How do I try to recover the contents of the cache disk?  I think it's mostly empty, but does (did?) have my appdata info stored on it so I'd like to recover if possible.

 

syslog.zip

What was your cache drive formatted as? This happened to me awhile back and if it is reiserfs you can use reiserfsck tools, which fixed my cache drive. If brtfs, you can attempt you use its version to repair what ever is corrupted but I didn't have any luck when my cache was formatted as brtfs. Although, I didn't have a drive shown up as a red ball so take my advice with a grain of salt.

I converted to btrfs with beta6 to use docker.  It's not red-balled (that's a different drive )

 

Once I finish 'fixing' the red-balled drive, I'll look into btrfs repair tools.

 

I may just decide that it's likely less work to just wipe the cache drive, and start over with the dockers, but I'm not there yet.

 

One hassle at a time.  :'(

I converted to btrfs with beta6 to use docker.

Check to make sure the format type on the cache drive is set to BTRFS. If it's set to Reiserfs, a BTRFS drive would indeed be unformatted.

 

With all the little (and big) issues you have been having lately, is it possible your flash drive is corrupt / dying?

I just realized the picture i uploaded had cut off the formatting section, not sure why.

 

But, yes, it still shows btrfs.  I just checked it, and it's definitely not doing well (see attached).

 

I had tried to copy a file in windows, and it failed, then I realized it was probably trying to copy to the cache.  It apparently tried really, really hard 

Any suggestions on how to resolve this one?

 

I would like to get my cache drive back into action, but I'm not sure if there are any good tools to 'fix' a btrfs formatted drive.

 

Assuming there is a tool to try to repair this, how do I run it?

 

thanks!

nothing?  no way to fix?

 

I just need to pull it, pre-clear it, then put it back into the server and start over?

From a shell, can you see any data under /mnt/cache?

no.  I've removed it in the GUI, thinking that might help the array work better.  it didn't.

 

I could try adding the drive back, but I don't think I could see it before either.

I added the drive back.  No, nothing under mnt...

 

root@media:/mnt# ls -l
total 0
drwxrwxrwx 10 nobody users 232 Sep 21 08:25 disk1/
drwxrwxrwx 10 nobody users 256 Sep  5 14:18 disk10/
drwxrwxrwx 10 nobody users 232 Jun 16 15:08 disk2/
drwxrwxrwx 10 nobody users 232 Jun 16 15:08 disk3/
drwxrwxrwx  4 nobody users  80 Sep 27 09:44 disk4/
drwxrwxrwx 10 nobody users 232 Sep 23 14:03 disk5/
drwxrwxrwx  9 nobody users 208 Aug 31 16:45 disk6/
drwxrwxrwx  9 nobody users 200 Sep 23 14:06 disk7/
drwxrwxrwx 10 nobody users 232 Jun 16 15:08 disk8/
drwxrwxrwx  1 nobody users 232 Sep 27 09:44 user/

If you don't have it assigned to the cache in the array config it wont mount.  You can either put it back in the config, or mount it manul similar to the discussion about disk9..

If you don't have it assigned to the cache in the array config it wont mount.  You can either put it back in the config, or mount it manul similar to the discussion about disk9..

 

sorry, that's what I meant, I assigned it to the cache in the array/config when i stopped the array to test mounting disk9.  When i restarted the array, the disk is listed for the cache, but shows unformatted.

 

It's still not visible under mnt 

You may want to do the same, and stop the array and mount the first partition on the cache disk to see what data is on it.  You could ether copy it off and/or let it format it again.

Ah, I mistyped and mounted sdg1 to cachedisk, so now disk9 is mounted to 2 locations. 

 

root@media:/mnt/somedisk# mkdir /mnt/cachedisk
root@media:/mnt/somedisk# mount /dev/sdg1 /mnt/cachedisk

 

It should have been sdh1 

 

What's the command to unmount cachedisk?

 

 

just do a "umount /mnt/cachedisk"

that was my first guess, but it didn't work...

 

root@media:/mnt# umount /mnt/cachedisk
umount: /mnt/cachedisk: not mounted
root@media:/mnt# ls
cachedisk/  disk1/  disk10/  disk2/  disk3/  disk4/  disk5/  disk6/  disk7/  disk8/  somedisk/  user/

 

try a "df -k" and see what is mounted.

 

or you can "cat /etc/mtab"

ARGH!!!  I'm so mad at myself for making such a stupid mistake and not typing sdh1 

 

df -k doesn't show it...

 

root@media:/mnt# df -k
Filesystem       1K-blocks        Used  Available Use% Mounted on
/dev/sda1          3781376      107376    3674000   3% /boot
/dev/md1        2930177100  2926575788    3601312 100% /mnt/disk1
/dev/md2        2930177100  2925919324    4257776 100% /mnt/disk2
/dev/md3        2930177100  2920092756   10084344 100% /mnt/disk3
/dev/md4        1953454928       32840 1953422088   1% /mnt/disk4
/dev/md5        2930177100  2397599420  532577680  82% /mnt/disk5
/dev/md6         976732736   831994796  144737940  86% /mnt/disk6
/dev/md7         976732736   969651932    7080804 100% /mnt/disk7
/dev/md8        2930177100  2920310264    9866836 100% /mnt/disk8
/dev/md10       3906899292  3905021648    1877644 100% /mnt/disk10
shfs           22464705192 19797198768 2667506424  89% /mnt/user
/dev/sdg1       2930177100  2204406704  725770396  76% /mnt/somedisk

 

but MC does (see screenshot)

 

I'm going to try umount somedisk, and see if that clears both of them, and if not, I suppose I have to restart unRAID, and mount them again, but with more caution!

MC is only seeing the directory, but the disk isn't mounted on it.  I don't think the mount worked when you mounted it..

I tried this...

 

root@media:/mnt# mount /dev/sdh1 /mnt/cachedisk
mount: wrong fs type, bad option, bad superblock on /dev/sdh1,
       missing codepage or helper program, or other error
       In some cases useful info is found in syslog - try
       dmesg | tail  or so

Might need a -t btrfs on the mount to specify the fs type

then I tried this...

 

root@media:/mnt# mount -t btrfs /dev/sdh1 /mnt/cachedisk
mount: wrong fs type, bad option, bad superblock on /dev/sdh1,
       missing codepage or helper program, or other error
       In some cases useful info is found in syslog - try
       dmesg | tail  or so

root@media:/mnt# dmesg | tail
REISERFS (device sdg1): checking transaction log (sdg1)
REISERFS (device sdg1): Using r5 hash to sort names
BTRFS info (device sdh1): disk space caching is enabled
parent transid verify failed on 350928896 wanted 183324 found 183322
parent transid verify failed on 350928896 wanted 183324 found 183322
BTRFS: open_ctree failed
BTRFS info (device sdh1): disk space caching is enabled
parent transid verify failed on 350928896 wanted 183324 found 183322
parent transid verify failed on 350928896 wanted 183324 found 183322
BTRFS: open_ctree failed

 

it looks like a disk problem; but I know so little about any of this, that's more of a stab in the dark

I guess you could try a btrfs check on the cache device to see if there are any errors in the btrfs filesystem.  I am assuming your cache disk was btrfs?  I think by default it is reiserfs.

 

 

yeah, it's btrfs, due to beta6 and docker

 

it the command just btrfs check, or do I need to add more to that to specify which disk?

 

(I'm googling now, but perhaps you already know)

I think it is just btrfs check {device|mount point}