Network Utility User's Guide

Chapter 10. Software Maintenance

This chapter describes what you need to know to receive and install fixes for Network Utility software problems, and to upgrade to new software releases containing new function.

This information includes:

• How the software is named and packaged
• How to download new software versions from the World Wide Web
• How to load software onto Network Utility
• How to call for product service and support

Software Versions and Packaging

Version Naming

The software that operates Network Utility is called Multiprotocol Access Services, or MAS. MAS also operates the IBM 2216-400, but there are different, separate packages of MAS for each product. The MAS packages for Network Utility are characterized by:

• Preset configuration defaults, to tune Network Utility for its intended applications.

• Specialized function packaging oriented toward the key uses of Network Utility. For example, some of the general multiprotocol routing functions of the 2216-400 such as IPX, Appletalk, Banyan Vines, and DECNet, are not available in the Network Utility packages.

Specific levels of MAS are identified by the following numbers:

Version

A new function release occasionally requires a new version number. Sometimes this is related to a price increase, but it could also be related to a shift in how IBM is distributing the software. A new version number does not mean that the release has any more new function than a release that only has a new release number.

Release

This number changes with every new function release.

Modifier

This number signals a new function release that is a small change to a larger base new function release. It follows the decimal point in the format "MAS Vv.r Mod m PTF p".

PTF

This number represents a maintenance level, described below.

The initial code base for Network Utility is: MAS V3R1.0 PTF 1. Because IBM is using the same release numbering as for the 2216-400 packages of MAS, you can easily correlate the function and maintenance level of software on the two products.

To see the software level of the code that is actively running in your Network Utility, move to the base talk 5 menu and type c (for "configuration"). The software version part of the output of this command uses the format "MAS Vv.r Mod m PTF p".

To see the software level of the code loads on the Network Utility hard disk, move to the base talk 6 menu, type boot to enter the boot Config subprocess and then type describe.

Maintenance Levels

When you access the World Wide Web pages that contain recent versions of Network Utility software, you see some of the following terms for different maintenance levels of the Network Utility packages:

GA Level

The software level first made "generally available" to IBM customers. This is the level initially shipped on the hard disk of new Network Utility boxes. GA level software undergoes an extensive product-level and system-level test before it is released. General availability normally corresponds to a new Version or Release of the software (the initial release of Network Utility in a PTF is an exception to this rule).

PTF

A major maintenance release ("program temporary fix") that accumulates a large number of fixes and undergoes a regression test of most major software functions. After a Release has been deployed for some time, IBM will typically start to ship a stable PTF on the hard disk of new products.

EPTF

A small maintenance release ("emergency PTF") that comes out on a more frequent basis, involves fewer fixes, and undergoes a regression test of the specific areas affected by the fixes.

PTFs and EPTFs are cumulative, in that each supersedes all previous PTFs and EPTFs. You only need to install the latest PTF or EPTF to obtain all previous fixes.

Feature Packaging

There are two feature packages of Network Utility software, corresponding to the two different models of Network Utility:

Based on the model you purchased, your Network Utility comes preloaded with the proper software package in both banks of the hard disk. When you load a new maintenance level of software, you load the same package that is already on the Network Utility.

Note that there is only one version of the Configuration Program, and it supports the software functions in all the software packages. If you configure functions that are not supported in the particular software package you have on the router, the router software ignores that part of the configuration.

From the command line, you cannot configure or monitor software functions that are not present in the software load you are running.

Getting Web Access to the Software

To update your Network Utility software, you must first download the appropriate maintenance level from the World Wide Web. To find the new software, start with the main Network Utility product page at:

http://www.networking.ibm.com/networkutility

Click on Support and then Downloads to reach the following information and links:

• General information about accessing the software

• Detailed procedures for downloading and installing the software

• Links to the latest maintenance levels of the Configuration Program, with associated README files

• Links to the latest maintenance levels of MAS, with associated PTF or EPTF content files

When you follow the links to a particular maintenance level of the Configuration Program, you can access packed binary versions of the 2216/Network Utility Configuration Program for each of its supported operating systems. Anyone can download these files. The associated README file gives instructions for unpacking and installing the new version of the Configuration Program.

When you follow the links to a particular maintenance level of MAS, you can access compressed packed binary versions of each of the Network Utility software features listed above.

