System suddenly reboots on high load with 4.13.0-26-generic

[Marco Kleiss]

Hi, I administer a system which encounters high load regularly over nighttime. Since Kernel 4.13.0-26-generic it suddenly reboots shortly after the load starts. When switching back to Kernel 4.10.0-42-generic everything works fine. I guess earlier versions of 4.13.0 worked too.
The system is running Ubuntu 16.04.3 LTS on a AMD EPYC 7351.
regards Marco

# lsb_release -rd
Description: Ubuntu 16.04.3 LTS
Release: 16.04

# apt-cache policy linux-image-4.13.0-26-generic
linux-image-4.13.0-26-generic:
  Installed: 4.13.0-26.29~16.04.2
  Candidate: 4.13.0-26.29~16.04.2
  Version table:
 *** 4.13.0-26.29~16.04.2 500
        500 http://de.archive.ubuntu.com/ubuntu xenial-updates/main amd64 Packages
        500 http://security.ubuntu.com/ubuntu xenial-security/main amd64 Packages
        100 /var/lib/dpkg/status

[Marco Kleiss (mkleiss)]

I recognized that after a crash I see the following entries in dmesg
# dmesg |grep BERT
[ 0.000000] ACPI: BERT 0x00000000DA1DB7C0 000030 (v01 AMD AMD BERT 00000001 AMD 00000001)
[ 2.457831] BERT: [Firmware Bug]: Invalid error record.

So I now rather think it is a Hardware failure than a bug. I am setting the status from new to invalid (hoping thats the correct state).