Hi All,

 

Thanks for taking the time to read my post, I am kind of scared that I have lost my data.

 

I have a NAS that I build during the summer which I decided to run unRAID on. Currently, I have an ASRock Motherboard, Intel Core i3 w/ ECC support, 8GB ECC DDR3 RAM, and 3 WD Red 3 TB, plus 1 2TB Seagate drive. One of the WD Red's is my parity drive. I also have a 120GB Samsung 840 EVO SSD set up as a cache drive.

 

I have been running Linuxserver.io/Plex LS.IO/PlexRequests, LS.IO/PlexPy and A Glances Container.

 

I also have been running Sickrage plugin from Phaze, as well as Couchpotato and Transmission plugins.

 

I have the Nerd pack, Community Apps, Preclear disks, and a couple other standard plugins installed as well.

 

My configuration was very stable with 6.1, and I never had any issues once I got all my data directories set up correctly. My sorrows started when I excitedly upgraded to 6.2 last week. At first, everything was fine, and I had no problems. I ran some updates, and after a couple days I noticed my server was unresponsive. Checking into the issue, I found it was completely hung, albeit powered on. I couldn't reach it via SSH, WebGUI, or any of the various services.

 

I rebooted the server manually, and it came back online. I upgraded all my plugins and docker images - thinking it was lack of updates that caused the issues. Then I started a parity check. The parity check yeilded about 166 errors, but I could not find a way to check them or see any logs of these errors. I did some reading here and on Reddit, and the conclusion I came to was that I had some corruption from the loss of power and the partiy check had probably resolved the issue.

 

Last night I queued up a bunch of movies on Transmission and noticed it became unresponsive after a minute. I thought it was odd, but figured I had just flooded it and it would be fine in the morning since everything else was up.

 

This afternoon, the WebGUI was unresponsive and so was Transmission. Plex and other services were fine, as well as my shares were accessible. I tried to reboot via SSH to clear the issue with Transmision, and now I cannot access anything.

 

I booted unRAID in safe mode, and still cannot access the WebGUI or any shares, I saw a stack crash occur on my console, and when I am SSH'd into it, I can now only see 2 disks and my cache drive.

 

I hope that I have not lost any data, but what is the safest way to reinstall unRAID without damaging my current situation any more?

 

Thanks for reading my book!

 

Rhys

 

 

In an effort to provide some more information, as well as ask a question I am going to add this comment.

 

I just booted my unRAID server off a Ubuntu ISO to look at the drives and their smart data.

All drives pass a SMART test, and their various data values all look normal.

 

I found out apparently I did not set up my cache drive correctly, since it still has my old OSX partitions on it. So there is that.

 

I cannot mount the XFS partitions though. When I attempt to, 3 of my drives give the same error saying the filesystem needs to be cleaned.

 

One of the drives (My parity drive? I Hope?) says that it cannot read the superblock.

 

Should I be able to mount the partitions in unbuntu and verify that my data is safe and determine which drive is my parity drive so that I can make sure I reassign things correctly if I have to reinstall unRAID?

I just booted my unRAID server off a Ubuntu ISO to look at the drives and their smart data.

All drives pass a SMART test, and their various data values all look normal.

The one thing you should know is that the data is almost always fine!  There *might* be a bit of corruption, but it should be minimal at most, may not be any data lost.

 

I cannot mount the XFS partitions though. When I attempt to, 3 of my drives give the same error saying the filesystem needs to be cleaned.

Did you read the Additional Upgrade Advice?  There's a section about XFS errors with 6.2, I had some myself, but once you've run the xfs_repair on the drives, all is well again.  If it instructs you to use the -L option, do so.

 

One of the drives (My parity drive? I Hope?) says that it cannot read the superblock.

The parity drive does not have a file system, so never run a file system repair tool on it!  You'll damage parity info.

 

Should I be able to mount the partitions in unbuntu and verify that my data is safe and determine which drive is my parity drive so that I can make sure I reassign things correctly if I have to reinstall unRAID?

You can if you want, although I don't think it's necessary.  You can't mount a file system on the parity drive, (and don't try to), and there's nothing to see on it, as it has nothing but binary parity data.

RobJ,

 

Thanks so much for your advice, I linked this post on Reddit and a user there suggested the same thing.

 

I ran the xfs_repair -v as instructed, and it suggested I run with the -L option since it could not repair.

 

After running with -L, my drive was cleaned and when I started my array all my files and applications are there and working just fine!

 

Thank you so much!