Symptom: My server is unable to run longer than 24 hours without some kind of error. Its seemingly random, I can't predict or force trigger it to act up.

Its generally one of the following

• Drives disconnecting/unable to be accessed from my PC

• Whole system lockup/freeze, can't do anything, can't access the gui, nothing. Have to hard-reboot.

• Docker Container errors, sometimes the containers freeze, lockup, docker crashes ( "Docker service failed to start" error )

• Unresponsive unraid GUI, plugging a monitor in, there isn't any longer the wall of text, but just a black screen. (drives are still accessible from my pc)
  
  This system has been running well for the last 2 years, but starting at the beginning of this year, its been constant chasing of errors, system instability, and unable to run for more than 24-48 hours max without error.

Things I've tried:

• Changing Ram ( I have 2 sets, both have passed 12 hours of MemTest86 runs )

• Replaced previous cache ssd with a new one. ( Samsung 990 evo 1tb )

• Changed Boot Flash Drives

• Upgraded to 7.3.0 and set up Internal boot to see if it was a flash drive error ( issue was happening on prior 7.2.4 release )

• Downgraded again, and then updated back to 7.3.0

• Bios updates on motherboard

• Disabling XMP ram settings

• Booting with 1 ram sticks

• Disabled/Deleted the majority of my Plugins

• Booted Windows on the hardware and ran Stress Test/Benchmarks on loop, and I wasn't getting any crashes.

• Bare minimum hardware, no usb's, no monitors, keyboards, etc. Just drives, cpu & ram.

• I'm currently testing changing my docker to run on 'ipvlan' over 'maclan'

Hardware

• Intel 12700k ( stock settings, not OC'd )

• MSI Z790 Pro Wifi DDR4

• 16gb generic DDR4 ram

• 64gb Corsair LPX DDR4 ram

• Corsair 750w PSU connected to a 1500va UPS ( with new batteries, replaced fall 2025 )

• 6 14tb HDD's, 64gb m.2 ( unraid boot drive ) , Samsung 990 evo 1tb Cache drive

• Noctua DH-15 with plenty of fans and airflow ( its not a thermal shutdown issue )

• No GPU

• The hardware is all about 2 years old, and aside from the HDD's, everything was purchased new.

I'm not running any VM's, or doing anything CPU heavy, for now, I've been slimming my docker to the basic plex/arr stack, and get crashes.

Forgive me for not being an IT professional, but I've put everything into chatgpt/claude and its just giving me loops of ram/ssd failure, which I find hard to believe that 2ssd's ( one new ) and 4 sticks of ram have all failed.

I don't know if its a hardware issue, the only thing I haven't touched is the cpu, psu, and motherboard, and don't want to blindly start throwing money at things hopping its the problem when I don't even know if its a hardware or software issue.

I've attached my full logs.

Thank you all.

epcot-diagnostics-20260421-2158.zip

Edited April 22Apr 22 by clayton.g.finley

Enable the syslog server to save to a share and post that afgter the next crash.

6 hours ago, JorgeB said:

Enable the syslog server to save to a share and post that afgter the next crash.

Will do.

When I made this post my latest attempt was to change my docker containers from macvlan to ipvlan, and that didn't change anything, this morning had a black screen, drives that were still accessible, but unable to access the web GUI and the docker containers were inaccessible. ( I was only running one container this time, plex )

Next on my list of things to do is change my cache drive from BTRFS to XFS to see if that helps anything.

Edited April 22Apr 22 by clayton.g.finley

Suggest stop changing stuff til Jorgeb can go through your syslog. You run the risk of introducing new problems every time you change something while flailing around blindly.

Updated with a system log of my most recent crash. Same as usual, sometime in the night drives are not able to be accessed from my PC, unresponsive GUI, system freeze/crash.

per AI, here's the crash

Apr 23 04:14:38 Epcot kernel: BUG: kernel NULL pointer dereference

Apr 23 04:14:38 Epcot kernel: smartctl_type[569660]: segfault...

epcot-syslog-20260423-1424.zip

Looks like you posted the regular syslog; it only covers 2 minutes of uptime.

STUPID question, do you use SAS drives?

I had something similar. A drive would be perfect for a few hours, then SOMETHING would make it error out and die.

SMART said it was fine.

Turns out, I had the stupid power cable jammed in and that was causing the errors. Rearranged the drives so the SAS power wasn't tight up against the power supply, and haven't had an issue since.

Cables man, sometimes it's just cables.

On 4/22/2026 at 1:10 PM, clayton.g.finley said:

Symptom: My server is unable to run longer than 24 hours without some kind of error. Its seemingly random, I can't predict or force trigger it to act up.

Its generally one of the following

• Drives disconnecting/unable to be accessed from my PC

• Whole system lockup/freeze, can't do anything, can't access the gui, nothing. Have to hard-reboot.

