svn commit: r239598 - head/etc/rc.d

[Ian Lepore]

On Thu, 2012-09-06 at 13:03 -0700, David O'Brien wrote:
> 
> Does 'ps' vary that much across the two invocations that we had in
> 'initrandom'?  Please post a diff to back up any "yes" answer.
> 
> We already have an invocation of 'ps'.  Please suggest a *different*
> command invocation.

When I was playing with this stuff to come up with those command
sequences I suggested, one thing I noticed was that the ps (with those
extra parms for sorting and detailed stats) did differ if used as the
first and last commands in the overall sequence.  Especially they
differed by being sorted into a slightly different order in the second
invocation.

I eventually came to the conclusion that the first one was still
superfluous, and the way I should leverage that difference was to put
any commands that report on ever-changing kernel statistics nearer to
the end of the list, so that the commands that run first get a chance to
perturb those stats as much as possible.

-- Ian