January 4, 20179 yr Hello guys, New unRAID user here and first time poster. I’ve been using unRAID for about 6 months without a hitch. A few days ago the server started experiencing a problem where I wake up in the morning and find the array stopped with the main tab on the webGUI showing “Stopped - Unclean shutdown detected”. I guess this must be because the sever is randomly rebooting itself. When I look at the syslog I can see the logging starting at a specific time that I guess must be the time where the reboot occurred. I left memtest running overnight for about 15+ hs and could not find any errors with the RAM that could trigger a system reboot. I would like to totally discard the software side of things before start changing hardware parts. A have looked around the forums but have been unable to find an answer to the following question, is there a way to capture the syslog before the reboot occurs? I'm attaching the syslog/diagnostics file. System Specs: (All the hardware is brand new except for the PSU and Storage Drivers) unRAID: version 6.2.4 Plus licence CPU: Core i3-6100 MB: MSI B150M Mortar RAM: Kingston HyperX FURY 16GB 2133MHz DDR4 PSU: Zalman ZM600-HP 600W Cache Drive: SanDisk X400 512G Parity Drive: WD RED 3TB Storage Drives: WD RED 3TB, WD BLUE 1TB UPS: APC 1200VA BR1200GI Installed plugins: -Community Applications -Dynamix System Temperature -Fix Common Problems -Nerd Tools -Open Files -Preclear Disks -Tips and Tweaks -Unassigned Devices Thanks in advance for any tips/guidance you can provide David tower-diagnostics-20170104-1852.zip
January 4, 20179 yr Your best bet would be to go to Fix Common Problems, hit troubleshooting mode, and then upload the last diagnostics file it generates and possibly the tail of the syslog that's all stored on the flash drive after the next reboot.
January 5, 20179 yr I checked the spec of your power supply and found that it has four independent 12 volt rails. I notice you don't have very many hard disks in your server but the recommendation is always to use a power supply with a single (and therefore pretty hefty!) 12 volt rail.
January 5, 20179 yr Author Thanks for the quick replies guys. John, the PSU is my prime suspect as well. Even though the system has pretty low power requirements, the PSU is quite old (around 8 years) and before putting it into the system it was sitting on a shelf for about 4 years. Probably a dried out electrolytic capacitor or cracked solder is producing unstable voltage/excessive ripple . I will borrow the PSU form my desktop system and put it into the server for a few days and see how it behaves.
January 5, 20179 yr There's a thread dedicated to power supplies. You might want to take a look. https://lime-technology.com/forum/index.php?topic=12219.0
January 11, 20179 yr Author After borrowing the PSU from the desktop PC and putting in on the server for a couple of days the reboot problem went away. I decided then to buy and new PSU for the server and ended up getting a Seasonic SS-460FL2. What a difference a modern and more efficient PSU makes. With the previous PSU, the server consumption with all disk spun down would idle at around 25watts. Now that figure went down to 15watts. I'm marking the post as "Solved". Thanks for the advice provided!
Archived
This topic is now archived and is closed to further replies.