June 4Jun 4 Hello all, I am sorry if this question has already been asked and answered multiple times. I am experiencing an incredibly frustrating issue where my Unraid server keeps suffering hard reboots. I have my array configured to not automatically start when the server is up. Typically, I log in headlessly to my server from my main computer and start the array. A few hours later, when I log back in, I can see that my server has hard rebooted because the reported uptime is recent (such as only being up for 30 minutes even though the last time I logged in and started my array was 4 hours ago). When I go to the main tab, Unraid tells me that there was an unclean shutdown detected. The only Docker containers I have on my server are Redis, PostGRE, and Immich. I am primarily using this server as a photo backup alternative to iCloud for my iPhone.My server specs:HP ProDesk 600 G4 Desktop Mini PCi5-8500T CPU12 GiB DDR4HP 83EF Motherboard, Version KBC Version 07.D5.00HP BIOS, Version Q22 Ver. 02.33.00BIOS dated: Fri 13 Feb 2026 12:00 AMCache is a Toshiba KBG30ZMV256G 256Gig NVMe SSDArray consists of a 2TB Seagate Barracuda Compute 7200RPM HDD and a 4TB SEAGATE EXOS Enterprise ST4000NM0035 7200RPM HDD, both of which are attached to a USB 3.0 CENMATE 2-bay hard drive enclosure.Here are things I have tried to do to fix my server:Enabled mirror logs to flash drive. After each reboot, I went through the logs myself and asked Gemini. Gemini and me were both unable to find any serious errors that could cause a reboot. It appears that the server reboots out of nowhere.Disabled runtime power management and extended idle power states in my HP BIOS.Originally, the mini PC (90W power supply) and the 2-bay enclosure were plugged into daisy-chained power strips. I moved both the power supply for my mini PC and the power supply for my 2-bay enclosure to a single power strip plugged directly into a wall outlet to rule out delivery issues. I tested two different power strips on this outlet; the first shared power with a Roomba charger, and the second was completely isolated. The server still rebooted in both scenarios.I ran Memtest86+ from the Unraid boot menu - it did 3 passes with no effort.I let Unraid run without my array starting - Unraid doesn't seem to reboot if I don't start my array. As long as I start my array, it will eventually reboot and report "Unclean shutdown detected". It doesn't matter if i turn on my Docker containers or not.I tried starting the array, but in maintenance mode. It still crashes after a few hours.I tried different ports for my 2-bay hard drive enclosure - I tried the rear USB ports, the front USB ports, and also the USB-C port on the front. They all ended up rebooting.I unplugged the USB data cable for my 2-bay hard drive enclosure and started my array in maintenance mode with 0 array devices, just running my NVMe cache. It seemed to have run for a very long time (from 3am, then it reboots at 12pm). I noticed that when it did reboot, Unraid reported "Stopped. Configuration valid" instead of saying "Unclean shutdown detected". This might have happened because I had 0 array devices?Tried configuring Docker from macvlan to ipvlan, didn't do anything.If anyone has any idea, I would really appreciate the help. Thank you!
June 5Jun 5 Author Here is the diagnostics file. I made sure to get a diagnostics file right after the system hard rebooted, so the syslog.txt and syslog-previous.txt files should show how the typical hard reboot happens, with my array started and pool devices attached. tower-diagnostics-20260604-2122.zip
June 5Jun 5 Community Expert Server rebooting by itself is almost always a hardware problem, or bad power.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.