Part III. IBM System z Architecture - Installation and Booting

The installation process assumes that you are familiar with the IBM System z and can set up logical partitions (LPARs) and z/VM guest virtual machines. For additional information on System z, refer to http://www.ibm.com/systems/z.

For installation of Red Hat Enterprise Linux on System z, Red Hat supports DASD and FCP storage devices.

Before you install Red Hat Enterprise Linux, you must decide on the following:

Decide whether you want to run the operating system on an LPAR or as a z/VM guest operating system.
Decide if you need swap space and if so how much. Although it is possible (and recommended) to assign enough memory to z/VM guest virtual machine and let z/VM do the necessary swapping, there are cases where the amount of required RAM is hard to predict. Such instances should be examined on a case-by-case basis. Refer to Section 23.15.5, "Recommended Partitioning Scheme".
Decide on a network configuration. Red Hat Enterprise Linux 6 for IBM System z supports the following network devices:
- Real and virtual Open Systems Adapter (OSA)
- Real and virtual HiperSockets
- LAN channel station (LCS) for real OSA

You require the following hardware:

Disk space. Calculate how much disk space you need and allocate sufficient disk space on DASDs ^[9] or SCSI ^[10] disks. You require at least 2 GB for a server installation, and 5 GB if you want to install all packages. You also require disk space for any application data. After the installation, more DASD or SCSI disk partitions may be added or deleted as necessary.
The disk space used by the newly installed Red Hat Enterprise Linux system (the Linux instance) must be separate from the disk space used by other operating systems you may have installed on your system.
For more information about disks and partition configuration, refer to Section 23.15.5, "Recommended Partitioning Scheme".
RAM. Acquire 1 GB (recommended) for the Linux instance. With some tuning, an instance might run with as little as 512 MB RAM.

18.2. Overview of the System z Installation Procedure

You can install Red Hat Enterprise Linux on System z interactively or in unattended mode. Installation on System z differs from installation on other architectures in that it is typically performed over a network and not from a local DVD. The installation can be summarized as follows:

Booting (IPL) the installer
Connect with the mainframe, then perform an initial program load (IPL), or boot, from the medium containing the installation program.
Installation Phase 1
Set up an initial network device. This network device is then used to connect to the installation system via SSH or VNC. This gets you a full-screen mode terminal or graphical display to continue installation as on other architectures.
Installation Phase 2
Specify which language to use, and how and where the installation program and the software packages to be installed from the repository on the Red Hat installation medium can be found.
Installation Phase 3
Use anaconda (the main part of the Red Hat installation program) to perform the rest of the installation.

Overview of the System z installation process

Figure 18.1. The Installation Process

18.2.1. Booting (IPL) the Installer

After establishing a connection with the mainframe, you need to perform an initial program load (IPL), or boot, from the medium containing the installation program. This document describes the most common methods of installing Red Hat Enterprise Linux 6 on System z. In general, you can use any method to boot the Linux installation system, which consists of a kernel (kernel.img) and initial ramdisk (initrd.img) with at least the parameters in generic.prm. The Linux installation system is also called the installer in this book.

The control point from where you can start the IPL process depends on the environment where your Linux is to run. If your Linux is to run as a z/VM guest operating system, the control point is the control program (CP) of the hosting z/VM. If your Linux is to run in LPAR mode, the control point is the mainframe's Support Element (SE) or an attached IBM System z Hardware Management Console (HMC).

You can use the following boot media only if Linux is to run as a guest operating system under z/VM:

z/VM reader - refer to Section 20.1.1, "Using the z/VM Reader" for details.

You can use the following boot media only if Linux is to run in LPAR mode:

SE or HMC through a remote FTP server - refer to Section 20.2.1, "Using an FTP Server" for details.
SE or HMC DVD - refer to Section 20.2.2, "Using the HMC or SE DVD Drive" for details

You can use the following boot media for both z/VM and LPAR:

DASD - refer to Section 20.1.2, "Using a prepared DASD" for z/VM or Section 20.2.3, "Using a prepared DASD" for LPAR
SCSI device that is attached through an FCP channel - refer to Section 20.1.3, "Using a prepared FCP-attached SCSI Disk" for z/VM or Section 20.2.4, "Using a prepared FCP-attached SCSI Disk" for LPAR
FCP-attached SCSI DVD - refer to Section 20.1.4, " Using an FCP-attached SCSI DVD Drive" for z/VM or Section 20.2.5, "Using an FCP-attached SCSI DVD Drive" for LPAR

If you use DASD and FCP-attached SCSI devices (except SCSI DVDs) as boot media, you must have a configured zipl boot loader. For more information, see the Chapter on zipl in Linux on System z Device Drivers, Features, and Commands on Red Hat Enterprise Linux 6.

