I have had an Unraid system up for a while now, predominantly file storage and Plex (which is having issues I will post in another thread).  

Xeon E3 1245v3  16GB ram, 8TB x 3 disk (16TB and parity) and a 240GB SSD cache drive

Unraid 7.6.2 - recently updated - not sure if that was the start of my issues

 

I decided to just relook at my whole system.  So with the Plex docker not running, my CPU is varying from 30-60% over all load.  That seems high to me for a server that is just sitting there...   Is that normal?

 

The dashboard comes up and looks ok.  But when I go to Plugins, the page loads but the "spring" just keeps moving and it never returns. I had some of the usual applications installed, and was going to list them for this thread, but ...

 

Any ideas where to go from here?

 

 

Thanks,

 

Ross

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your next post.

Diagnostics attached.   I was hoping to fix anything with Unraid and then move on to Plex.  Storage and Plex are the primary uses of this server.

 

The current issue started last week, Plex interface looked fine, but when I tried to play anything or use Live Tv DVR, the hour glass would not stop.  As if there was a transcoding issue.  Was working fine for the last year or so.

 

Diagnostics attached.

 

Thanks for any help.  

wserver1-diagnostics-20190727-0257.zip

11 hours ago, Ross said:

That seems high to me for a server that is just sitting there...   Is that normal?

At the time the diags were taken there's a parity check going on, so moderate CPU usage is expected.

True, but what’s odd is that I’ve run parity checks weekly for the last year and they take 15 hours, speed 146 MB/sec

 

This one currently reports:

Elapsed time: 2 days, 23 hours, 20 minutes
Current position: 1.47 TB (18.3 %)
Estimated speed: 6.1 MB/sec
Estimated finish: 12 days, 8 hours, 39 minutes
Sync errors corrected: 0

 

Why would the speed be so low?  Server not being used for much else.  

 

Thanks for any advice.

 

Ross

Your syslog has a lot of this.

Jul 26 16:55:12 WServer1 kernel: EDAC MC0: 1 UE UE overwrote CE on any memory ( page:0x0 offset:0x0 grain:8)
### [PREVIOUS LINE REPEATED 2 TIMES] ###
Jul 26 17:50:00 WServer1 kernel: EDAC MC0: 1 UE ie31200 UE on mc#0csrow#2channel#0 (csrow:2 channel:0 page:0x0 offset:0x0 grain:8)
Jul 26 18:05:59 WServer1 kernel: EDAC MC0: 1 UE UE overwrote CE on any memory ( page:0x0 offset:0x0 grain:8)
### [PREVIOUS LINE REPEATED 1 TIMES] ###
Jul 26 18:44:06 WServer1 kernel: EDAC MC0: 1 UE ie31200 UE on mc#0csrow#2channel#0 (csrow:2 channel:0 page:0x0 offset:0x0 grain:8)
Jul 26 21:36:57 WServer1 kernel: EDAC MC0: 1 UE UE overwrote CE on any memory ( page:0x0 offset:0x0 grain:8)

Memory?

 

 

So does that imply bad memory?  I have 16GB installed, which sounds like it should be ok, although I have thought of going to 32. 

 

I could shutdown and reseat current memory to see if that changes anything...

3 hours ago, Ross said:

So does that imply bad memory

No implication.   Memory errors are happening

I have reset the memory, swapped slots, and it appear stable for now.

 

I am surprised that if there was an ECC memory error that there was no notification system for that, either on the dashboard or email.   Maybe not enough of us use server ECC memory to be worth the trouble.  

 

Does anyone know of a way to add a monitor for ECC memory errors?

 

Thanks for your help!

 

Ross