• Docker Container errors, sometimes the containers freeze, lockup, docker crashes ( "Docker service failed to start" error )

• Unresponsive unraid GUI, plugging a monitor in, there isn't any longer the wall of text, but just a black screen. (drives are still accessible from my pc)
  
  This system has been running well for the last 2 years, but starting at the beginning of this year, its been constant chasing of errors, system instability, and unable to run for more than 24-48 hours max without error.

Things I've tried:

• Changing Ram ( I have 2 sets, both have passed 12 hours of MemTest86 runs )

• Replaced previous cache ssd with a new one. ( Samsung 990 evo 1tb )

• Changed Boot Flash Drives

• Upgraded to 7.3.0 and set up Internal boot to see if it was a flash drive error ( issue was happening on prior 7.2.4 release )

• Downgraded again, and then updated back to 7.3.0

• Bios updates on motherboard

• Disabling XMP ram settings

• Booting with 1 ram sticks

• Disabled/Deleted the majority of my Plugins

• Booted Windows on the hardware and ran Stress Test/Benchmarks on loop, and I wasn't getting any crashes.

• Bare minimum hardware, no usb's, no monitors, keyboards, etc. Just drives, cpu & ram.

• I'm currently testing changing my docker to run on 'ipvlan' over 'maclan'

Hardware

• Intel 12700k ( stock settings, not OC'd )

• MSI Z790 Pro Wifi DDR4

• 16gb generic DDR4 ram

• 64gb Corsair LPX DDR4 ram

• Corsair 750w PSU connected to a 1500va UPS ( with new batteries, replaced fall 2025 )

• 6 14tb HDD's, 64gb m.2 ( unraid boot drive ) , Samsung 990 evo 1tb Cache drive

• Noctua DH-15 with plenty of fans and airflow ( its not a thermal shutdown issue )

• No GPU

• The hardware is all about 2 years old, and aside from the HDD's, everything was purchased new.

I'm not running any VM's, or doing anything CPU heavy, for now, I've been slimming my docker to the basic plex/arr stack, and get crashes.

Forgive me for not being an IT professional, but I've put everything into chatgpt/claude and its just giving me loops of ram/ssd failure, which I find hard to believe that 2ssd's ( one new ) and 4 sticks of ram have all failed.

I don't know if its a hardware issue, the only thing I haven't touched is the cpu, psu, and motherboard, and don't want to blindly start throwing money at things hopping its the problem when I don't even know if its a hardware or software issue.

I've attached my full logs.

Thank you all.

epcot-diagnostics-20260421-2158.zip

I am having the same issue, where the GUI won't work after a very short amount of time, I have only started using Unraid though.

8 minutes ago, The-Vault Master said:

I am having the same issue, where the GUI won't work after a very short amount of time, I have only started using Unraid though.

If you need help, please start your own thread and post the diagnostics, or it can get confusing in this one.

35 minutes ago, JorgeB said:

If you need help, please start your own thread and post the diagnostics, or it can get confusing in this one.

Done, thank you

4-24 crash

Docker crashed

"Docker Service failed to start"

GUI is accessible, mapped drives on my pc are accessible. System was up for no longer than 17 hours before crash.

epcot-syslog-20260424-2004.zip

full_diagnostics.tar.gz

Edited April 24Apr 24 by clayton.g.finley

A .NET application (likely a Docker container) crashed, could be a corrupt container image, but most likely a hardware issue. RAM, CPU would be the main suspects.

I have 4 sticks of ram, the likely hood of all 4 failing/being bad seems highly unlikely to me, I think for now I'll try a new motherboard, then move onto the cpu and ram if that doesn't fix anything.

4 minutes ago, clayton.g.finley said:

I have 4 sticks of ram, the likely hood of all 4 failing/being bad seems highly unlikely to me

Only one would need to fail for memory to be bad so how likely it is all 4 are failing doesn't matter.

3 hours ago, trurl said:

Only one would need to fail for memory to be bad so how likely it is all 4 are failing doesn't matter.

I've been testing with a single stick in the system, and the issues persist

Today 4-26 I've had the GUI fail twice, drives are still accessible via my pc, but can't log into the unraid GUI, only fix was rebooting, and then gone back the "docker service failed to start"

Edited April 26Apr 26 by clayton.g.finley

I think i'm going to just replace my motherboard, start there.

Memtest86 has passed all my ram sticks running multiple sessions, and the motherboard is the cheapest part i can start with.

Looking for suggestions for the following:

• 6 sata port connections

• 2.5gb network

• LGA 1700

• DDR4
  
  I can get by with 1gb for now, and add a 2.5gb network card later, if I HAVE to I can get by with less sata ports.
  
  Just looking for something cheap and stable to test with and hopefully I can RMA my motherboard, but knowing how that goes, its a coin-flip.

Disable Docker and VM Manager in Settings. Boot in SAFE mode. See if it still has problems like that.

So far motherboard swap has fixed it, hopefully I can get it RMA'd since you can't buy good lga 1700 ddr4 motherboards anymore.