You need an IBM Networking customer id and password to be able to download these files. You create the id and password yourself by registering on the Web, and you can immediately use it to download files. This id and password spans multiple IBM Networking products, and allows you to subscribe for e-mail notification of product updates. If you do not have one, the Web pages will take you to the registration page the first time you download a Network Utility code package.

Downloading and Unpacking Files

The Web pages for downloading a particular MAS maintenance release contain files for each of the supported software features. Each file contains a complete set of software for Network Utility. When you install a maintenance level of Network Utility software, you completely replace all the existing software with the new level.

To download the software in a particular file and transfer it to the router, you:

1. Use your Web browser to download the complete file in binary to your workstation.

2. Transfer the file to the workstation from which you will load it into the router. This is called the server workstation because it acts as a file server to the router. You can use FTP or any other file transfer method for this step.

3. At the server workstation, unpack the single downloaded file into a number of router software files. These files are called load modules and have the file extension "ld" (or "LD" if your file system does not support mixed case).

4. Using TFTP or Xmodem, you transfer the load modules into the router.

Depending on the MAS release, the Web page may contain two files for each software feature, each constructed by a different packing utility. Choose the version that your server workstation software can unpack. Normally you would choose as follows:

When you unpack the router software, make sure that all ".ld" files are in the same directory, and have file system permissions to give appropriate read access. Do not change the names of any of the .ld files. Do not mix files between different Network Utility feature packages, or between different maintenance levels of the same package. Keep each package distinct and separate with a different path name on your server workstation.

Loading New Operational Code

Operational code (op-code, for short) is the software that runs the normal packet forwarding and system services functions of Network Utility. Op-code includes the base operating system, protocols, features, diagnostics, and the command-line interface code. The vast majority of software changes in PTFs and EPTFs are changes to the operational code.

To load and activate new operational code, you must:

1. Transfer the unpacked load modules from your server workstation into one of the two op-code banks on the Network Utility hard disk.

2. Set the router to boot from the bank with the new op-code.

3. Reboot the router, or schedule it to reboot at a later date and time.

Table 10-1 summarizes the different ways you can transfer operational code from a server workstation to a Network Utility hard disk. Which method you choose depends on how you can attach the workstation to the router, what software you have on your workstation, and your own preferences. Here are some important points to consider:

• The size of all the .ld files combined is over 10 MB. If you can possibly use a LAN or network interface instead of the service port or modem, you should do so to avoid hours of file transfer time.

• The TFTP-based methods from the op-code and firmware automatically transfer all .ld files in a single operation. With Xmodem, you must manually specify the name of each of the roughly 20 .ld files that make up a software load.

Table 10-1. Loading Operational Code

Using the Operational Code

As Table 10-1 shows, the transfer procedures you can initiate from the op-code all use TFTP as the file transfer protocol.

Using TFTP

The op-code procedure for using TFTP to transfer op-code and firmware files to a Network Utility hard disk is:

1. Configure the IP addresses you will be using.

   If you are using a standard network interface including an Ethernet or Token-Ring LIC, use the Configuration Program or talk 6 to configure an IP address for the interface in the normal way. (From talk 6, you use add address in the IP subprocess.) Activate this configuration change before proceeding.

   If you are using the PCMCIA EtherJet card, use system set ip to set the following addresses:

   • IP address: the IP address for the EtherJet card
   • Netmask: the mask for the subnet attached to the EtherJet card
   • Gateway address: the IP address of any intermediate router to reach the TFTP server workstation, or the IP address of the workstation itself if there is no intermediate router.

   If you are using SLIP, you cannot change the IP addresses but must use those given in Table 10-1.

2. Transfer the opcode and firmware files.

   From the * prompt, follow this sequence:

     *t 6
     Config>boot
     Boot configuration
     Boot config>tftp get load mod
   

   Respond to the prompts as follows:

   • Server IP address: Put the address of the TFTP server workstation.

   • Remote directory: Put the path name to the directory on the server workstation where all the .ld files are. Use slashes in the direction expected by your server. Uppercase versus lowercase matters only if it matters to your server.

   • Destination bank: Select bank A or bank B. You cannot select the currently active bank.

   Based on the server IP address and the configured Network Utility interface IP addresses, the router selects which of its interfaces to use to reach the server. The router gives success or failure status messages as appropriate.

