My array began going offline periodically a few days ago. Guessing I might have gotten some corruption, I have tried copying the flash files from the SAN disk to another directory, downloading latest files, reformatting the flash drive, then copying config folder back. Array started, all drives & shares in place and config valid. Then over night, the array stopped again. I am attaching the diagnostics zip file.

 

Can anyone provide some help on why this is happening? Array was running fine for months. Could the Cruzer be going bad? I noticed that it wasn't recognized by the bios a couple of times when trying to reboot, but that seemed to stop after I reformatted it and reloaded the Unraid 6 files.

 

Last things I did to the array:

* Added the Crucial SSD as a cache device and move some system files there. Not getting any errors on the device that I can see.

* Created a "Linux Mint" VM. The VM started ok, but the array stopped by itself with the Unclean Shutdown Error, so I removed that VM in case it was related.

 

Thanks in advance for any help.

 

Motherboard: Asrock 970 Extreme4

Processor: AMD FX6120 Processor (6 cores x 3.5GHZ)

Memory: Crucial Ballistic Sport 8GB (2 x4GB) 240 Pin 1600 MHz DDR3

                      Kingston 16GB (2 x 8GB) 240 Pin 1600 MHz DDR3

Realtek Gigabit LAN (R8168-Driver Version 8.037.00-NAPI)

Unraid 6 Flash Drive ID: 0781-5571-0002-000704122413

UNRAID version 6.2.4 with PLUS License

 

HARD DISK DRIVES:

Device Type Address Serial # Manufacturer Model Size Speed

 

BOOT Flash Drive San Disk Cruzer 8GB

PARITY SATA  HDD JK1131YAH3W89V Hitachi HUA722020ALA330 2TB 3GB/sec

DISK 1 SATA  HDD JK1130YAGRUN7T Hitachi HUA722020ALA330 2TB 3GB/sec

DISK 2 SATA  HDD JK1131YAGGLVXV Hitachi HUA722020ALA330 2TB 3GB/sec

Disk 3 SATA  HDD YGHTJV8A Hitachi HUA723020ALA641 2TB 6GB/sec

Disk 4 SATA  HDD W1E3Y8ZF Seagate ST2000DM001-1CH164 2TB 6GB/sec

Cache

SSD 14310CCECA30 Crucial FCCT256MX100SSD1 256GB 6GB/sec

Disk 5 SATA HDD YGHE19UA Hitachi HUA723020AL641 2TB 6GB/sec

tower-diagnostics-20161205-0746.zip

Unclean shutdown typically means a hard powerdown, such as holding the power button until it powers off, unplugging your server, power loss, or power supply issue. You can see the uptime in the upper right of the webUI, and your syslog indicates your server has only been up for a few hours. If you are unaware that it is losing power, perhaps you have your BIOS configured to automatically start up again after a power outage.

Another possibility is problems with the CPU fan so that the system is powering itself off to prevent damage to the CPU due to it overheating.

Itimpi and trurl. Thanks for the ideas.

I have checked the bios and it is set to "Power Off" after power failure. I am now monitoring the array uptime in powered on but not started state to see if there is an unrecognized power failure which is resetting the counter.

 

I am also checking the CPU & motherboard temp using the Dynamix System Temp Plugin. Currently they are sitting at 82 & 72 degrees F. respectively, but as I said the array is not started so no work being done. My plan is to run it this way for a day or so and see if the uptime counter resets. If not, I will start array and monitor again.

 

Another observation I see is that in the logs I see the following message right before the unclean shutdown. It is consistently there when the array goes offline. Could the license key somehow be corrupted? Does anyone know if it is normal to "detect" the licence key periodically when the array is online?

 

Dec  5 04:25:37 Tower emhttp: Plus key detected, GUID: 0..3 FILE: /boot/config/Plus.key

Dec  5 04:25:37 Tower emhttp: unclean shutdown detected

 

The way it detects an unclean shutdown is by reading config/super.dat on the flash. This file contains your disk assignments as well as the running/stopped state of the array. If super.dat says it was running when it boots, it is an unclean shutdown because it should be stopped before powerdown. Possibly it reads the key file before reading super.dat.

I'm not sure why this solved the issue, but I noticed that the "Clear CMOS" lite was lit on the back of the tower. I performed the reset, re-entered the correct BIOS settings, and the array has been running fine.

 

I'm posting this to assist anyone who has experienced this issue.

 

Thanks for the help.

• • I'm having trouble with my Unraid Server going offline intermitantly with "unclean shutdown detected" error. I had this problem a month ago, but it went away for a few weeks after reseting the cmos and reloading the bios settings. Now it is back. Very frustrating.
     
     
    Motherboard: Asrock 970 Extreme4
    Processor: AMD FX6120 Processor (6 cores x 3.5GHZ)
    Memory: Crucial Ballistic Sport 8GB (2 x4GB) 240 Pin 1600 MHz DDR3
                            Kingston 16GB (2 x 8GB) 240 Pin 1600 MHz DDR3
    Realtek Gigabit LAN (R8168-Driver Version 8.037.00-NAPI)
    Unraid 6 Flash Drive ID: 0781-5571-0001-310923120232
    UNRAID version 6.2.4 with PLUS License
     
    Symptoms:
    Server hangs and gui goes away. Sometimes the gui comes back and the server is always "offline" with the "unclean shutdown " error showing .
    The server NEVER powers down when this happens, and I have bios set to "power off" after power failure.
    Sometimes the gui doesn't come back and server needs to be rebooted. It always reboots with no issue.
    Parity check always starts, and most times I cancel it. "Parity Valid" message when array is started. Few times I've done parity check, it is always good.
     
     
     
     
    Things I've Tried This time:
    -- Replaced Flash Drive with new one. Downloaded latest Unraid files and only copied config folder from original flash drive. New license key installed also due to new ID of flash.
    -- Replaced cmos battery - Sometimes when it fails the "Reset CMOS" light is on so I replaced the battery.
     
    I have attached the syslog as a word doc with the multiple times that "unclean shutdown detected" error occurs highlighted in yellow.
    Also attaching the syslog file created by tower as well.
     
    Any help is appreciated.
     
    

 

Unraid_Syslog_Jan_16_15.doc

tower-syslog-20170116-1518.zip

You posted this last in V5 support. I moved it and merged it with your previous thread since it had some posts that are probably still relevant. What is the exact model of your power supply?

Thanks for organizing the thread. I'm kinda new at this. The power supply is KDM-X8875C. 875 Watt ATX Supply. I included a picture of the specs.