I was hoping you all could help me solve an issue I have been having. My unRAID server, version 5.0.5 Pro is rebooting randomly about once a day. It is plugged into a UPS so I am confident it is not a power issue. This seems to be the common line in my system logs right before it restarts:

 

Dec  3 04:03:47 Tower syslogd 1.4.1: restart

 

Every time after it reboots the server runs a parity check which takes several hours. I've attached the log file(s) from the last two days when it has restarted on me. Any help you all can provide would be greatly appreciated. Also if I need to provide additional information like hardware specifications please let me know!

 

 

syslog-2014-12-03.txt

syslog-2014-12-04.txt

Here is the hardware I am running:

 

SUPERMICRO MBD-X10SL7-F-O uATX Server Motherboard

Kingston 4GB 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) ECC Unbuffered Server Memory Model KVR16LE11L/4

Norco RPC-4220

SeaSonic SS-750KM3 750W

Intel Core i3-4130

(1)WD Red WD40EFRX 4TB 64MB - Parity

(1)WD Red WD40EFRX 4TB 64MB

(1) SAMSUNG_HD203WI 2TB

(1) Seagate ST31500541AS_6XW00GEB 1.5TB

(1) SAMSUNG_HD154U 1.5TB

(1) Seagate ST31000340AS 1TB

(1) Western Digital Black WD1001FALS 1TB - Cache

 

I reattached the log from today so that it is easier to read.

syslog-2014-12-04_1.txt

Run Memtest overnight.

I ran a memory test using a bootable Memtest86+ flash drive and it did not find any errors. Any other thoughts? I have also check all the connections inside my case and verified there are no loose cables. Is there anyway to reinstall unRAID from scratch in case it could be a plugin that is causing issues? I do have plex media server installed.

Start in SAFE-MODE.

Thanks dgaschk. I reverted to a stock configuration of unRAID using your instructions and the server has been running for over 48 hours without incident!

My server stayed online from December 17 and rebooted this morning. Yesterday I logged into the unRAID web interface and disk 1, a WD red 4TB had a red ball next to it. After looking over the logs I found a write error on December 29th which I suspect took the drive offline. I went ahead and stopped the array, unassigned the disk, re-enabled the array, stopped the array again, and then reassigned the disk. It did a parity reconstruction which took over 10 hours.

 

Based on the current log file, the server restarted and came back online this morning on 4:40AM. I checked on the parity reconstruction this morning about 10 and it still had about 2 hours left. That being said, it appears it rebooted BEFORE the parity reconstruction finished. Would unRAID restart for any reason during a parity reconstruction? The server is online now and working with disk 1 running with no issues.

 

Does anyone think these random reboots I am having could be related to a possible failing drive? If so I will RMA the WD Red. Any help would be greatly appreciated!

Was it really a reboot, or just a log rotation?

Never thought about that! Does unRAID set a file size on the log which causes it to rotate periodically?

 

If it was a random reboot this morning, I would think the parity sync would still be running now versus finishing early this afternoon.

It's simple to find out, what does the following command show: uptime

 

It's simple to find out, what does the following command show: uptime

 

Many thanks! It did not restart and has been online for 17 days..