3. Place your configuration file into the target bank

   Transfer the configuration file you want into a position in the bank where you just placed the new code load. If the new code load is a new MAS release, see Migrating a Configuration to a New MAS Release for important background on this step.

   • If your new code load is not a new MAS release, or you use only the command-line interface to configure your Network Utility, use the copy config command to copy your current configuration where the new load can pick it up.
   • If your new code load is a new MAS release and you use the Configuration Program at all, use the Configuration Program to upgrade your configuration. Then use the command tftp get config (or any of the other methods described in Loading New Configuration Files) to transfer the upgraded configuration to the target bank.

4. Reboot or schedule a reboot

   To activate the new load immediately, use the following procedure, starting from the Boot config> prompt:

   1. Use the set command to select the bank you just loaded to boot next, and to select the configuration you just copied or transferred.
   2. Press Ctrl-p and then type reload to reboot the router

   To activate the new load later, type timedload activate from the Boot config> prompt to select the bank and configuration and to specify the date and time for the router to reboot. You can answer "no" to the questions about loading the bank, because you already did this step.

See the MAS Software User's Guide chapter "Configuring Change Management" for more information on the commands in the above procedure.

Using the Firmware

As Table 10-1 shows, you can use either Xmodem or TFTP from the firmware to transfer op-code to the Network Utility hard disk. Xmodem is not recommended because modem speeds are too slow for these large op-code files and Xmodem requires regular interaction. TFTP over LAN interfaces is the preferred transfer method when you are working from the firmware. Nevertheless, this section summarizes all the possible procedures in case you need to use them.

Using Xmodem

The firmware procedure for using Xmodem to transfer op-code and firmware files to a Network Utility hard disk is:

1. Access the firmware main menu using the procedures described in Boot Options: Fast Boot and Reaching Firmware.

2. Make the following sequence of menu selections:
   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 12, "Change Management"
   3. Change Management Software Control: Option 12, "XMODEM software"
   4. Select type: "Load Image"
   5. Select bank: choose Bank A or Bank B

   The firmware tells you when to start the file transfer.

3. Go to your terminal emulation package and start the transfer of the file LML.ld from your workstation server.

4. After transferring LML.ld, you must transfer every other ".ld" module on your workstation server, one by one. LML.ld must be first, but after that the order does not matter. You must include Firm.ld.

   When the file transfer begins, the status of the bank changes to CORRUPT, to indicate that it does not contain a complete valid code load. When the Network Utility has received the last load module, the status of the bank changes to AVAIL. You can verify that this has happened using option 7, "List Software", from the firmware Change Management menu.

5. Boot the router using the op-code you just loaded .

   Use Option 9 "Set Boot Information" to select the new op-code bank (and configuration) to boot from. Press Esc to reach the main menu, then F9 to boot the Network Utility with the new op-code.

Using TFTP

The firmware procedure for using TFTP to transfer op-code and firmware files to a Network Utility hard disk is:

1. Access the firmware main menu using the procedures described in Boot Options: Fast Boot and Reaching Firmware.

2. Configure the IP addresses you will be using:

   Follow the menu sequence:

   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 11, "Remote Initial Program Load Setup"
   3. Network Parameters: Option 1, "IP Parameters"

   Set the following addresses:

   • Client IP address: an IP address for the Network Utility LAN card. This is a temporary address that need not be related to the router operational address for that interface.
   • Server IP address: the IP address of the workstation's LAN adapter
   • Gateway IP address: the IP address of any intermediate router, or repeat the workstation's IP address if there is none
   • Netmask: the mask for the subnet attached to the Network Utility LAN card

3. Initiate the transfer through these menu selections:
   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 12, "Change Management"
   3. Change Management Software Control: Option 10, "TFTP software"
   4. Select Type: "Load Image"
   5. Select Bank: choose Bank A or Bank B
   6. Select Load Type: "Modules"

4. Enter the path on your workstation to the directory with all the load modules.

5. If prompted, select the interface through which you want the firmware to do the file transfer.

   The firmware transfers each load module in turn and gives status messages. On completion, you will be back at the Change Management menu.

