svn commit: r290010 - vendor/ntp

[Ian Lepore]

On Tue, 2015-10-27 at 19:45 +0300, Dmitry Morozovsky wrote:
> On Tue, 27 Oct 2015, Ian Lepore wrote:
> 
> > > > Log:
> > > >   Document NTP import procedure, using experience from the last
> > > one.
> > > 
> > > BTW:
> > > 
> > > marck at freefall:~> ntpq -c pe
> > >      remote           refid      st t when poll reach   delay  
> > > offset  jitter
> > >
> =====================================================================
> > > =========
> > > +ntp0.ysv.freebs 95.87.161.134    4 u  296 1024  377    0.422   
> > > -0.620   0.598
> > > *ntp1.ysv.freebs 39.86.174.157    3 u  443 1024  377    0.405   
> > > 0.904   0.239
> > > +ntp2.ysv.freebs 65.182.224.39    3 u  617 1024  377    0.434   
> > > 1.225   1.014
> > > marck at freefall:~> ntpdc -c pe
> > > localhost: timed out, nothing received
> > > ***Request timed out
> > > 
> > > I did not bisect to find when this brokes, but it is for quite
> some
> > > time.
> > > 
> > > [snip]
> > > 
> > > 
> > 
> > That's not broken, it's a "feature" of the newer ntpd.  The ntpd
> folks
> > decided to stop supporting the old mode-7 control packets because
> they
> > say the mode-6 packets used by ntpq can do everything necessary. 
> That
> > means we all have to learn some new commands within ntpq to get the
> > info and controls we're used to with ntpdc.  There are things I'm
> used
> > to doing with ntpdc that I haven't figured out how to do with ntpq
> yet,
> > foremost being 'addserver'.
> 
> well, why not to exclude ntpdc from building then? A least for 11.*
> 

I think it's because you may be running on 10-stable or 11 but still
need to administer downrev ntpd running on 8.2 or whatever, so you
would still need ntpdc to talk to those.  Note that this is
speculation... I don't know for sure that they had to add anything to
ntpd itself to help ntpq take over all the functionality of ntpdc.

-- Ian