Hi,

 

My unraid server has been restarting randomly. Sometimes it will go days/weeks, but occasionally it will only go hours between a restart. I haven't had time to diagnose it until recently, but am very new to this so don't really know a lot of steps to try. So far nothing has fixed it. I do not have a UPS (one is currently on the way), but I don't think that is the problem due to the number of times that the server restarts, and also that other desktops at my house don't restart. Any thoughts on what else to try?

 

One other point - I recently installed the crashplan and plex media dockers. The restarts were happening even before install, so I don't think that's the problem, but still wonder if it is making it happen more often somehow.

 

Thanks in advance for any help.

johnny-diagnostics-20161116-0624.zip

syslog.txt

It is always worth checking all your fans.  In particular the CPU fan as that can cause the server to shut down unexpectedly if the CPU overheats.

I am going to ask a question here. 

 

First, what do you mean by "restart"?    You mention that you thought it might be the result of a short (possibly very) short power outage.  If this happens, your server would normally be in an 'off' state when you discovered it.  (Having it restart automatically when power comes back can be a bad option after you install a UPS!)

 

A list of Hardware in your server might be useful.  Particularly information on the PS.  You should also double check that all of your cooling fans are running and that the cooling fins on your CPU cooler are not filled with dust and dirt.  (Don't blow this  stuff out in the house.  You would be surprised how much crap accumulates inside a case in a few months.)

 

After checking this out, get the UPS installed and see if the problem is still there.  PS differ in their ability to bridge very short power interruptions (less than a fraction of a second).  And PS do age and may not be as rugged as when new.

 

 

OK, thanks to you both for the replies. I checked the fan and there was some dust on the outside of the case, but very little on the CPU fan. Cleaned everything out anyway. there are air filters on a few of the air intakes, so its possible this was restricting the airflow. Still, I didn't see a lot of dust and things looked pretty clean. I also ran a memtest and it made it all the way through 1 pass with no issues.

 

I do have the tower setup to auto restart in the event of power failure. Do I need to disable this now, or just when I install the UPS? Why is it bad to have it auto-restart when connected to a UPS? Just wondering. I set it up to auto-restart so I (or my wife when I'm not home) wasn't also having to access to turn it back on. it isn't super easy to access.

 

I have a corsair power supply in there, I'm pretty sure i purchased the one that was recommended for an unraid box on the forum when I purchased everything. Everything is almost exactly 1 year old - 11/17/15 purchase date.  Here is the full hardware list

 

• ASRock Z97 Extreme4 LGA 1150 Intel Z97 HDMI SATA 6Gb/s USB 3.0 ATX Intel Motherboard
  
• CORSAIR RMx Series RM650X 650W 80 PLUS GOLD Haswell Ready Full Modular ATX12V & EPS12V SLI and Crossfire Ready Power Supply
  
• Intel Pentium G3258 Haswell Dual-Core 3.2 GHz LGA 1150 53W BX80646G3258 Desktop Processor Intel HD Graphics
  
• G.SKILL Ripjaws Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model F3-12800CL9D-8GBRL
  
• Fractal Design Define R4 Blackout with Window Silent ATX Mid Tower Computer Case
  
• Seagate NAS HDD ST4000VN000 4TB 64MB Cache SATA 6.0Gb/s Internal Hard Drive
  
• Seagate Desktop HDD ST4000DM000 4TB 64MB Cache SATA 6.0Gb/s 3.5" Internal Hard Drive Bare Drive
  
• I also have 2 older 2TB HDDs in there, they are showing as green and not reporting any errors
   
  

 

UPS is on the way - more to come when it arrives and I get it installed. Thanks again to you both for your help!

 

I do have the tower setup to auto restart in the event of power failure. Do I need to disable this now, or just when I install the UPS? Why is it bad to have it auto-restart when connected to a UPS? Just wondering. I set it up to auto-restart so I (or my wife when I'm not home) wasn't also having to access to turn it back on. it isn't super easy to access.

 

UPS is on the way - more to come when it arrives and I get it installed. Thanks again to you both for your help!

 

The problem is that the the shutdown process will use up a goodly amount of the battery reserve when it shuts the server down the first time.  If the power is restored and the server automatically restarts and then a second outage occurs with seconds or minutes (Not an unusual condition in many situations), the battery will not have recharged enough to do a second shutdown.  (Four to six hours is about the minimum recharge time!) You will then have an unclean shutdown which will require a Parity rebuild and perhaps even file corruption in the worst case scenario.

 

You mention that it is difficult to access the server.  Is it in a confined area where heat build-up is possibility?  (Most CPU's automatically shutdown when they get too hot!)

Gotcha, that makes sense. I'll be sure the change the configuration when I install the UPS. Thanks for the advice.

 

The server is under the bed in our guest room. Likely not the best place for airflow, but not shut up in a cabinet or anything. I make sure all the air intakes don't have anything next to them. Just looking at the disk temperatures it is running a little cooler than it was before I cleaned everything out, so perhaps that was the problem. I remember seeing disk temps in the high 80s or low 90s, now I see them in the low 80s. I've never gotten any warnings for high disk temperatures and have the high disk temperature set at the default of 113. Is there an area where I can check CPU temp within unraid? It would be good to know definitively if it is running hot.

WOW!!!  I have never seen a disk temperature above about 40 degrees--- Centigrade as that is the way they are reported. 

 

Here is the temperatures specs for WD Red Drives:

 

Temperature (Metric) Operating 	-0° C to 70° C
Temperature (Metric) Non-operating 	-40° C to 70° C

Ha. I guess I changed my settings to Fahrenheit at some point. The HDDs certainly aren't boiling!

Is there an area where I can check CPU temp within unraid? It would be good to know definitively if it is running hot.

 

Try the    System Temp   Plugin.  If you can get the right probe setup, it can provide with the CPU temperatures.  You can find it here:

 

    https://lime-technology.com/forum/index.php?topic=36543.0

 

     

OK, so my UPS arrived and I installed it yesterday. I turned off the autorestart feature in the BIOS, meaning that if the power is interrupted it will stay shut down and won't boot back up when the power comes back on. I woke up this AM and had a notification that there was an unclean shutdown at 1am last night, the server had rebooted and was 75% of the way through a parity check. Long story short, even with the UPS I'm still having the problem. I'm a little confused why it restarted instead of powering down since I changed the BIOS setting. I will go in and make sure the BIOS setting saved, but since I'm already most of the way through a parity check, I figure I'll until it finishes before rebooting the server.

 

Here is what I've tried so far:

 

1- Ensured unraid OS and all plugins are up to date and running the latest stable versions

2- Turned off both docker apps I have running (crashplan and plexmediaserver): I'd had restarts before installing these apps, but turned these off just in case to try it.

3 - Installed fix common problems app: I posted the log files after a recent restart above, let me know if it would be helpful to have the most recent one and I'll pull and attach it.

4 - ran a full cycle memtest: no issues

5 - vacuumed out case and made sure all fans are spinning

6 - installed system temp plugin and have been monitoring system temperatures for both HDDs and CPU. I have yet to see a temperature over 100F for HDDs and 106F is the highest I've seen for CPU. I've checked them pretty regularly including when the CPU was performing some intensive operations. At this moment both HDDs and CPU are running at ~80F (for the HDDs that are not spun down

7 - installed and turned on UPS

 

Any ideas on what additional steps I should take, or any of these steps I should go back and take an additional look at? I'm happy to provide any additional detail, just let me know.

 

Thanks for the continued support.