Unraid 6.7 continuously reboots

[christopherhughes]

Hi, recently build my first Unraid server (previously used nap-it on Solaris 11.3 and Centos7).

 

Asus x570 motherboard

rhyzen 3900x CPU

32gb EEC Ram
2 x M2 drives for cache and dockers
Sas9220-8i HBA

nvidia quadro P2000

APC UPS

system was running fine for 3 weeks, last night it mysteriously rebooted,  and I’m not able to boot into headless Unraid. Attached is the footage of the boot up before screen goes blank and system reboots again.   

very grateful for and recommendations.  Booting into safe mode with no plugins makes no difference and also no memory errors reported after running the mem test from the menu

 

thanks in advance

chris

 

 

[christopherhughes]

Still having this problem.

 

What I've tried so far.

 

1) Updated Asus x570 Bios to latest

2) Copied backed up Unraid install to new USB

3) Attempted to boot from clean new 6.7.2 install

 

 

[Squid]

Enable the syslog server and set it to mirror the log to the flash drive (Tools - Syslog server).  Wait until the next reboot and then post the syslog from the flash drive here.

 

Also would be great if you could also upload your diagnostics

[christopherhughes]

Hi Squid,  thanks for your reply.  I can’t even get the system to boot to be able to access the menu.  Is there a way to enable syslog server without accessing the menu?  perhaps a config hack?.   
 

Spent a fun day playing with Bios settings and different USB sticks / versions.  Nothing is working,  but I’ve just ran the memtest83 from a clean install of the latest 6.8 RC and it’s found errors.

 

I’m running with ECC memory, (Crucial 2x16GB PC4-2666V-EE1 CT16G4WFD8266).  I can’t find any ECC memory reported on ASUS’s QVL list,  so I’ve ordered 2 sticks of Corsair non ECC memory arriving tomorrow from Amazon to try.

I’ll report back!

 

 

 

[Squid]

2 minutes ago, christopherhughes said:

Spent a fun day playing with Bios settings and different USB sticks / versions.  Nothing is working,  but I’ve just ran the memtest83 from a clean install of the latest 6.8 RC and it’s found errors.

Even though you've already ordered memory, would be good to try disabling XMP / any overclocking you've got.  

[christopherhughes]

I can confirm XMP is disabled and cpu is not overclocked. 

[christopherhughes]

Replacing the memory fixed the problem!  Back up and running. 😁 Replaced my USB stick also as the Sandisk Cruzer plastic was slipping of the metal body after all the inserting and pulling out trying to test different config options.   

New;

Corsair 163301 Vengeance LPX 32 GB (2 x 16 GB) DDR4 3200

 

Faulty;

Crucial 2x16GB PC4-2666V-EE1 CT16G4WFD8266

[John_M]

Was there any reason for thinking that ECC RAM would work or did you just happen to have some and thought you'd give it a try?