hi - just got out of a slightly adventurous build with supermicro h11ssl-c rev2 (epyc 7281, separately tested w/ two sets of ram: 4 x supermicro-certified micron 32gb ram / 4x32gb from nemix, seagate barracuda 10tb w/ or w/o m.2 samsung 1tb, couple of gpu sets: nvidia geforce rtx2080 / k80 / no gpu).
after a couple of calls with supermicro tech support was advised to rma the board (suspected hw problem).
h11ssl-c showed some regular instabilities: either a spontaneous reset or hang up with ledsas blinking red, most similar to post below
community.amd.com
if there was one regularity in terms of contributing factors it was a switch between the system running on idle and on 50%+ load; uptime between resets varied between 15min and 48hrs depending on hw combination / load
tested multiple scenarios (gradual build from just cpu+1 memory stick and no chassis through full build), various os flavors (tested a few of supermicro recommended variants but mainly ubuntu 18.04.2 lts server), various bios settings recommendations from online forums and supermicro online help (turned off core performance boost, global c-state control, pcie spread spectrum, watchdog, determinism slider set to performance etc etc); no meaningful system logs
while waiting for replacement / planning for next steps thought i might ask for some help (sorry, i know i'm casting a fairly wide net here):
1) build experience with similar hw: specifically epyc 7001 series and h11ssl-c; any hw / os / bios settings recommendations or observations?
2) regardless of specific motherboard choice, any recommendations for linux distro / bios settings / kernel options that are expected with epyc 7001 series running on supermicro?
3) experience with other (non supermicro) mb running epyc 7001, asrock rack rome d8-2t v. supermicro?
thank you!
after a couple of calls with supermicro tech support was advised to rma the board (suspected hw problem).
h11ssl-c showed some regular instabilities: either a spontaneous reset or hang up with ledsas blinking red, most similar to post below
epyc 7551 spontaneously resets after 10mins rendering
I just finished building a system, dual 7551 epyc cpus using the supermicro H11DSi-NT motherboard. The build went very well, and the system is running just fine, and the performance is extraordinary. I am using Fedora 27 linux, but have access to about 20 different linux distros, as I maintain...
if there was one regularity in terms of contributing factors it was a switch between the system running on idle and on 50%+ load; uptime between resets varied between 15min and 48hrs depending on hw combination / load
tested multiple scenarios (gradual build from just cpu+1 memory stick and no chassis through full build), various os flavors (tested a few of supermicro recommended variants but mainly ubuntu 18.04.2 lts server), various bios settings recommendations from online forums and supermicro online help (turned off core performance boost, global c-state control, pcie spread spectrum, watchdog, determinism slider set to performance etc etc); no meaningful system logs
while waiting for replacement / planning for next steps thought i might ask for some help (sorry, i know i'm casting a fairly wide net here):
1) build experience with similar hw: specifically epyc 7001 series and h11ssl-c; any hw / os / bios settings recommendations or observations?
2) regardless of specific motherboard choice, any recommendations for linux distro / bios settings / kernel options that are expected with epyc 7001 series running on supermicro?
3) experience with other (non supermicro) mb running epyc 7001, asrock rack rome d8-2t v. supermicro?
thank you!