Subject: CS20 disappointments
To: None <port-alpha@netbsd.org>
From: Stephen M. Jones <smj@cirr.com>
List: port-alpha
Date: 09/23/2002 12:49:41
I've corresponded with a few people from this list about the CS20 .. while
its a kickarse microcomputer, its sadly unsuitable for production in remote
locations (unless you have a minon who can press a reset switch).  The
I2C remote management via the primary ethernet interface is a myth .. its
implementation was never completed.  At best you can compile a program to
query the temperature of the machine and have it automatically shut off at
a particular temperature .. but that has to be done from the OS.  

On most ALPHAs we have RCM or later RMC which is such a big win.  It looks
like API wanted that type of feature but never finished the implementation.

A barfy kludge would be to wire up x10 boxes to the machine(s) so that when
a hang does occur, you could power cycle the thing.  But there is another
problem .. SRM states that you can set a variable to poweron when AC is
applied but this doesn't seem to give the desired results!  Again, unfinished
implementation .. its quite sad because the CS20 is so cool.