Hi Shad,
You are right about hostnames panel (I did say I was a newbie !

). The reason for this is because the older bind version uses ndc to reload. Not knowing how to properly use rndc (the replacement of ndc), I took a different course.
A quick (not necessarily elegant) solution was to do this :-
Edit the file /etc/e-smith/events/actions/restart-named
Change the following section by commenting out all the other lines and modify / add the last 2 as follows :-
# my @cmd = ("/usr/sbin/ndc");
# my $chroot = db_get_prop(\%conf, 'named', 'chroot') || 'no';
# if ($chroot eq 'yes')
# {
# push(@cmd, '-c', '/home/dns/var/run/ndc');
# }
# push(@cmd, "-q", "reload");
# system(@cmd)
# and warn("could not run ndc to reload config");
system("/etc/init.d/named", "reload") == 0
or warn("could not reload named");
That should take care of the hostnames user panel.
Many thanks for pointing this out Shad. In case anyone reading this have not already found out, I am working off a test system, not the production server for obvious reasons. Once I managed to iron out all the problems, then I'll put in onto the production server.
By the way, reading the messages log, I discovered that there are 2 options being written to the named.conf file that is not supported and just ignored in the new version of bind. As it does not seem to interfere with the operation of named, I'm going to put off looking at it just at the moment and concentrate on possible problems that might stop named working correctly.
Kelvin