Configuration Methods - Configuration File
Configuration File
Configuration files are used to specify system-wide configuration values. This method is last in the order of precedence; that is, configuration file option values can be overridden by every other method of configuration (see Configuration Methods).
For most Universal Agent components, some options can be specified only in a configuration file, while other options can be overridden by individual command executions. The Reference Guide for each component identifies these options.
Universal Broker maintains the for all Universal Agent components, including itself. The components do not read their configuration files themselves (except for Universal Enterprise Controller, which does read its own configuration file).
At initial start-up, Universal Broker reads the configuration files of all components and places the configuration data in Universal Broker memory. When a component starts, it first registers with its local Universal Broker. As part of the registration process, Universal Broker sends that component's configuration data to the component. Thereafter, if a configuration file is modified, Universal Broker must be refreshed. This directs Universal Broker to re-read all component configuration files and update the configuration data in memory (see Configuration Refresh).
Universal Broker can operate in managed or unmanaged mode:
- In unmanaged mode, the configuration information for the various Universal Agent components can be modified either:
- Locally (either by editing the configuration files or, on Windows systems, via the Universal Configuration Manager).
- Remotely, via the Universal Enterprise Controller I-Management Console application.
- In managed mode, the configuration information for the various Universal Agent components is "locked down" and can be modified or viewed only via the I-Management Console.
(For information on unmanaged and managed modes, see Remote Configuration).
z/OS | Configuration files are members of a PDSE. The data set record format is fixed or fixed block with a record length of 80. No line numbers can exist in columns 72-80. All 80 columns are processed as data. |
---|---|
UNIX | Configuration files are regular text files on UNIX. The files can be edited with a text editor. See #Configuration File Syntax, below, for the configuration file syntax.
|
Windows | Although configuration files can be edited with any text editor (for example, Notepad), the Universal Configuration Manager application, accessible via the Control Panel, is the recommended way to set configuration options. Universal Configuration Manager provides a graphical interface and context-sensitive help, and helps protect the integrity of the configuration file by validating all changes to configuration option values. |
IBM i | The configuration files on IBM i are stored in a source physical file named UNVCONF in the UNVPRD520 library. The files can be edited with a text editor. See #Configuration File Syntax, below, for the configuration file syntax. |
HP NonStop | The configuration files on HP NonStop are stored as EDIT files, file code 101, within the $SYSTEM.UNVCONF subvolume. The files can be edited with the EDIT editor. See #Configuration File Syntax, below, for the configuration file syntax. |
Configuration File Syntax
Configuration files are text files that can be edited with any available text editor.
The following rules apply for configuration file syntax:
- Options are specified in a keyword / value format.
- Keywords can start in any column.
- Keywords must be separated from values by at least one space or tab character.
- Keywords are not case sensitive.
- Keywords cannot contain spaces or tabs.
- Values can contain spaces and tabs, but if they do, they must be enclosed in single ( ' ) or double ( " ) quotation marks. Repeat the enclosing characters to include them as part of the value.
- Values case sensitivity depends on the value being specified. For example:
- Directory and file names are case sensitive.
- Pre-defined values (such as yes and no) are not case sensitive.
- Each keyword / value pair must be on one line.
- Characters after the value are ignored.
- Newline characters are not permitted in a value.
- Values can be continued from one line to the next either by ending the line with a:
- Plus ( + ) character, to remove all intervening spaces.
- Minus ( - ) character, to preserve all intervening spaces between the end of the line being continued and the beginning of the continuing line.
- Ensure that the line continuation character is the last character on a line.
- Comment lines start with a hash (#) character.
- Blank lines are ignored.
Note
If an option is specified more than once in a configuration file, the last instance is used.