HI Everyone. 

 

I'm asking for help again! 

 

All my docker apps disappeared after a power cut here last week. I followed Squids excellent advise to delete the corrupted Docker image and reinstall the apps. 

 

All worked fine for a week or so until all the docker apps stopped and would not restart. 

 

I deleted the docker image again and started reinstalling the apps but I now get an error when installing. 

 

Oct 9 10:20:22 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/plex (28) No space left on device
Oct 9 10:21:36 UnRaid shfs/user: cache disk full
Oct 9 10:21:36 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/plex (28) No space left on device
Oct 9 10:28:13 UnRaid shfs/user: cache disk full
Oct 9 10:28:13 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/plex (28) No space left on device
Oct 9 10:28:24 UnRaid shfs/user: cache disk full
Oct 9 10:28:24 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/openvpn-as (28) No space left on device
Oct 9 10:28:33 UnRaid shfs/user: cache disk full
Oct 9 10:28:33 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/plex (28) No space left on device
Oct 9 10:29:08 UnRaid shfs/user: cache disk full
Oct 9 10:29:08 UnRaid shfs/user: err: shfs_mkdir: assign_disk: appdata/openvpn-as (28) No space left on device
Oct 9 10:31:49 UnRaid emhttp: shcmd (345): /usr/sbin/hdparm -y /dev/sdg &> /dev/null

 

It states that my Cache disk is full but it is at approx 60% full. 

 

 

I get the attached error when trying to start Plex or open VPN from the docker. 

 

 

Further to this, when the mover runs I get this error for what looks like every file on the array!

 

 

SSH into your host and post the output of:

 

btrfs fi show /mnt/cache

and

 

btrfs fi df /mnt/cache

 

Edited October 9, 20178 yr by MowMdown

Hi, 

 

root@UnRaid:~# btrfs fi show /mnt/cache
Label: none  uuid: ac9204a2-b518-428c-85d0-198999695b79
        Total devices 2 FS bytes used 54.96GiB
        devid    1 size 55.90GiB used 55.90GiB path /dev/sdg1
        devid    2 size 111.79GiB used 55.90GiB path /dev/sdb1

 

 

 

root@UnRaid:~# btrfs fi df /mnt/cache
Data, RAID1: total=54.87GiB, used=54.87GiB
System, RAID1: total=32.00MiB, used=16.00KiB
Metadata, RAID1: total=1.00GiB, used=89.97MiB
GlobalReserve, single: total=70.42MiB, used=0.00B
 

 

Quote

 devid    1 size 55.90GiB used 55.90GiB path /dev/sdg1

This tell me that the cache has been fully allocated ^

 

You'll need to run:

btrfs balance start -dusage=5 /mnt/cache

and once that is finished, run the previous two commands and post the results again.

 

Edit: Here is an explaination on whats happening.

Edited October 9, 20178 yr by MowMdown

Thank you. 

 

root@UnRaid:~# btrfs balance start -dusage=5 /mnt/cache
Done, had to relocate 0 out of 58 chunks
 

root@UnRaid:~# btrfs fi show /mnt/cache
Label: none  uuid: ac9204a2-b518-428c-85d0-198999695b79
        Total devices 2 FS bytes used 54.96GiB
        devid    1 size 55.90GiB used 55.90GiB path /dev/sdg1
        devid    2 size 111.79GiB used 55.90GiB path /dev/sdb1
 

 

root@UnRaid:~# btrfs fi df /mnt/cache
Data, RAID1: total=54.87GiB, used=54.87GiB
System, RAID1: total=32.00MiB, used=16.00KiB
Metadata, RAID1: total=1.00GiB, used=89.97MiB
GlobalReserve, single: total=70.42MiB, used=0.00B
 

 

 

 

2 minutes ago, monkiefone said:

Thank you. 

 

root@UnRaid:~# btrfs balance start -dusage=5 /mnt/cache
Done, had to relocate 0 out of 58 chunks
 

root@UnRaid:~# btrfs fi show /mnt/cache
Label: none  uuid: ac9204a2-b518-428c-85d0-198999695b79
        Total devices 2 FS bytes used 54.96GiB
        devid    1 size 55.90GiB used 55.90GiB path /dev/sdg1
        devid    2 size 111.79GiB used 55.90GiB path /dev/sdb1

 

root@UnRaid:~# btrfs fi df /mnt/cache
Data, RAID1: total=54.87GiB, used=54.87GiB
System, RAID1: total=32.00MiB, used=16.00KiB
Metadata, RAID1: total=1.00GiB, used=89.97MiB
GlobalReserve, single: total=70.42MiB, used=0.00B

 

What size is your cache disk? Not combined. 

Your out of space on your cache, when using different size devices the free space is incorrectly reported, your max usable space is 60GB.

I see whats going on,

 

You're running RAID1 which is mirroring, your smallest disk is a 60GB disk and you've used 55.9GiB which is pretty much 60GB

 

not sure why 90GB is shown but I don't belive that is accurate, it should read 60GB

 

Paging @johnnie.black, he might have a solution to regain your cache. I don't use a cache pool (single 250Gb for me)

 

Edit: Johnnie beat me too it.

Edited October 9, 20178 yr by MowMdown

If you want to keep the raid1 profile free up space or convert to single profile and you can use the space of both combined:

 

 

 

Thank you, everyone. 

 

I have sorted it!

 

 

I have taken the cache pool out of Raid 1 and put it in Raid 0 for now. I've changed where Plex now stores its metadata and I've just ordered a new 1tb SSD as a new cache drive as these are far too small. 

 

 

 

 

Raid0 doubles the available space, but note that you'll only have 120GB usable and the free space will still be incorrectly reported, only the single profile can use the combined 180GB.

 

 

Thank you.

 

I'm only going to run this setup for a couple of days until the new drive shows up. 

 

thank you again for your help.