unRAID constant restarts

[fonzie]

my unraid server has been having constant restarts on it's own without me prompting them.  When it restarts, it says there was an unclean shutdown and begins a parity check.

 

My Fix Common Problems is showing two things of note:

 

- Call traces found on your server: It asked me to run a memtest so I ran one pass and it came back with no errors.

 

and also

 

irq 16: nobody cared found on your server

I've attached the diagnostics out.  I'm not sure what the problem is and any help would be greatly appreciated.

media-diagnostics-20170319-2018.zip

[burtjr]

In my experience restarts such as these are caused by a power supply that is either going or overloaded, but can be other items also. I would start by shutting down all dockers and VM's and see if that brings stability to your system. If it does bring one back at a time to test for stability until you find the culprit.

 

As for the power supply, what type and voltage is it, and the 12V rail amperage? Do you have any maintenance scheduled to occur that would spin up your disks (mover ect.)prior to the computer restarting?

 

Looking through your smart reports you have around 3 disks that have reported errors. I haven't spent time educating myself on these yet, but they look like they happened a while ago.

 

I have know idea on the call trace, and nobody cared found on your server means.

[BobPhoenix]

Your USB controller and your MVSAS controller are sharing an interrupt and is causing one of your call traces.  If I could read logs better I might be able to tell you which device you should try moving to a new port to stop that call trace.  I think it might be this device:Hewlett-Packard v165w Stick - but I could be wrong.  I also see some of your disks are still ReiserFS drives.  More people are reporting weird problems with ReiserFS drives and you should probably convert them if your server can be made to stay up long enough.  One thing to do to help that out is to prevent all writes to your ReiserFS drives while you convert them.  XFS is now the default format and what I would recommend you changing too.  Another possibility is your Super MIcro controllers that use a Marvel chipset along with the Marvel 9230 controller on your MB (I can't use mine - Marvel 9230 - because it continually drops drives).  Also many have been reporting problems with Marvel controllers on the Intel CPU platform.  Many on the AMD platform are still working well with Super Micro controllers based on posts on this forum anyway.

 

So if no one else has better suggestions

1. You should switch from ReiserFS disks to XFS.
2. Switch to LSI based controller(s) in place of your SuperMicro Marvel based controller(s).
3. Put no devices on the Marvel 9230 controller on your MB. 

Those are my suggestions but I do not know which suggestion would help your reboot problems.  They are just suggestions I've seen posted to other users with problems.  So maybe someone else will have better ones for you.  Hope these suggestions are helpful.