[pkg-ntp-maintainers] Dynamic interface scanning and restarting.

[Kurt Roeckx]

On Tue, Apr 24, 2007 at 09:35:10PM +0200, Peter Eisentraut wrote:
> Am Sonntag, 15. April 2007 12:40 schrieb Kurt Roeckx:
> > We currently seem to have lots of options to restart ntp at various
> > places.  The dynamic interface scanning seems to be working properly
> > now.  5 minutes after I change the interface, it deletes the old one and
> > creates a new one.
> 
> Btw., how do you check that?  I couldn't get any information about when and 
> why it loses interfaces and readds them.

Part of the /var/log/daemon.log file:
Apr 15 12:21:29 intrepid ntpd[2863]: Listening on interface #6 eth0, 10.0.0.2#123 Enabled
Apr 15 12:21:29 intrepid ntpd[2863]: Deleting interface #5 eth0, 10.0.0.1#123, interface stats: received=5209, sent=5291, dropped=0, active_time=6301 secs
Apr 15 12:23:33 intrepid ntpdate[7088]: the NTP socket is in use, exiting
Apr 15 12:23:34 intrepid ntpd[2863]: sendto(XXX) (fd=22): Invalid argument
Apr 15 12:23:50 intrepid last message repeated 8 times
Apr 15 12:24:49 intrepid ntpd[2863]: sendto(XXX) (fd=22): Invalid argument
Apr 15 12:25:07 intrepid ntpd[2863]: sendto(XXX) (fd=22): Invalid argument
Apr 15 12:25:45 intrepid ntpd[2863]: sendto(XXX) (fd=22): Invalid argument
Apr 15 12:26:01 intrepid last message repeated 8 times
Apr 15 12:26:29 intrepid ntpd[2863]: Listening on interface #7 eth0, 10.0.0.1#123 Enabled
Apr 15 12:26:29 intrepid ntpd[2863]: Deleting interface #6 eth0, 10.0.0.2#123, interface stats: received=9, sent=9, dropped=20, active_time=300 secs

I've changed from 10.0.0.1 to 10.0.0.2 and back to 10.0.0.1.

The line with ntpdate is when the interface comes up.

After the first 2 lines it was working, and after the last line it was
also working.


Kurt