sshd dying on BeagleBone

[Ronald Klop]

On Tue, 12 Feb 2013 01:16:08 +0100, Brett Wynkoop <wynkoop at wynn.com> wrote:

> Greeting-
>
> While looking at the console I saw that sshd died on signal 11!  So now
> we have a clue.  If I recall correctly signal 11 is a seg fault.
>
> -Brett
>

You recall correctly. But a seg fault can have multiple reasons like a  
programming error in ssh or corrupted memory (like errors in the kernel  
which shuffle pages or whatever) or faulty memory chips so a pointer is  
flipped or ....
Do you have a core dump from ssh? If not you should configure the machine  
to make them.

You can instruct where core files are dumped. My ARM computer says:
$ sysctl kern | grep core
kern.corefile: %N.core
kern.nodump_coredump: 0
kern.coredump: 1
kern.sugid_coredump: 0

But my amd64 desktop has this:
kern.corefile: /var/tmp/%U.%N.core

See 'man core' for more information.

Regards,
Ronald.