[pkg-ntp-maintainers] Bug#484974: Bug#484974: ntpdate's ip-up script runs even for virtual interfaces

Josip Rodin joy at debbugs.entuzijast.net
Sun Jun 15 23:53:22 UTC 2008


On Mon, Jun 16, 2008 at 01:43:18AM +0200, Peter Eisentraut wrote:
> Josip Rodin wrote:
> > So, where is the demand for ntpdate to be run on eth* interfaces? All I see
> > is people asking for it with PPP connections (typically, dialups and/or
> > desktops). Why aren't we doing this for ppp interfaces only?
> 
> It seemed like a useful generalization.  PPP use has declined much since the 
> bug was filed, but other usages have become possible.  Say, if I'm on a 
> laptop, plug in the cable or activate the wifi, everything is taken care of 
> automatically.  For that reason, for example, running ntpdate on eth* 
> interfaces is definitely useful.
> 
> What are you using ntpdate for?

I have had it on numerous servers since forever, and it's used ad hoc; most
of the machines run a ntpd (sometimes the NTP ntpd, sometimes openntpd).
The machines can tolerate ntpdate running on network interface upping time
on boot, but otherwise it's generally a problem when you do something random
with interfaces, yet ntpdate runs and changes system time, sometimes
upsetting openntpd because of a separate bug, and otherwise upsetting me
because it's fiddling with system time behind my back (causing scripts that
depend on time to warn or error, such as RRD updates).

I don't want to have to go rm the conffile or remove the package on N
machines just because of whatever happens on the some random desktop...

-- 
     2. That which causes joy or happiness.





More information about the pkg-ntp-maintainers mailing list