5.1.1 Migration Versus
Pristine Installation
5.1.1 Migration Versus
Pristine Installation
This section briefly discusses the CID concept and its implementation.
First, let us describe briefly what CID is. Some of the CID architecture's primary goals are to:
Software that is capable of being distributed and configured through the LAN is called CID-enabled. CID-enabled products can be configured and installed remotely on LAN-attached clients with limited or no interaction required locally at each client. CID-enabled software must be able to use a response file to determine which options to install, use a redirected drive, and also log results of the installation to a file.
In the context of migration, then, the purpose of CID is to enable remote, unattended (or at least lightly attended) software installation. This is achieved by providing answers to installation questions through response files and the actual procedures required for the installation itself.
Response files provide predefined responses to any prompts normally aimed at the user during the installation or configuration process. This allows user interaction with the installation process to be bypassed.
Response files are product-specific ASCII files that contain sequences of keyword-value pairs. They are interpreted during the installation and configuration process of a product by the installation (and configuration) program.
CID also supports the capability to install from a drive other than A:. This drive could be an alternate drive on the target system, a redirected drive on a LAN or other network, or some other device that appears to the operating system as a logical drive, such as a CD-ROM device.
The workstation that uses a remote (redirected) drive is known as the client or redirector, and the workstation that provides a remote (redirected) drive is known as the server, software distribution server, or code server.
The client workstations will access the drive on the server where the product images reside and will perform the installation. Depending on the method of communications used, there are different ways to connect to a code server. In most cases, the redirected drive will be accessed through a Local Area Network (LAN).
Before starting a CID installation, a code server is required. A code server is the system that contains the source files (or installation diskette images) to be used during the installation or maintenance process. It also contains the response files for each product and an area for log files produced by the installation routines.
Aside from containing the files and programs required for installation, in some environments, the server may also initiate and/or manage the installation of code in one or more of its clients. In this case, the code server provides more functions than just file sharing.
Some software distribution managers, such as NetView DM/2, implement, for example, functions to schedule or remotely invoke software installation processes. Others, such as LAN CID Utility (LCU), do not have a scheduling capability.
The features of the particular software distribution manager also determine within which system environments it is able to drive the automated installation process. Additionally, these features decide whether this process is required to be invoked locally (at the target workstation) or whether it may be invoked remotely (at the client or server) or at the central site.
A system that is being installed, configured, or maintained, is called the client. It utilizes the resources of a code server to gain access to the files and programs it requires, and in some cases, will operate under the direction of a software distribution manager.
In unusual cases, where only very few machines must be installed, or if no network connection exists, an image of the code server can be provided on CD-ROM or on the local hard disk.
5.3 LAN CID
Utility