[15587] in Kerberos_V5_Development
Re: Project review: Parallel KDC
daemon@ATHENA.MIT.EDU (Nicolas Williams)
Thu Mar 11 18:33:54 2010
Date: Thu, 11 Mar 2010 17:32:50 -0600
From: Nicolas Williams <Nicolas.Williams@sun.com>
To: Greg Hudson <ghudson@mit.edu>
Message-ID: <20100311233250.GO19297@Sun.COM>
MIME-Version: 1.0
Content-Disposition: inline
In-Reply-To: <1268350090.18898.502.camel@ray>
Cc: "krbdev@mit.edu" <krbdev@mit.edu>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: krbdev-bounces@mit.edu
On Thu, Mar 11, 2010 at 06:28:10PM -0500, Greg Hudson wrote:
> On Thu, 2010-03-11 at 17:54 -0500, Nicolas Williams wrote:
> > - I don't see the need for two options. One option will do.
>
> To clarify, I am only adding one option (-w NUM). The project proposal
> mentions an existing option (-n) as being exclusive with -w, because
> "don't fork, but fork off 5 worker threads" doesn't make sense.
>
> Or at least, that was my thinking at the time. Since then, it became
> clear to me that there needs to be a supervisor process to make the
> existing -P pidfile option meaningful. -n -w 5 could mean "fork off
> five worker threads and run the supervisor in the foreground." I don't
> know what would be a good application for that, other than perhaps the
> test suite.
I agree, these are not really mutually exclusive.
Nico
--
_______________________________________________
krbdev mailing list krbdev@mit.edu
https://mailman.mit.edu/mailman/listinfo/krbdev
