[pkg-ntp-maintainers] Bug#535049: Bug#535049: ntpd does not work (removes every server) when no network at boot time

[Vincent Lefevre]

On 2009-06-30 20:55:00 +0200, Kurt Roeckx wrote:
> I think your problem is that you have a local named that returns
> a permanent error to ntpd when it tries to resolv.

Yes, because at boot time, the machine isn't connected yet.

> As far as I know ntpd will behave correctly when it can't reach a
> nameserver.

But in that case, why isn't there a script in /etc/network/if-up.d
to restart ntpd? In fact there was one in the past, but it has been
removed probably because:

ntp (1:4.2.4p0+dfsg-1) unstable; urgency=low

  This version of ntp will periodically rescan the network interfaces to
  pick up new and deleted interfaces.  This should supplant most or all
  of the various workarounds in use such as restarting the daemon in
  /etc/network/if-up.d/ or /etc/ppp/ip-up.d/.

 -- Peter Eisentraut <petere at debian.org>  Thu,  3 May 2007 11:32:29 +0200

So, ntpd doesn't behave as documented.

-- 
Vincent Lefèvre <vincent at vinc17.org> - Web: <http://www.vinc17.org/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/>
Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)