Erling Ringen Elvsrud wrote:
This is due to a limitation in CA. According to the CA-documentation the
hostname can be maximum 26 characters long, but in my case the
hostnames can exceed 35 characters!
I trust you've raise a bug against CA-unicenter. Even the unqualified
hostname can easily exceed 26 bytes: if I remember rightly the limit
imposed by the DNS is 63 bytes and then there's another limit for the
number of component which makes the maximum possible FQDN rather long.
The RHEL Deployment guide (system -> documentation -> deployment
guide) says this about HOSTNAME:
"HOSTNAME=<value>, where <value> should be the Fully Qualified Domain
Name (FQDN) such as hostname.example.com, but can be whatever hostname
is necessary."
That makes a lot of sense.
Do you know if the RH practice of storing the fqdn in HOSTNAME is
common practice and sensible? or is it better to use only the hostname
in HOSTNAME and add the fqdn to /etc/hosts?
Different distros do different things. SLES always sets the hostname to
the unqualified name so that $(hostname) and $(hostname -s) are the same
and, although it's been a while since I did a SLES installation, I seem
to remember that it doesn't allow you to have anything with dots in it
for the hostname.
At one stage, a really good way to mess up a Red Hat system was to use
an unqualified name as the hostname but I think the last of the bugs
that assume $(hostname) == $(hostname -f) were fixed a while ago. You
should exercise a little caution though: various scripts may well assume
that $(hostname) is fully qualified -- I know I had to go around a fix a
bunch of mine about a year ago.
jch
_______________________________________________
rhelv5-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv5-list