6. Boot the router using the op-code you just loaded .

   Use Option 9 "Set Boot Information" to select the new op-code bank (and configuration) to boot from. Press Esc to reach the main menu, then F9 to boot the Network Utility with the new op-code.

Upgrading Firmware

Introduction

Firmware is low-level software that drives the power-on and boot logic of Network Utility. It resides in nonvolatile flash memory rather than on the hard disk, so in the event of a failure such as corruption of your operational software load on disk, you can retrieve new software or configuration files and get back up and running. To upgrade the firmware means to write a new version of it to flash, replacing the old version.

You need to upgrade the firmware under two conditions:

1. IBM ships a PTF or EPTF that you need to fix a problem, and that PTF or EPTF requires a firmware upgrade. The documentation associated with each PTF or EPTF states whether firmware upgrade is required or not.

2. You want to install a new MAS functional release. Moving to a new release almost always requires a firmware upgrade.

On the Network Utility code download Web pages, there are no separate files containing new versions of the firmware. Instead, the firmware is one of the load modules packed inside the .zip and .tar files along with the operational code load modules. The firmware load module has the file name "Firm.ld". Every PTF and EPTF contains a new Firm.ld file, even if that file's contents are the same as for an older maintenance level.

When you follow the procedures described in Downloading and Unpacking Files and in Loading New Operational Code, you are downloading a new version of firmware from the web and transferring it to Bank A or Bank B of your hard disk. Placing Firm.ld into a disk bank and rebooting from that bank has absolutely no effect on the active firmware, which is running from flash memory. In order to upgrade to new firmware, you must write the new firmware to flash memory.

Procedure Overview

There are two general methods for downloading new firmware from the Web and getting it into flash memory on Network Utility. The recommended method is to do the firmware upgrade in conjunction with installing new operational code, as follows:

1. Download the new maintenance level of both op-code and firmware from the Web to a local server, as described in Downloading and Unpacking Files.

2. Transfer the new op-code and firmware ".ld"s to a bank in the Network Utility hard disk, using one of the TFTP or Xmodem procedures described in Loading New Operational Code.

3. Write the copy of Firm.ld that is now already on disk to flash memory, using one of the procedures described in Local Disk Procedures.

In addition to the recommended method, you can also independently transfer only the firmware into the Network Utility, and write it to flash without also transferring and activating operational code. You do this as follows:

1. Download the new maintenance level of both op-code and firmware from the Web to a local server, as described in Downloading and Unpacking Files. There is no way to download only the firmware from the Web, because it is packaged with operational code.

2. Transfer only Firm.ld to a non-bank location on the Network Utility hard disk, and write it to flash in the same procedure. You can use either Xmodem or TFTP for the file transfer, as described in File Transfer Procedures.

Transferring the firmware independently is not the recommended upgrade method, simply because it duplicates the Firm.ld file transfer you already did when you installed new operational code into bank A or B of the hard disk. The local disk procedures are faster and simpler.

Local Disk Procedures

Follow either of these procedures after you have already transferred a new set of operational code and firmware to hard disk bank A or B, to activate the firmware in that disk bank.

Using the Operational Code

Note: This procedure is available only when you are running MAS V3.2 or later operational code. If you are installing such a level for the first time, you must reboot to the new operational code before you can use this procedure to upgrade the firmware to the same level.

1. Type talk 6, then boot to reach the Boot Config subprocess.

2. Type update to start the firmware upgrade

3. When prompted, select the bank (A or B) into which you transferred the new level of operational code and firmware.

   There is also a "P" option, which you can use to rewrite flash with a valid firmware level previously saved on disk (not in bank A or B). You can select this if your flash becomes corrupted (perhaps the system lost power during a flash write) and you want to return to the previous firmware level.

4. The system writes flash memory with the new firmware level from the source location you specified, and creates a new "recovery image" (the one selected by "P") automatically as appropriate. Do not turn off Network Utility power while the firmware is being updated in flash memory.

The update command writes the new firmware level to flash memory, but the updated firmware does not begin to run until your next reboot. The easiest way to install a new maintenance level that requires firmware upgrade from the Boot config> prompt is therefore:

1. Use tftp get load m to transfer the new op-code and firmware to disk

2. Use update to write the new firmware to flash

3. Use copy to copy your configuration file to the new code bank

4. Use set to select the new code bank for the next boot

5. Type Ctrl-p then reload to reboot the Network Utility to use the new firmware and the new operational code at the same time.

