J E wrote:
On Oct 18, 2007, at 8:23 PM, John Summerfield wrote:
J E wrote:
Installing chrooted bind on a RHEL 5 system, and I'm wondering about
what the docs are telling me vs. what I see in reality on my system.
I can just copy over our current configs from our debian system, but I'd
Not too exactly, there are some differences in the config file. The
zones would be okay.
Well, the default is apparently for a very basic caching configuration -
There is nothing there that I will retain for my config, other than
adjusting directory structures to match where RHEL puts the chroot.
What else would you do?
I use it as a starting point to add my zones. I don't change stuff I
don't need to.
prefer to stick to doing things the way the distro I am using does
things.
So, that being said, docs located here
https://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Deployment_Guide-en-US/s1-bind-rndc.html
say:
I've not yet set up BIND on Tikanga. When I did it on CentOS4 and
WBEL4, it just happened that way. I didn't do anything special, and I
didn't read the docs:-)
Hrm - well, it looks like either they broke something, or I did. I did
install and run the system-config-bind tool to see what it was all
about. I don't think it would have, but perhaps it did something
unexpected and blew away the files that were there?
rpm -Va bind\*
a) I'm not missing some other step/package that would actually place
this file there
or
b) if I put my own in place it won't get whacked by some future
upgrade to one of the bind packages.
Like Debian, RH is good at not whacking config files. For any package,
you can find out what the config files are using the rpm command.
Unlike Debian, RH doesn't keep asking inane questions about what to do
with changed configuration files. It automatically saves the old one, or
Good to know - I was hoping that this was the case, but as I noted,
RHEL5 and Yum are still new here, so we haven't really had much chance
to see how updates were handled in this whole process. It looks like Yum
is a bit nicer than up2date was, at least at first blush.
up2date has the ability to get source packages. I had it set to download
source for the binaries it retrieved. I'm not happy with yum.
Presumably, you plan on testing this system before placing it in
production? I suggest you take a note of your concerns as you go, and
then verify that those are all in the mind when you can.
Testing will be very basic. I'm not too worried about having to do some
of these things by hand -- I just wanted to get some confirmation on
what I hoped was the case, namely that updates down the road won't kill
me. Basically, just knowing that it does make a backup - along with some
good tape backups just in case - are all I need :-)
cp -pr /etc /root/etc
plus any /var directories you're worried about.
I'd not persist in doing that, but while you're a beginner it will be
comforting to have recovery so close to had.
Of course, my 4 Gb sandisk cruzer would be fairly handy too;-)
Not relevant to your Q, but remember selinux can cause grief. The best
way of dealing with it is to find what to do to make it do what you
want. For example, Apache can't by default serve from anywhere but
/var/www. It took me a while to figure how to serve my install images
from /var/local/mirrors
As I understand it, if the standard POSIX permissions says you can do
it, then selinux gets to vote.
--
Cheers
John
-- spambait
[EMAIL PROTECTED] [EMAIL PROTECTED]
Please do not reply off-list
_______________________________________________
rhelv5-list mailing list
rhelv5-list@redhat.com
https://www.redhat.com/mailman/listinfo/rhelv5-list
