I have this relatively old Supermicro server with m/b X7DBR-3, it has dual Xeon L5420 cpu and 32G ECC memory. I am trying to install Linux on it, tried both Ubuntu 16.04.4 and CentOS 7 (the distribution does not matter, it will run Docker either way). Installation goes smoothly, the system installs and reboots but then it never comes up completely. Ubuntu gets stuck after "Started Dispatch Password Requests to Console Directory Watch". CentOS is stuck after "Starting udev Wait for Complete Device Initialization...". I can remove kernel parameter "quiet" on CentOS to get more info, in this case, it stops at this place (this is the last line):
EDAC MC: Ver: 3.0.0
This seems to be related to RAM so I ran memtest. It passed memtest fine (took 4 hours to complete). This answer: Debian error : edac mc0 internal error describes similar error that is caused by un-initialized ECC RAM. I have tried to turn "quick boot" off in BIOS to let it run the memory test and then boot the system but this did not help, it got stuck at exactly the same place. Note also that unlike in the referenced answer, I do not actually get EDAC errors.
Another answer I have found https://askubuntu.com/questions/772494/stuck-on-ubuntu-installation-trial-started-dispatch-password-requests seems to be related and suggests adding kernel boot parameter "intel_idle.max_cstate=1". I have tried this and it did not help either.
I did not mess with CPU and chipset parameters in the BIOS. This system used to run VMWare ESXi just fine.
What could be the problem?
Side question: how do I get more detailed boot log on Ubuntu?
