Yo Alan! On Wed, 29 Aug 2012 17:08:43 -0400 (EDT) Alan Stern <[email protected]> wrote:
> > Uh, not easy. This is a production machine. Acceptable downtime is
> > very small. There should be a way to turn it off at runtime.
>
> No, there shouldn't. It's a debugging feature; while debugging we
> want to see all occurrences of these messages.
>
> Production machines should not run debugging kernels.
We'll have to agree to disagree. My experience is the fun bugs only happen
in production. A big NASA study years ago proved that after spending $1M
per line of code that only half the bugs could be caught before live
missions.
> > Sadly the problem went away after a reboot. I suspect an
> > initialization issue, but doing a git bisect of an intermittent
> > problem on a production machine is not gonna happen.
>
> Especially if the problem doesn't reappear. One-time issues are very
> hard to fix.
Certainly not a one-time issue. I've seen it before, and will see it
again, so I'll just keep gnawing at it.
RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
[email protected] Tel:+1(541)382-8588
signature.asc
Description: PGP signature
