I am having problems with the GUI becoming unresponsive.  I am still able to telnet into the system and send commands, but the system doesn't typically respond to the commands.  I have tried to stop running dockers, kill running dockers and shutdown from a telnet session but the server continues to stay up and running with no response.

 

I have attached an export of the diagnostic log from the last time this happened.  Can anyone help out with troubleshooting to determine why my system is locking up and becoming unresponsive?

rebel_base-diagnostics-20170201-1928.zip

Your server had an Out Of Memory error:

 

Feb  1 05:11:27 Rebel_Base kernel: Plex Script Hos invoked oom-killer: gfp_mask=0x2400840, order=0, oom_score_adj=0

Feb  1 05:11:27 Rebel_Base kernel: Plex Script Hos cpuset=c3e706411203ee1a09a69f255cc72f4e5a459bc5326d75409139281dc5fe626a mems_allowed=0

Feb  1 05:11:27 Rebel_Base kernel: CPU: 3 PID: 8204 Comm: Plex Script Hos Not tainted 4.4.30-unRAID #2

Feb  1 05:11:27 Rebel_Base kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./H77 Pro4-M, BIOS P2.00 08/06/2013

Feb  1 05:11:27 Rebel_Base kernel: 0000000000000000 ffff88013591b320 ffffffff8136f79f ffff88013591b4f8

Feb  1 05:11:27 Rebel_Base kernel: ffffffff8177d510 ffff88013591b390 ffffffff8110857d 0000000000000206

Feb  1 05:11:27 Rebel_Base kernel: 0000000000000206 ffff88013591b370 ffffffff810a936b ffff88021f5f2b30

Feb  1 05:11:27 Rebel_Base kernel: Call Trace:

Feb  1 05:11:27 Rebel_Base kernel: [<ffffffff8136f79f>] dump_stack+0x61/0x7e

Feb  1 05:11:27 Rebel_Base kernel: [<ffffffff8110857d>] dump_header+0x55/0x1d2

 

unRAID doesn't use virtual memory so you might like to look at how much RAM your applications are using and consider whether you need to add more. You might be able to free some up by reducing the amount reserved for the integrated graphics in the BIOS - it only need the minimum amount. You currently have 7685 MB available to unRAID.

 

I noticed a couple of other things.

 

Unassigned disk ST3000DM001-1CH166_W1F30S73 doesn't have SMART reporting enabled. You can turn it on with:

 

smartctl --smart=on /dev/sdc

 

Your disks are repeatedly spinning up and down:

 

Jan 30 13:42:48 Rebel_Base kernel: mdcmd (770): spindown 0

Jan 30 13:44:52 Rebel_Base kernel: mdcmd (771): spindown 2

Jan 30 14:13:13 Rebel_Base kernel: mdcmd (772): spindown 0

Jan 30 14:13:13 Rebel_Base kernel: mdcmd (773): spindown 2

Jan 30 14:42:48 Rebel_Base kernel: mdcmd (774): spindown 0

Jan 30 14:42:49 Rebel_Base kernel: mdcmd (775): spindown 2

 

You might want to increase the delay before they spin down.

 

Thanks for the feedback! I will take a look and see what I can change.  I have 8GB of RAM currently so it may be time to upgrade.