Installation in an LPAR
The following terms are used in this document.
RHEL - Red Hat Enterprise Linux
SLES - SUSE Linux Enterprise Server
SE - Support Element
HMC - Hardware Management Console
Disclosure - This document is based on Red Hat Enterprise Linux 6.0 Installation Guide and SLES Prep zSeries (SLES10 SP4) Document.
Installation of Linux on an IBM System z LPAR
1. Before Installation
1.1 Pre-Installation 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. Configure your IBM System z LPAR to start in ESA/S390 or LINUX-only mode with an appropriate activation profile and IOCDS. For additional information on System z, refer to http://www.ibm.com/systems/z.
1.2. Overview of the System z Installation Procedure You can install Linux on System z interactively or in unattended mode. Installation on System z differs from installation on other architectures, since it is typically performed over a network and not from a local DVD. This document describes interactive network installation on an LPAR .
1.You need access to the Hardware Management Console or the Support Element to start Linux installation on LPAR.
2.An SSH client. If you use Linux on your workstation, ssh is available. For Windows, you can use SSH -client 'Putty'.
3.A VNC client. You can use the VNC Viewer or a similar program for Windows and Vinagre for Linux. These tools will be needed to connect to the LPAR and perform the installation.
4.An installation repository which is accessible to the installation Linux on LPAR by nfs, HTTP or FTP.
5.At least two devices: a DASD (Direct Attached Storage Device = a disk) and a network connection device.
2. Linux Installation Installation of Linux on System z includes the following stages.
1. IPL (Initial Program Load = booting) the installer Linux must be loaded into main memory. Because the System z server can not use a built-in CD or DVD drive as other platforms, there are several other ways to do this task. See below "Booting (IPL) the installer" for details.
2. Installation Phase 1 - Initial Network Setup Since installation will be done over the network, one of the first steps will be to set up an initial network device. A question and answer script will ask for the network details like IP address, netmask or gateway server. This network device is then used to connect to the installation server via SSH or VNC. This gets you a full-screen mode terminal or graphical display to continue installing as on other architectures.
3. Installation Phase 2 - Connect to the installation source Once the network connection has been established it will be used to access the install server containing the installation repository. To establish the connection to the installation repository, specify the IP address of the installation server, access data ,and the corresponding directory .
4. Installation Phase 3 - Starting the distribution specific installer Different Linux Distributions use different installers. For RHEL use anaconda (the main part of the Red Hat installation program) to perform the rest of the installation. For SLES use Yast. Typically defining partitioning scheme and software package selection is done in this stage.
* Booting (IPL) the installer There are different ways to IPL the installation Linux in an LPAR. The preferred way is to use the Load from Removable Media or server feature of the SE or HMC.
IPL from an FTP server IPL over network using FTP requires the FTP server IP address, access data, and the target directory. The target directory should have a plain text file with the suffix .ins. This file is provided by the distributor and contains the mapping of each component of the installation Linux (kernel, ramdisk, parameter file, ...) to the storage address at which the component should be loaded in main memory. Here an example of such an .ins file:
- cat generic.ins
- minimal lpar ins file
IPL from DVD-ROM Here the installation Linux will be loaded from SE or HMC built-in DVD drive. Since the LPAR does not have access to this drive, it can only be used to load the installer, using a SE or HMC load function (Load from Removable Media or server). Once the installer has been started the remaining installation must be done via the network.
IPL from an FCP-attached SCSI DVD An alternative way to load the installation Linux into memory is to use a SCSI CD or DVD drive. Because System z does not accept SCSI attachments directly, SCSI devices must be attached through FCP (Fibre Channel Protocol). This requires an FCP SAN (Storage Area Network) environment including an FCP-to-SCSI bridge. The installation Linux will be loaded then using the SE or HMC SCSI IPL functionality. To load Linux, specify the device number of the FCP adapter, the target WWPN (world-wide port number), and the target FCP LUN (logical unit number).
3.References Refer to installation guide provided by your Linux distribution for more details.