Package: bind9
Version: 1:9.5.1.dfsg.P2-1+lenny1
I have some systems which are affected by symptoms similar to
#169577. This bug is not about that actual problem. This bug is
about the fact that /etc/init.d/bind9 stop does not work if rdnc is
broken for any reason. See transcript below.
Either:
(a) The init.d script should use start-stop-daemon with
appropriate options to forcibly kill named;
or:
(b) The bind9 daemon should fail to start if it cannot grab the
control channel, so that it is not possible to have an
uncontactable daemon.
The latter would likely cause some systems' nameds to stop working
entirely with the current state of things so I would strongly
recommend (a). I only mention (b) because it would have been an
alternative reasonable design at the start.
Ian.
-chiark:~> really su -
# /etc/init.d/bind9 stop
Stopping domain name service...: bind9rndc: connect failed: 127.0.0.1#953:
connection refused
.
# ps -ef | grep named
root 31549 1 2 Jul08 ? 00:54:44 /usr/sbin/named
root 20950 10426 0 11:12 pts/177 00:00:00 egrep ^Jul 8.*named
/var/log/daemon.log
root 26408 26135 0 11:17 pts/176 00:00:00 grep named
# kill 31549
# ps -ef | grep named
root 20950 10426 0 11:12 pts/177 00:00:00 egrep ^Jul 8.*named
/var/log/daemon.log
root 26464 26135 0 11:17 pts/176 00:00:00 grep named
#
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
