This chapter provides background information about configuring Network Utility, including:
Chapter 3, Performing the Initial Configuration introduced the basic methods of configuring Network Utility and provided guidance on choosing between them (see Choosing Your Configuration Method). This chapter gives additional details about each method and discusses using both of them together.
For specific procedures and commands dealing with configuration files, see Chapter 7, Handling Configuration Files. Some common configuration tasks are described in Chapter 4, Quick Reference to the User Interface.
A Network Utility configuration is a collection of data items that control how the software operates, including such elements as:
When you boot up a Network Utility, the system reads its configuration from a file on its hard disk, and activates interfaces and protocols according to the information in that file. You create the file in one of two ways:
You type commands to create configuration data items in memory and then write the configuration to the Network Utility hard disk.
You create the configuration on the workstation and then transfer it to the Network Utility hard disk.
Once the system is up and running, you can use the command-line interface to make the following types of configuration changes:
The Network Utility hard disk is organized to contain two logical banks, one for each of two operational code (software) loads. This allows you to have the active code load in one bank, transfer a new load to the other bank, test it, and be able to back off to the original load if necessary. The two banks are referred to as Bank A and Bank B.
Each of the two banks has room for four configuration files. You can select to boot up the code load in Bank A with any of the 4 configuration files in Bank A. The same holds true for Bank B. To use a Bank A configuration file with the Bank B code load, you must first copy the Bank A configuration file to one of the four file positions in Bank B.
There are four ways to transfer a configuration file into a bank on the hard disk:
Use this command if you are configuring your Network Utility with the command-line talk 6 process, rather than with the Configuration Program.
Note: If the term "talk 6" is unfamiliar to you, use Chapter 5, A Guided Tour through the Command-Line Interface as a tutorial on the command-line interface.
You can transfer a configuration file in, whether the file was created from the Configuration Program or was previously transferred from this or another Network Utility.
You initiate the file transfer from the Configuration Program. This method is available only from the Configuration Program.
You initiate copies and other configuration file management operations from the Network Utility console under talk 6 in the boot subprocess.
See the section Loading New Configuration Files in Chapter 7, Handling Configuration Files for specifics on these operations.
To use the command-line interface, first bring up a local or remote console to a Network Utility. For details on how to do this and reach the * or Config (only)> prompt, refer to Chapter 2, Bringing Up a User Console.
If you have an active console at the * prompt, use talk 6 to access the Config process. If you are at the Config (only)>, the Config process is the only process available to you. From the Config process, you navigate menus and issue commands to configure interfaces and protocols, and write these changes to configuration files on the Network Utility hard disk.
In most cases, you use the command-line interface to configure only the Network Utility to which you are attached. But you could easily use a single Network Utility to produce configuration files to be transferred into other Network Utilities. Simply use the write command under talk 6 to store a configuration to a disk file, then use tftp put under the boot subprocess to transfer the file off the Network Utility. From then on, you have a file that can be loaded into the target Network Utility just as if it had come from the Configuration Program.
One option available only from the command line is Quick Config. As described in step 3, Quick Config guides you through an initial configuration of a subset of the protocols in Network Utility. The system asks you questions, instead of the normal mode where it waits for you to type commands.
The ability to dynamically activate configuration changes without rebooting the Network Utility is also exclusive to the command-line interface. Example: Dynamic Reconfiguration described using talk 5 to activate an IP address change made under talk 6. Dynamic Reconfiguration gives more background on the dynamic reconfiguration capabilities of Network Utility.
Network Utility is supported by the same graphical configuration program that you can use to configure the 2216-400. You run this program on a PC or workstation and send the configurations you produce to one or more 2216s or Network Utilities. A version of the 2216/Network Utility Configuration Program is available for each of the following operating systems:
IBM distributes major releases of the Configuration Program on CD-ROM and on the Web. Regular maintenance PTFs are available only on the Web. The Configuration Program User's Guide describes system requirements and contains instructions for installing and using the program.
When you start a new configuration with the Configuration Program, it presents a drop-down list for you to select whether the new configuration is for a 2216-400 or for a Network Utility. Your choice affects the following:
Configurations for the 2216-400 and Network Utility are not interchangeable.
The Configuration Program deals with three different formats of configuration files:
You use this format with the Configuration pull-down commands Open, Save, Save as, and Delete. Contents are software release-dependent; the Configuration Program automatically migrates data items when you do an Open.
You use this format when you want to create a file to transfer to the router, or when you want to read in a file you have transferred from a router.
