Hello,

 

I've had an unRAID server set up and running as a media server for over a year.  I recently went on a 2 week vacation, and immediately had issues with the server upon return. 

• Server turns on and is accessible through the webUI, but the webUI loads very slowly or sometimes not at all. 
• Some docker webUIs fail to load entirely(Sonarr, Radarr). 
• Some of the dockers seem to be working fine at least some of the time, I see the results of automated scripts in Plex. 
• When I try to access applications like Plex through streaming devices (Roku, Shield, iPad), most of the time, nothing loads. 
• Strangely, if I use a PC and browse directly to the folder containing media files on the server, I can access and play videos without any issues. 
• If I use a PC and try to copy files from my appdata share (which prefers cache disk), the process struggles and intermittently pauses.

 

I updated unRAID version and all dockers to latest versions.  CPU, memory, cache disk space all look OK from the dashboard at least.

 

Any help would be immensely appreciated.

tower-diagnostics-20171205-1958.zip

Edited December 10, 20178 yr by stickman2000

Diagnostics after after rebooting, so not much help, next time try to grab them before, you can type diagnostics on the console if the webGUI is unresponsive.

Sorry about that, to clarify, these issues are continuous.  Everything is bogged down from the moment the server is restarted.  Restarting doesn't seem to have any effect.  There are times when the problem is worse, but it seems completely random.

 

I've attached a new diagnostics file 24 hours or so from my last restart.  I am having issues still with slow webGUI response.

tower-diagnostics-20171206-1911.zip

You have writes errors on your cache device:

 

Dec  5 23:34:04 Tower kernel: sd 3:0:0:0: [sdd] tag#11 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x06
Dec  5 23:34:04 Tower kernel: sd 3:0:0:0: [sdd] tag#11 CDB: opcode=0x2a 2a 00 06 68 c6 b8 00 05 40 00
Dec  5 23:34:04 Tower kernel: blk_update_request: I/O error, dev sdd, sector 107529912
Dec  5 23:34:04 Tower kernel: BTRFS error (device sdd1): bdev /dev/sdd1 errs: wr 3, rd 0, flush 0, corrupt 0, gen 0
Dec  5 23:34:04 Tower kernel: sd 3:0:0:0: [sdd] tag#14 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x06
Dec  5 23:34:04 Tower kernel: sd 3:0:0:0: [sdd] tag#14 CDB: opcode=0x2a 2a 00 07 71 f4 e8 00 08 00 00
Dec  5 23:34:04 Tower kernel: blk_update_request: I/O error, dev sdd, sector 124908776
Dec  5 23:34:04 Tower kernel: BTRFS error (device sdd1): bdev /dev/sdd1 errs: wr 4, rd 0, flush 0, corrupt 0, gen 0

These are hardware errors, start by replacing the SSD cables.

 

Also, your cache is full:

 

Dec  6 03:30:35 Tower shfs/user: cache disk full
Dec  6 03:30:35 Tower shfs/user: cache disk full
Dec  6 03:30:35 Tower shfs/user: cache disk full
Dec  6 03:30:35 Tower shfs/user: cache disk full
Dec  6 03:30:36 Tower shfs/user: cache disk full
Dec  6 03:30:36 Tower shfs/user: cache disk full

This is because your cache filesystem is fully allocated, see here for how to fix:

 

https://forums.lime-technology.com/topic/54533-624-cache-drive-reporting-quotno-space-left-on-diskquot/#comment-532931

 

Johnnie,

 

Thanks, looks like the cache SSD was the issue.  I replaced it with a spare and everything is up and running.  Thanks!