[SOLVED] [6.8.3] unRaid Constantly Freezing/Crashing

[coffeeroasted]

I'm at my wit's end. I've been running unRaid for approximately 2 weeks now and I can't figure out why my system keeps freezing and/or crashing. I just finished a new build and am using all new parts with the exception of my HDDs and an SSD. For the first few days, I had no issues. Starting off slow, I added a couple plugins, created a Windows 10 VM, and figured out how to pass through a USB device. Then I replaced the 500 GB NVME drive I had available with a WD Black 1 TB NVME drive. From that point, it started freezing up. I'd see two different screens (see images at bottom). I started a syslog server and have scoured the log every time it crashes and I can't see anything that indicates why the system crashes. Whenever it freezes/crashes, the GUI is unavailable, I can't ssh into the computer, and it has to be hard shutdown. There doesn't seem to be any specific task that causes this issue. It's frozen/crashed in the middle of pre-clear, loading a plug-in, accessing an SMB share from my laptop, etc.

 

--EDIT: I'm on my 3rd clean install of unRaid 6.8.3, and my second USB device. Both flash drives were brand new Sandisk Ultra USB 3.1 devices. This issue happens regardless of completely clean install or if I'm running plugins. I've tried different data drives as well.

 

I thought maybe there was something wrong with the NVME drive I used, so I swapped the old one back in. It ran for an entire day, even allowing a full pre-clear (although the drive failed post-read). It's still doing the same thing. I ran a Memtest for 36 hours with 0 faults. At this point, I'm thinking it's MB, CPU, or PSU. I don't see any signs of PSU problems, though. Here's what I'm currently running:

 

AMD Ryzen 5 Pro 3400GE

Gigabyte B450 AORUS M mobo (BIOS on most recent update)

32GB HyperX Black DDR4 3200MHz 

SilverStone Tek 550W 80 Plus Gold PSU

 

Syslog is attached. Something weird to point out: time shifts by 4 hours sometimes in the log - I'll set my BIOS time to the correct time locally, but syslog shows it 4 hours early until ntpd resolves. Then after a while, the time in my BIOS is set to 4 hours later than local time. The sylog is quite long, I've been running it since Sunday. I'm open to any recommendations.

 

Thanks in advance!

syslog_2May_0958.log

Edited May 6, 2020 by coffeeroasted
Added info about install and flash drives

[JorgeB]

Start here:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

 

[coffeeroasted]

16 minutes ago, johnnie.black said:

Start here:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

 

Awesome, thanks so much! I've been banging my head against the wall trying to figure this out. I had seen the P-State issue listed, but saw it was only an issue pre-6.8 and also not really a problem with Zen+ or higher. Didn't catch the note about OC.

 

Gonna have to take a look at my setup now. Gigabyte's QVL for this CPU listed my setup as supported up to 2933 MHz, but it seems I was sent DR RAM even though I ordered SR (based on part number).

 

I've returned the RAM speed back to stock @ 1.2v and will report in a couple days.

[coffeeroasted]

On 5/2/2020 at 10:32 AM, johnnie.black said:

Start here:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

 

Thanks, again. Everything seems to be working swimmingly now. I've had a couple crashes since making this change, but it had to do with an HDD issue. Is there a way to tag this as solved or do I just edit the title? Or don't worry about it?

[itimpi]

1 minute ago, coffeeroasted said:

Thanks, again. Everything seems to be working swimmingly now. I've had a couple crashes since making this change, but it had to do with an HDD issue. Is there a way to tag this as solved or do I just edit the title? Or don't worry about it?

Edit the title on the first post to add (SOLVED).