18.2.2. Installation Phase 1

After the kernel boot, you will configure one network device. This network device is needed to complete the installation.

The interface you will use in installation phase 1 is the linuxrc interface, which is line-mode and text-based. (Refer to Chapter 21, Installation Phase 1: Configuring a Network Device.)

18.2.3. Installation Phase 2

In installation phase 2, you need to specify what language to use and where phase 3 of the installation program and the software packages to be installed from the repository on the Red Hat installation medium can be found. On System z, the installation sources are usually transferred from the DVD to a network server. Phase 3 of the installation program and the repository can be accessed in one of the following ways:

Over the network using one of the FTP, HTTP, HTTPS, or NFS protocols. A separate network server (FTP, HTTP, HTTPS, or NFS), which holds all the required installation sources, must be set up in advance. For details on how to set up a network server, refer to Section 19.1, "Preparing for a Network Installation".
Hard disk (DASD or a SCSI device attached through an FCP channel). You need to set up a disk that holds the required installation sources in advance. For details, Refer to Section 19.2, "Preparing for a Hard Drive Installation".
Through an FCP-attached SCSI DVD. This is handled automatically if booted from FCP-attached SCSI DVD.

The interface you will use in installation phase 2 is the loader, which provides a full-screen text-based interface with a blue background by default. For unattended installations in cmdline mode, the loader offers line-mode, text-based output. (Refer to Chapter 22, Installation Phase 2: Configuring Language and Installation Source.)

18.2.4. Installation Phase 3

In installation phase 3 you will use anaconda in graphical, text-based, or cmdline mode:

Graphical mode
This can be used through a VNC client (recommended) or through an X11 server. You can use your mouse and keyboard to navigate through the screens, click buttons, and enter text in fields.
Text-based mode
This interface does not offer all interface elements of the GUI and does not support all settings. Use this for interactive installations if you cannot use a VNC client or X11 server.
cmdline mode
This is intended for automated installations on System z. (Refer to Section 26.6, "Parameters for kickstart installations")

If you have a slow network connection or prefer a text-based installation, do not use X11 forwarding when logging in over the network and do not set the display= variable in the parameter file (refer to Section 26.4, "VNC and X11 parameters" for details). In Red Hat Enterprise Linux 6 the text-based installation has been reduced to minimize user interaction. Features like installation on FCP-attached SCSI devices, changing partition layout, or package selection are only available with the graphical user interface installation. Use the graphical installation whenever possible. (Refer to Chapter 23, Installation Phase 3: Installing using anaconda.)

18.3. Graphical User Interface with X11 or VNC

To run anaconda with the graphical user interface, use a workstation that has either an X Window System (X11) server or VNC client installed.

You can use X11 forwarding with an SSH client or X11 directly. If the installer on your workstation fails because the X11 server does not support required X11 extensions you might have to upgrade the X11 server or use VNC.

To use VNC, disable X11 forwarding in your SSH client prior to connecting to the Linux installation system on the mainframe or specify the vnc parameter in your parameter file. Using VNC is recommended for slow or long-distance network connections. Refer to Section 28.2, "Enabling Remote Access to the Installation System".

Table 18.1, "Parameters and SSH login types" shows how the parameters and SSH login type controls which anaconda user interface is used.

Table 18.1. Parameters and SSH login types

Parameter	SSH login	User interface
none	SSH without X11 forwarding	VNC or text
`vnc`	SSH with or without X11 forwarding	VNC
none	SSH with X11 forwarding	X11
`display=IP/hostname:display`	SSH without X11 forwarding	X11

18.3.1. Installation using X11 forwarding

You can connect a workstation to the Linux installation system on the mainframe and display the graphical installation program using SSH with X11 forwarding.

You require an SSH client that allows X11 forwarding. To open the connection, first start the X server on the workstation. Then connect to the Linux installation system. You can enable X11 forwarding in your SSH client when you connect.

