Hi!

I ran the parity swap procedure over the weekend to exchange a 250GB drive to a 12TB drive, and reuse the 4TB parity drive as a data drive. 

Everything went smooth until this morning I came to step #15. Under Array operation had the "Start" button appeared again after the parity copy was done. I clicked "Start" but it didn't start the array like it says in the manual (and it was no "Yes I want to do this" checkbox either). Everything just stalled.

 

Current state is this:

 

"Array Stopped•/usr/local/emhttp/plugins/user.scripts/backgroundScript.sh /tmp/user.scripts/tmpScripts/Enable write cache for all disks by-id/script >/dev/null 2>&1"

 

I checked the script for write cache and entered the new 12TB drives serial number (Edit: and removed the 250GB serial number). But I'm not able to run the script?

The script is this from the thread (SOLVED) Set write cache permanently? (with my drives S/N):

 

I've had the script running perfect for almost a year, because I had some problem that every time I rebooted the server write cache was disabled and I had to run hdparm (if I remember correctly, should really keep personal logg...).

 

 

I haven't rebooted yet, don't want to redo the parity copy if not necessary.

Any ideas of what I should try now?

Syslog attached.

 

 

tower-syslog-20201207-1100.zip

Edited December 8, 20205 yr by kim_sv
Changed title

Anyone?

Your title says 

Quote

copy of data drive failed

Parity swap copies parity, but it rebuilds the data disk as mentioned in that wiki. Your screenshot suggests the data disk hasn't been rebuilt yet.

 

That syslog shows parity copy completed and then it seems to show array started and disks mounted but then nothing in syslog after that. Is there anything new in syslog after the syslog you attached?

 

I have no experience with anything that needed to have write-cache enabled so I don't know if that is related in any way or not.

 

In any case, if you still have the original disk2 your data should be OK.

 

You can probably force it to rebuild disk2 with New Config / Trust Parity then stop/unassign disk2/start/stop/reassign disk2/start to begin rebuild.

 

You might wait to see of there are other opinions on what happened or what to do.

 

Might be useful to have the complete diagnostics instead of just syslog.

Well, I lost my cool and hit reboot.

 

First I couldn't even connect or ssh to it, but then I remembered I have an unassigned drive connected and maybe it tried booting of that. Removed the external USB-drive and then it booted normal. Since the write-cache script runs on startup and I now had changed it I guess everything was ok so unraid started the Data-Rebuild automatically. Will run a parity check after to check if everything still is ok.  

 

20 minutes ago, trurl said:

Your title says 

Quote

copy of data drive failed

Parity swap copies parity, but it rebuilds the data disk as mentioned in that wiki. Your screenshot suggests the data disk hasn't been rebuilt yet.

 

That syslog shows parity copy completed and then it seems to show array started and disks mounted but then nothing in syslog after that. Is there anything new in syslog after the syslog you attached?

You are correct, I will try and change the title.

 

21 minutes ago, trurl said:

I have no experience with anything that needed to have write-cache enabled so I don't know if that is related in any way or not.

It is really strange but since the script worked I haven't looked into it more, should I? In this case I guess the problem wasn't that the new drive had write-cache problem but that the script included a removed drive?

 

OK, the data rebuild went without a glitch. It wasn't a problem to restart before the data rebuild, at least not this time. I also ran an extra parity check to be sure. So now everything is working. I guess it was the script that messed it up and I'll have to look into that more.