On Mon, Jan 15, 2007 at 11:04:01AM +0100, Sjoerd Simons wrote: > On Mon, Jan 15, 2007 at 01:15:13AM +0100, Kurt Roeckx wrote: > > The postint script is doing just the same thing, with the line > > before it doing the right thing: > > > > # Ask the bus to reload the config file > > if [ -x "/etc/init.d/dbus" ]; then > > invoke-rc.d dbus force-reload || true > > fi > > /etc/init.d/avahi-daemon restart > > > > Since the policy doesn't allow dbus to get started, avahi-daemon > > also failed to start. > > No the postinst script does the right thing. Why ? Because avahi isn't > normally > started via the normal /etc/rc*.d symlink, so invoke-rc.d just doesn't work > for > avahi.
So, avahi shouldn't be run (or stopped) in any runlevel, but should always run anyway? > Putting avahi in both /etc/init.d/ and /etc/dbus/event.d was a choice of > the original packager, which has both it's good and bad sides :) If I understand you right, dbus should start it for you? Then I don't understand why you need to start it in the first place, specially not in the postinst. If dbus doesn't start it when it should, I think dbus needs to be fixed. Kurt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
