10.3-stable random server reboots - finding the reason

[tech-lists]

On 18/04/2017 13:34, Kurt Jaeger wrote:

> 1)
>     echo '-Dh -S115200' > /boot.config
> 2)
>     vi /boot/loader.conf
> ------------
> console="comconsole"
> comconsole_speed=115200
> ------------
> 
> Then: connect it to some kermit session from a different box,
> and write a session log. This might capture some last gasp.
> 

A bit more detail:

Problem seems to be it's DRAC is broken. The server is a Dell R710 in a
datacentre so I've not got physical access. I'd normally admin the
hardware via the DRAC but it's non-functional now.

The system won't boot without someone there plugging in a keyboard and
pressing F2 to continue... arrrgh! "idrac failed press F2". It does
boot, though. The DRAC normally shows fan speed and voltage amongst
other things. Right now I'm looking for an in-band method of doing that
on FreeBSD as obv the out-of-band method has failed. cpu temps look
fine, cool.

AIUI the DRAC is a chip on the motherboard and not a module that can be
whipped out and replaced. For all I know, the DRAC might be causing the
reboot if it's only intermittently dead. I don't know this though. It
might be CPU or RAM, both of these can be swapped out, just need to
capture the info...

arrgh!
-- 
J.