• Install NX (info from this page: http://www.nomachine.com/download-package.php?Prod_Id=355 )

  CentOS 5.5 64bit

  yum install wget #must be logged on as root
NXVER=3.4.0
CVER=7
NVER=13
SVER=12
ARCH=x86_64
NXURL=http://64.34.161.181/download
wget $NXURL/$NXVER/Linux/nxclient-$NXVER-$CVER.$ARCH.rpm
wget $NXURL/$NXVER/Linux/nxnode-$NXVER-$NVER.$ARCH.rpm
wget $NXURL/$NXVER/Linux/FE/nxserver-$NXVER-$SVER.$ARCH.rpm
rpm -i nxclient-$NXVER-$CVER.$ARCH.rpm
rpm -i nxnode-$NXVER-$NVER.$ARCH.rpm
rpm -i nxserver-$NXVER-$SVER.$ARCH.rpm

Working with yum, an intro from IBM: http://www.ibm.com/developerworks/linux/library/l-lpic1-v3-102-5/index.h...

Note that this now only works for the 64 bit version of CentOS

Here is Redhat's own guide: http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.5/html/Virtu...
and for bridging:
http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.5/html/Virtu...

system-config-securitylevel #set to permissive, not disabled
yum install kvm

This is very "quick and dirty" using the CentOS gui tools as much as possible

-install server & open firewall to samba
-create office and supervisor groups and a /home/common folder
-create all users on server and make them part of office, (+supervisor)
-add all users to Samba with Windows names and passwords
-share all home drives and set user access on shares as appropriate
-set permissions in the directory at the root of each share, appropriate to users
-chmod g+w -R * #at the top level of each shared directory so groups can read and write

The process is very straightforward, with only one minor complication: the MBR and GRUB must be set up on both drives for alternate boot if one drive fails.
1. First setup a boot raid partition (100MB) on each drive
-Then combine them into one RAID1 ext3 partition mounted at /boot
2. Make two 4096GB raid partitions for the swap drive (adjust size as needed)
-Combine them into RAID1 of type swap (not mounted)
3. Next use the rest of the disks to make big raid partitions
-Combine them into RAID1 of type LVM
-Create a Physical file group from the RAID


yum install subversion mod_dav_svn
mkdir -p /var/www/svn/test #place to keep repository
svnadmin create --fs-type fsfs /var/www/svn/test #make a test one
chown -R apache.apache /var/www/svn
vim /etc/httpd/conf/httpd.conf #make sure the following two lines are there
LoadModule dav_svn_module modules/mod_dav_svn.so
LoadModule authz_svn_module modules/mod_authz_svn.so
#
vim /etc/httpd/conf.d/subversion.conf #and add the following

<Location /svn>
   DAV svn
   SVNParentPath /var/www/svn
   # Limit write permission to list of valid users.

cat /etc/redhat-release; rpm -q centos-release #should be 5.4
yum clean all
yum update
cat /etc/redhat-release; rpm -q centos-release #should be 5.5

Please be aware that I was not 100% successful using this method:
Note that you cannot install Fedora 10 on RHEL/CentOS 5.2 using virt-install since the image structure has changed. You will need to use the new virt-install in 5.3

• Prepare LVM with:
  lvcreate -L 5G -n vm_fred vg
  lvcreate -L 1G -n vm_fred_swp vg
  /sbin/mkswap /dev/vg/vm_fred_swp
• for installation under RHEL/CentOS (note that 32 bit domU's can only run under 32 bit dom0's)
• Set up the install location with:
  

The best resources I found were:
How to create a modeline: http://xtiming.sourceforge.net/cgi-bin/xtiming.pl
Some general tips: http://www.x.org/wiki/FAQVideoModes
A complex example: http://www.mail-archive.com/fedora-list@redhat.com/msg12682.html
Some detailed debugging tips: http://redhatcat.blogspot.com/2007/09/xorg-for-sharp-aquos-and-nvidia-88...

Using the last link, I found that there were three video ports on the card (DVI0, VGA0 and VGA1) and all my settings were being applied to the wrong one.

Here is the first part of redhatcat, on how to do debugging: