November 27, 20169 yr I recently upgraded to v6.2.4. Immediately after the restart Plex stopped working. I've also noticed that TimeMachine on my iMac (which backs up to my Unraid) is complaining that the server is full. My recollection is that it started complaining prior to installing v6.2.4 Looking at Plex logs, it seems to think the array cache is full. Array cache is has 22GB used and 92GB free. This persists even though I have run Mover several times to try to empty the cache. I've rebooted the server a couple of times and reinstalled Plex, without any success. I suspect an underlying problem in Unraid causing Plex and TimeMachine to think the volumes are full. Diagnostics file is attached. I'd be grateful for any assistance. Happy to pay for someone to log in remotely and just fix it. Thanks Shane tower-diagnostics-20161127-1930.zip
November 27, 20169 yr Your minimum cache free space is set to 100Gig on your global share settings. So any write any write to anything referenced by /mnt/user/cacheOnlyShare is going to be bounced back as disk full. ie: Any write to a cache-only share will fail with 92gig available
November 27, 20169 yr Author I reduced the 100GB to 10GB and it's all working ok. Still need to know why Mover isn't emptying the cache. Perhaps that will just fix itself (he says hopefully) Thank you
November 27, 20169 yr Community Expert You have 25 user shares and most of these are set to not use cache. Many have names that begin with d and end with a number. I suspect you have misconfigured some application so that it is writing top level folders. Any top level folder on cache or any array disk is automatically a user share.
November 27, 20169 yr Author Looking in the web gui I can only see 7 User Shares. The only Apps I have installed are PlexMedia Server, Digikam (turned ok) and Zoneminder (turned off). Could it be a Plugin causing the problem ? The only Plugins are produced by Andrew Zawadzki and Bergware. I'm lost here. Happy to pay for some remote support to just fix this.
November 27, 20169 yr Community Expert From the command line, what do you get with this? ls -lah /mnt/user
November 28, 20169 yr Author root@Tower:~# ls -lah /mnt/user total 11G drwxrwxrwx 1 nobody users 74 Nov 28 04:40 ./ drwxr-xr-x 14 root root 280 Nov 27 20:16 ../ -rw-rw-rw- 1 shane users 4.0K Aug 28 09:26 ._Media drwxrwxrwx 1 nobody users 6 Sep 4 11:23 Backup/ drwxrwxrwx 1 shane users 22 Nov 7 19:19 Media/ drwxrwxrwx 1 shane users 30 Nov 27 19:46 Private/ drwxrwxrwx 1 nobody users 110 Nov 28 15:59 TimeMachine/ drwxrwxrwx 1 nobody users 30 Aug 29 21:58 appdata/ -rw-rw-rw- 1 nobody users 10G Sep 16 08:45 docker.img drwxrwxrwx 1 nobody users 0 Sep 16 19:43 domains/ drwxrwxrwx 1 nobody users 14 Sep 16 19:43 system/
November 28, 20169 yr I reduced the 100GB to 10GB and it's all working ok. Still need to know why Mover isn't emptying the cache. Perhaps that will just fix itself (he says hopefully) Thank you Normaly, the Plex-Database is also stored on the Cache-Drive. This could explain your 22GB on the SSD.
November 28, 20169 yr Community Expert The diagnostics you posted earlier has 25 .cfg files in the shares folder. Perhaps these are just old files that are no longer used.
Archived
This topic is now archived and is closed to further replies.