Re: ppp rcvar

[Marc Balmer <marc%msys.ch@localhost>]

Am 13.01.2013 um 13:10 schrieb Patrick Welche <prlw1%cam.ac.uk@localhost>:

> On Tue, Aug 21, 2012 at 05:26:26PM +0100, David Brownlee wrote:
>> On 21 August 2012 16:56, Patrick Welche <prlw1%cam.ac.uk@localhost> wrote:
>>> On Tue, Aug 21, 2012 at 10:36:56AM -0500, Jeremy C. Reed wrote:
>>>> On Tue, 21 Aug 2012, Patrick Welche wrote:
>>>> 
>>>>> Is the lack of rcvar in rc.d/ppp intentional or an oversight?
>>>>> 
>>>>> (Now I'm not sure if I had put ppp=NO or pppd=NO in rc.conf, and was then
>>>>> surprised that the phone was engaged hours after booting)
>>>> 
>>>> http://mail-index.netbsd.org/tech-misc/2008/02/23/msg000005.html
>>>> 
>>>> Some of the thread answers why it happened.
>>> 
>>> It seems that ppp isn't considered in the thread. I would expect ppp=YES to
>>> mean at boot run ppp and connect to $ppp_peers | /etc/ppp/peers. If ppp=NO
>>> then wait for a /etc/rc.d/ppp onestart.
>>> 
>>> Thoughts?
>> 
>> Makes perfect sense to me - providing the defaults are left at YES I
>> cannot imagine why almost all rc.d files should be checking an
>> rcvar...
> 
> I was about to apply the attached, but now I'm two minds:
> 
> The default would be ppp=YES, ppp_peers="", which means that by default we
> would get
> 
> ./ppp: WARNING: ${ppp_peers} is not set - pppd was not started.
> 
> Opinions?

The default must not be ppp=YES (at least not in /etc/rc.conf) and in no case 
it is ok that have this warning at every boot.

> 
> Cheers,
> 
> Patrick
> <ppp.diff>