For example, with OpenSSH enter the following in a terminal window on your workstation:

ssh -X [email protected]

Replace linuxvm.example.com with the hostname or IP address of the system you are installing. The -X option (the capital letter X) enables X11 forwarding.

18.3.2. Installation using X11

The direct connection from the X11 client to an X11 server on your local workstation requires an IP connection from your System z to your workstation. If the network and firewalls prevent such connections, use X11 forwarding or VNC instead.

The graphical installation program requires the DNS and hostname to be set correctly, and the Linux installation system must be allowed to open applications on your display. You can ensure this by setting the parameter display=workstationname:0.0 in the parameter file, where workstationname is the hostname of the client workstation connecting to the Linux image. Alternatively, you can set the display environment variable and run loader manually after having logged in with SSH as user root. By default you log in as user install. This starts the loader automatically and does not allow overriding the display environment variable.

To permit X11 clients to open applications on the X11 server on your workstation, use the xauth command. To manage X11 authorization cookies with xauth, you must log in to the Linux installation system using SSH as user root. For details on xauth and how to manage authorization cookies, refer to the xauth manpage.

In contrast to setting up X11 authorizations with xauth, you can use xhost to permit the Linux installation system to connect to the X11 server:

xhost +linuxvm

Replace linuxvm with the hostname or IP address of the Linux installation system. This allows linuxvm to make connections to the X11 server.

If the graphical installation does not begin automatically, verify the display= variable settings in the parameter file. If performing an installation under z/VM, rerun the installation to load the new parameter file on the reader.

18.3.3. Installation using VNC

Using VNC is recommended for slow or long-distance network connections. To use VNC, disable X11 forwarding in your SSH client prior to connecting to the temporary Linux installation system. The loader will then provide a choice between text-mode and VNC; choose VNC here. Alternatively, provide the vnc variable and optionally the vncpassword variable in your parameter file (refer to Section 26.4, "VNC and X11 parameters" for details).

A message on the workstation SSH terminal prompts you to start the VNC client viewer and provides details about the VNC display specifications. Enter the specifications from the SSH terminal into the VNC client viewer and connect to the temporary Linux installation system to begin the installation. Refer to Chapter 31, Installing Through VNC for details.

18.3.4. Installation using a VNC listener

To connect from your temporary Linux installation system to a VNC client running on your workstation in listening mode, use the vncconnect option in your parameter file, in addition to the options vnc and optionally vncpassword. The network and firewalls must allow an IP connection from your temporary Linux installation to your workstation.

To have the temporary Linux installation system automatically connect to a VNC client, first start the client in listening mode. On Red Hat Enterprise Linux systems, use the -listen option to run vncviewer as a listener. In a terminal window, enter the command:

vncviewer -listen

Refer to Chapter 31, Installing Through VNC for details.

18.3.5. Automating the Installation with Kickstart

You can allow an installation to run unattended by using Kickstart. A Kickstart file specifies settings for an installation. Once the installation system boots, it can read a Kickstart file and carry out the installation process without any further input from a user.

On System z, this also requires a parameter file (optionally an additional configuration file under z/VM). This parameter file must contain the required network options described in Section 26.3, "Installation network parameters" and specify a kickstart file using the ks= option. The kickstart file typically resides on the network. The parameter file often also contains the options cmdline and RUNKS=1 to execute the loader without having to log in over the network with SSH (Refer to Section 26.6, "Parameters for kickstart installations").

For further information and details on how to set up a kickstart file, refer to Section 32.3, "Creating the Kickstart File".

18.3.5.1. Every Installation Produces a Kickstart File

The Red Hat Enterprise Linux installation process automatically writes a Kickstart file that contains the settings for the installed system. This file is always saved as /root/anaconda-ks.cfg. You may use this file to repeat the installation with identical settings, or modify copies to specify settings for other systems.

^[9]Direct Access Storage Devices (DASDs) are hard disks that allow a maximum of three partitions per device. For example, dasda can have partitions dasda1, dasda2, and dasda3.

^[10]Using the SCSI-over-Fibre Channel device driver (zfcp device driver) and a switch, SCSI LUNs can be presented to Linux on System z as if they were locally attached SCSI drives.

Installation Guide