On 9/25/06, Vasil Dimov <[EMAIL PROTECTED]> wrote:
On Mon, Sep 25, 2006 at 11:47:46AM +0200, Erwin Van de Velde wrote:
> On Monday 25 September 2006 11:26, Vasil Dimov wrote:
> > IMHO in this case the administrator should be changed,
> > not the way ports operate. Building idiot-proof system is impossible and
> > leads to the:
> >
> > Shaw's Principle:
> > Build a system that even a fool can use, and only a fool will
> > want to use it.
>
> This is not about being idiot-proof but being handy to use. It also reduces
> service downtime by immediately restarting the service after update instead
> of waiting for the administrator to restart it.
I think that implementing this feature the right way would be too
complex and would bring too little benefit. By "right way" I mean that
a given service should be started upon installation only if it was
actually stopped during the upgrade process.
Ofcourse if someone thinks that it is worth implementing I would be
happy to be asked
"service xyz was stopped during upgrade, do you want to start it now?" [y]
One could always minimize downtime by doing
portupgrade xyz && /usr/local/etc/rc.d/xyz start
Our ports system does not have a notion of upgrade action. We'll
be discussing this later this year. The problem is acknowledged,
but the only current guideline is not to stop a service at deinstall,
unless it's really necessary. Many maintainers follow this rule.
_______________________________________________
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
