[rescue] SGI quad IP25 cpu question

[Scott Newell]

At 11:37 PM 2/25/2003 -0600, Jonathan C. Patschke wrote:
>I'm sure it wasn't intended that way, but it's an interesting form of
>paranoia--if the user can't determine whether or not the CPU has failed
>from the result displayed on the front panel, cause the CPU to enter the
>only provably deterministic state.

Yeah, but if the led comes on to indicate a failed CPU, a cold solder joint
couldn't look like a CPU failure.  Now if the led illuminates to indicate a
good CPU, I can understand how the resulting open LED circuit could be
misdiagnosed as a bad CPU.  

Still, it's pretty funny, to me at least.


newell