From: Nico Kadel-Garcia on
Some time back, I growsed here about the difficulty keeping a static
Linux server's DNS information alive in a Windows Active Directory
server that had DNS expiration turned off. There turn out to be two
workable solutions.

1) Use dynamic DNS on the Linux client. This will keep refreshing the
DNS information, and on RHEL is done by setting up a '/etc/dhclient-
eth0.conf' file with the relevant information. This particularly
requires dynamic DNS to *not* require authentication on the Active
Directory server to set DNS information, so it requires either passive
cooperation by the Active Directory manager, or their being security
careless enough not to require authentication. But this is not an
uncommon setup, and it works.

2) Assign a subdomain for the Linux hostname that has only the Linux
hostname's A record, any relevant subdomains, and possibly NS records,
in it. Set *that* subdomain to be static.

For example, for mymachine.domain, assign a subdomain on Active
Directory for the name 'linuxhost.domain' that has NS records pointing
to the Active Directory servers, and an A record for linuxhost.domain,
and if you want a CNAME or A record for 'www.linuxhost.domain'. This
solution is more of a pain in the neck for the Active Directory
server, but it's very stable. And in environments where there are only
a few critical Linux servers and hundreds or thousands of Windows
desktops for whom expiring DNS is so useful, it's workable.

Mind you, I personally think that expiring DNS was a bad idea from the
start and should never have been permitted, but it has its uses for
laptops that hop from domain to domain.
From: Wanna-Be Sys Admin on
Nico Kadel-Garcia wrote:

> Some time back, I growsed here about the difficulty keeping a static
> Linux server's DNS information alive in a Windows Active Directory
> server

So, was this really a Linux issue/question?
--
Not really a wanna-be, but I don't know everything.