We have three quasi-embedded systems with a MiTAC Mitac PD11BI motherboards running Ubuntu 14.04.3 Desktop (application requires GUI) and Kernel 3.19.0-28-generic that have been randomly locking up/dropping off the network at random times. It appears that the servers are going into some type of floating state where system is completely unresponsive to any type of communication and all logging and applications stop.
We run these servers headless, so I can't get a visual on what might be happening, and plugging a monitor yields a blank screen. SSH/TeamViewer connections become unresponsive as well.
The only way to regain access to the system is to hard reboot.
I've pored over /var/log/syslog, /var/log/dmsg, and /var/log/kern.log to try to find any type of pattern during the crashes, but I can't find any consistency in the logs at the time of the crash. Running "last" after reboot indicates a "Crash"
The really odd thing is, we have two other units with the same hardware that are running Debian Jessie that have had no problems whatsoever.
I'm pulling my hair out as there seems to be no rhyme or reason for these issues. My gut reaction is that the problem is Hardware, but the Jessie systems run fine. In fact, one of the now locking-up Ubuntu systems ran Jessie fine for months and the issues popped up as soon as it was wiped and Ubuntu was installed. Any help provided would be very appreciated!
[link][7 comments]