🚀 go-pugleaf

In Reply to this Note From: <Joe Buck>
o 
o In article <RJC.92Mar5114747@brodie.cstr.ed.ac.uk>, rjc@cstr.ed.ac.uk (Richard Caley) writes:
o |> For an operating system in particular MTBF is very meaningful.
o 
o I partially disagree: MTBF means something for some types of OS defects
o and is meaningless for others.
o 
[text deleted]
o 
o But OSes also have bugs that cause instant crashes as soon as a particular
o program does a particular thing.  In this case, MTBF is meaningless;
o if a user is running crashme and the OS has a defective trap handler,
o "MTBF" might be a few seconds; otherwise, it might stay up for years,
o until a particular program comes along that triggers the bug, in which
o case the system dies instantly whenever that program is run.

This is where the 'M' in MTBF comes in.  You can simply add the total
times the processor is running, divided by the number of crashes, and
that is your MTBF for the month.  You need not consider the source of
the crashes, simply do the math.  -David-

# david@wubios.wustl.edu             ^     David J. Camp                #
# david%wubios@wugate.wustl.edu    < * >   +1 314 382 0584              #
# ...!uunet!wugate!wubios!david      v     "God loves material things." #
#        "I am a Member of the League for Programming Freedom."         #