1 minute ago, sawdustfarmer said:

Once I copy all the data to an external drive how do I format a drive in unraid?

Stop the array, click on that disk and change the filesystem to a different one, start array, format the disk, then stop the array once more and change back to original filesystem and format again.

I'm just about to start the process of backing up disk1.

 

Is it safe to move files with MC from /mnt/disk1 to the free space I have on /mnt/disk2 and /mnt/disk3 ? or will that cause my filesystem issues to spread to disk2 and disk3?

Edited November 26, 20187 yr by sawdustfarmer

It's safe to move from disk to disk, and every disk is a separate filesystem, so not a problem there, you just can't move from disk to share or share to disk.

 

 

great, I don't have enough room to move all the files off disk1 to disk2 and 3 but i'd rather move as much as I can within the server rather than spend hours copying it all to an external then hours copying back to the server

2 hours ago, sawdustfarmer said:

great, I don't have enough room to move all the files off disk1 to disk2 and 3 but i'd rather move as much as I can within the server rather than spend hours copying it all to an external then hours copying back to the server

If you are trying to save time, copying external is likely to be faster for the initial operation. Granted, you will then have to spend more time putting it back, but copying data from disk to disk within the array isn't very fast.

I'm trying to shut down Unraid and I get "Array Stopping•Retry unmounting disk share(s)..." again.

 

This time I made sure PuTTY with MC running was closed and docker was turned off (I dont run VMs) and there was no activity on the disks can the diags show whats stopping it?

 

tower-diagnostics-20181128-0630.zip

Just now, sawdustfarmer said:

can the diags show whats stopping it?

I can only see that the problem is something using disk1:
 

Nov 28 06:29:48 Tower root: umount: /mnt/disk1: target is busy.

 

That's the disk with the corrupt file system I'm trying to copy files off, I'm going to assume that wont be an issue once its replaced.

Is there a way to force kill it, or should l just click Power Down

If you're sure nothing else is open like SSH sitting on /mnt/disk1 then try typing reboot on the console, it should work after the clean shutdown timout, IIRC 60 seconds, but if not you'll need to force it.

4 minutes ago, sawdustfarmer said:

That's the disk with the corrupt file system I'm trying to copy files off, I'm going to assume that wont be an issue once its replaced.

Is there a way to force kill it, or should l just click Power Down

Note that if you have a corrupt file system then rebuilding the disk will not fix that.    You need to run the appropriate file system repair tool to recover from file system corruption.

It rebooted immediately after typing reboot into the web terminal

 

On 11/25/2018 at 7:02 PM, johnnie.black said:

Filesystem has generations errors, likely still mounting because the generation is very close, those type of errors can't be fixed and are usually the result of an unclean shutdown, you should backup that disk (to another one on the array or to an external disk), reformat and restore the data.

itimpi, I'm going to follow the instructions johnnie.black has given me above, I've already tried a new config then resync parity which didn't fix it. 

At the moment I'm just moving files off the corrupt disk onto other disks and then the remaining onto an external before the reformat (I also already have a full copy of the corrupt disk on an external as a 2nd backup as well)

 

On 11/25/2018 at 10:49 PM, johnnie.black said:

Stop the array, click on that disk and change the filesystem to a different one, start array, format the disk, then stop the array once more and change back to original filesystem and format again.

 
I formatted the drive to xfs then back to btrfs and the movie share has come back everything looks to be working again. 

 

I'll now retire the failing 4TB drive add in my new 8TB and copy over the files that I salvaged from the failing 4TB and all should be back to normal

 

Thanks again for all your help johnnie.black and the other members for getting me up and running again, I really appreciate it 

👍

On 11/27/2018 at 3:04 PM, sawdustfarmer said:

I've already tried a new config then resync parity which didn't fix it. 

Curious why you might think rebuilding parity would help fix something on another disk.

9 hours ago, trurl said:

Curious why you might think rebuilding parity would help fix something on another disk.

They were instructions I followed from issues I was having on November 20th. Until now I've thought a rebuilding parity could fix this issue because I would simply remove the failing disk insert a new disk and it would rebuild files from the parity over to a fresh disk, but I understand now that with a corrupt filesystem rebuilding parity isnt the answer. 

I believe this all started because I lost power in my apartment and didn't have a UPS attached to my server, I now know how critical it is to have a UPS on a BTRFS system and to have notifications setup so im alerted ASAP when an issue arises.

The reason I asked the question the way I did is because I thought you were actually talking about rebuilding the parity disk but apparently you were talking about rebuilding the data disk. Resync parity doesn't mean rebuild a data disk it means rebuild the parity disk.