You can write your configuration out into an ASCII flat file, make changes to it with a text editor, and read it back in.
There are two ways to transfer a configuration from the Configuration Program to a Network Utility:
Note that with each method, you transfer and activate an entire Network Utility configuration. There is no mechanism for the Configuration Program to dynamically send a small configuration change and activate it at the Network Utility without requiring a reboot of the Network Utility. You can only perform this type of dynamic reconfiguration using the command-line interface.
Features of the Configuration Program include:
When you use the Configuration Program's facility to send a configuration to a router, you can specify the date and time you want the router to restart and use the configuration.
You can create a list of target routers to receive configuration files, with the same or different configuration files, restart times, and so on, for each router.
You can use the workstation operating system command line, from which you start the Configuration Program, to automate configuration operations that are available in the program. You place arguments on the original command line or in an argument file, and the Configuration Program uses them to direct its operation.
From AIX, it is not necessary to have the operating system graphical environment (for example, Xwindows) installed to use this facility. You start the Configuration Program using the headless command.
You can use the Configuration Program to create and read configuration files in ASCII format. You can also convert configuration files from one format to another. An ASCII configuration file may be useful if you want to alter many configurations at one time without having to load the configurations into the Configuration Program. This feature is not intended to be used to create new configurations or to make major modifications to existing configurations.
The Configuration Program supports an extensive set of help files. Press F1 when you are positioned on any data item, and you will see a pop-up window describing the item and giving its default value and allowable range.
The ability to dynamically modify configuration parameters without
rebooting the Network Utility is available only from the command-line
interface. Table 6-1 summarizes the different ways you can change configuration
parameters from the command line, whether a change affects the running system
before a reboot, and whether the change is active following a reboot.
The column "Choose Write to Disk" indicates whether you issued the
write command from the main talk 6 menu to save the configuration
to disk, or requested a disk save after issuing the reload
command.
Table 6-1. Dynamic Reconfiguration Options
|
Method |
Choose Write to Disk | Affects
Running System |
Active After Reboot | ||
|---|---|---|---|---|---|
| Change in talk 6 | Yes | No (Note 1) | Yes | ||
| No | No (Note 1) | No | |||
| Change in talk 5 | Not applicable | Yes | No | ||
| Change in talk 6, then activate in talk 5 (Note 3) | Yes | Yes (Note 2) | Yes | ||
| No | Yes (Note 2) | No | |||
| |||||
As you can see, the general rule is that talk 6 changes become active following reboot or a talk 5 command to activate them. Talk 5 commands become active immediately but are lost upon reboot.
Not every configuration data item can be changed in all of the above ways. It depends on the part of the system (protocol, interface, and so on) to which a given data item belongs. For example, DLSw, SNMP, and ELS configuration all support most of the same commands in talk 6 and talk 5. You can make a change in either place depending on the permanence you want for the change. There is no talk 5 command to activate talk 6 changes, because a talk 5 command exists to make the same change.
In IP, however, there are no talk 5 commands corresponding to talk 6 commands. You use reset ip in talk 5 to activate the current talk 6 configuration. Interface reconfiguration is also activated using a single talk 5 command, because it involves taking the interface down and up.
See Configuring Physical Adapters and Interfaces for a few examples of common dynamic reconfiguration tasks involving adapters and interfaces.
If you decide to use only the command-line interface for configuration, you never need to use the Configuration Program. If you use the Configuration Program, you still need to use the command-line Config process for several reasons:
To use a combination of the Configuration Program and talk 6, you must keep the .CSF file at the Configuration Program synchronized with the configuration information at the Network Utility. A typical scenario might be:
See Chapter 7, Handling Configuration Files for specific procedures to transfer configuration files.
You will occasionally need to migrate your Network Utility to a new release of MAS, either for maintenance purposes or to pick up new function 9 release-specific information, you must also upgrade your configuration to the level of the MAS release you are installing.
If you use only the command-line interface to configure your Network Utility, you simply load and boot the new MAS release using one of the procedures in Chapter 10, Software Maintenance. When the new MAS release boots, it automatically adjusts your configuration to the new release level. These adjustments are made in memory and do not affect the disk copy of the configuration. You can issue the write command at the Config> prompt to save the upgraded configuration to disk. You can leave a copy of the old release config in the disk bank with the old level of code, in case you need to boot from the older release.
If you use the Configuration Program at all, even just occasionally, you must use the Configuration Program to upgrade your configuration. All new MAS releases are accompanied by a new release of the Configuration Program. Follow these steps to upgrade your configuration: