I am setting up a new system - it has a brand new 4TB drive as parity, a 250GB SSD as cache and and two 2TB drives for data - these drives were recycled from other PCs.  I was able to start setting up my system and I created a share called media.

 

One of the two 2TB drives has been disabled by unRAID.    Ever since then my media (and a couple of other shares) have disappeared and I cannot add a new share as when I try to create one it says "Share media has been deleted".  Why can't I create a share as I still have a parity drive and a working data drive?

 

My system log shows a ton of errors that look like this:

Jan 23 11:19:09 Hoylake kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 6507, rd 2552, flush 1, corrupt 0, gen 0
Jan 23 11:19:09 Hoylake kernel: loop: Write error at byte offset 2427752448, length 4096.
Jan 23 11:19:09 Hoylake kernel: blk_update_request: I/O error, dev loop0, sector 4741704
Jan 23 11:19:09 Hoylake kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 6508, rd 2552, flush 1, corrupt 0, gen 0
Jan 23 11:19:09 Hoylake kernel: loop: Write error at byte offset 2561703936, length 4096.
Jan 23 11:19:09 Hoylake kernel: blk_update_request: I/O error, dev loop0, sector 5003328
Jan 23 11:19:09 Hoylake kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 6509, rd 2552, flush 1, corrupt 0, gen 0
Jan 23 11:19:21 Hoylake kernel: XFS (md1): xfs_log_force: error -5 returned.

 

My diagnostics are attached

hoylake-diagnostics-20170123-1111.zip

Disk1 has dropped offline. Check connections and post a new diagnostic.

 

I don't understand why you are worried about creating shares when you have a disabled disk you need to rebuild. You shouldn't even be trying to write to your server in this state.

I will do that when I get home which won't be for a few hours.  There is no data to speak of on the drives.

 

But shouldn't I be able to use the system with a valid parity disk and a valid data disk, even if one disk is offline?

I will do that when I get home which won't be for a few hours.  There is no data to speak of on the drives.

 

But shouldn't I be able to use the system with a valid parity disk and a valid data disk, even if one disk is offline?

Depends on whether those disks are actually valid or not. Can you read files on the emulated disk?

You also have file system corruption on the emulated Disk 1:

 

Jan 22 21:43:13 Hoylake kernel: XFS (md1): xfs_do_force_shutdown(0x2) called from line 1197 of file fs/xfs/xfs_log.c.  Return address = 0xffffffff8128bb59

Jan 22 21:43:13 Hoylake kernel: XFS (md1): Log I/O Error Detected.  Shutting down filesystem

Jan 22 21:43:13 Hoylake kernel: XFS (md1): Please umount the filesystem and rectify the problem(s)

 

And your docker.img file looks to be corrupted:

 

Jan 22 21:43:14 Hoylake kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 4, rd 0, flush 1, corrupt 0, gen 0

Jan 22 21:43:14 Hoylake kernel: BTRFS info (device loop0): forced readonly

Jan 22 21:43:14 Hoylake kernel: BTRFS warning (device loop0): Skipping commit of aborted transaction.

 

To be honest I haven't really put many files on the disk.  I copied over a few movies last night to the media share before that disappeared.  Right now when I try to do an ls on /mnt I get: 

bin/ls: cannot access 'user 0': Input/output error

bin/ls: cannot access 'disk 1': Input/output error

You also have file system corruption on the emulated Disk 1.

Filesystem corruption often breaks user shares.

 

And I still say you shouldn't even be trying to write to your server in this state.

 

And if you aren't home, how are you accessing your server? If you aren't using a VPN then you are probably being hacked in addition to your other problems.

I run OpenVPN server on my router so I am using that to access my LAN.  My laptop (and mobile devices) have OpenVPN client

I shutdown, checked out the cables and rebooted.  Disk 1 is still disabled.  Here is a new diagnostics.  Should I remove the drive?

hoylake-diagnostics-20170123-1745.zip

The SMART report for that disk (ST32000542AS_5XW1FD76) is not good:

 

=== START OF READ SMART DATA SECTION ===

SMART overall-health self-assessment test result: FAILED!

Drive failure expected in less than 24 hours. SAVE ALL DATA.

 

You need to replace it.

 

Ok, I went to Tools | New Config and built a new config without that drive for now. The parity is now rebuilding which is forecasted to take about another 9 hours.  Was that the right way to remove a drive?

No. Absolutely not. You have just lost any chance of rebuilding the failed disk. I said you need to replace it, not remove it.

 

Here's the correct procedure: https://lime-technology.com/wiki/index.php/Replacing_a_Data_Drive

 

Ok,  - but there was nothing on this disk as the system was just turned on last night.  The only data on the drives was about 5 movies that I transferred over from another drive.

In that case, if you don't care about your data and you're just experimenting, doing the New Config will give you a new configuration with one less data disk. I'm a little nervous of someone else with data to lose stumbling across this thread in the future.

 

Your docker.img is still corrupt. The best way to fix that is to re-create it from scratch and download the containers again: https://lime-technology.com/forum/index.php?topic=48054.msg460524#msg460524

 

I redid my dockers.  How do you tell that they are corrupt?

It was the docker.img file that was corrupt. It's mounted via /dev/loop0 at /var/lib/docker, which was reporting BTRFS errors:

 

Jan 22 21:43:14 Hoylake kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 4, rd 0, flush 1, corrupt 0, gen 0

Jan 22 21:43:14 Hoylake kernel: BTRFS info (device loop0): forced readonly

Jan 22 21:43:14 Hoylake kernel: BTRFS warning (device loop0): Skipping commit of aborted transaction.

 

So it's the docker.img file that needs to be re-made, before re-downloading the containers.