Using The Firmware

Once you have transferred a new level of operational code and firmware to disk bank A or B, reboot to use the new operational code but stop in the old firmware to write flash memory with the new firmware as follows:

1. Access the firmware main menu using the procedures described in Boot Options: Fast Boot and Reaching Firmware.

2. Make the following sequence of menu selections:
   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 7, "Update System Firmware"
   3. F/W Update Options: Option 3, "Use a Local Image File"

   The firmware asks for a local file name. Enter one of:

   c:\sys0\firm.ld for Bank A

   c:\sys1\firm.ld for Bank B

3. Respond "yes" to the question "Do you want to continue?" The firmware then starts writing the new firmware to flash memory.

4. Wait and do not turn off the system while the update proceeds.

5. On completion, press Enter to restart the system. The new firmware boots up to the new operational code if you enabled autoboot in step 1 of this procedure.

File Transfer Procedures

Follow either of these procedures to transfer only the firmware from a local Xmodem or TFTP server to the Network Utility, and to activate that firmware. As shown in Table 10-2, you use the old firmware user interface in both procedures to initiate the file transfer over any of a number of connection types. As described in Procedure Overview, the local disk procedures may be faster than using these procedures.

Table 10-2. Loading Firmware

Using Xmodem

The firmware procedure for using Xmodem to transfer op-code and firmware files to a Network Utility hard disk is:

1. Access the firmware main menu using the procedures described in Boot Options: Fast Boot and Reaching Firmware.

2. Make the following sequence of menu selections:
   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 12, "Change Management"
   3. Change Management Software Control: Option 12, "XMODEM software"
   4. Select type: "Load Image"
   5. Select bank: choose Bank A or Bank B

   The firmware tells you when to start the file transfer.

3. Go to your terminal emulation package and start the transfer of the file LML.ld from your workstation server.

4. After transferring LML.ld, you must transfer every other ".ld" module on your workstation server, one by one. LML.ld must be first, but after that the order does not matter. You must include Firm.ld.

   When the file transfer begins, the status of the bank changes to CORRUPT, to indicate that it does not contain a complete valid code load. When the Network Utility has received the last load module, the status of the bank changes to AVAIL. You can verify that this has happened using option 7, "List Software", from the firmware Change Management menu.

5. Boot the router using the op-code you just loaded .

   Use Option 9 "Set Boot Information" to select the new op-code bank (and configuration) to boot from. Press Esc to reach the main menu, then F9 to boot the Network Utility with the new op-code.

Using TFTP

The firmware file transfer and upgradeprocedure using TFTP is:

1. Access the firmware main menu using the procedures described in Boot Options: Fast Boot and Reaching Firmware.

2. Configure the IP addresses you will be using:

   Follow the menu sequence:

   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 11, "Remote Initial Program Load Setup"
   3. Network Parameters: Option 1, "IP Parameters"

   Set the following addresses:

   • Client IP address: an IP address for the Network Utility LAN card. This is a temporary address that need not be related to the router operational address for that interface.
   • Server IP address: the IP address of the workstation's LAN adapter
   • Gateway IP address: the IP address of any intermediate router, or repeat the workstation's IP address if there is none
   • Netmask: the mask for the subnet attached to the Network Utility LAN card

3. Initiate the transfer with the following sequence of menu selections:
   1. System Management Services (main menu): Option 4, "Utilities"
   2. System Management Utilities: Option 7, "Update System Firmware"
   3. F/W Update Options: Option 1, "TFTP a Remote Image File"

   Enter the following file names:

   • Local file name: choose a name for a temporary file to be stored in the root directory of the Network Utility hard disk. Do not give a path name. Use a file name extension of 3 characters or fewer.

   • Remote file name: the path and file name (which must be "Firm.ld") of the firmware load module on your workstation. It should be in the directory where you unpacked the .zip or .tar file you downloaded from the Web.

   After you select the adapter and port the firmware should use, the router initiates the TFTP get operation.

4. After TFTP completes, respond "yes" to the question "Do you want to continue?" at the firmware console. The firmware then starts writing the new firmware to flash memory.

5. Wait and do not turn off the system while the update proceeds.

6. On completion, press Enter to restart the system. The new firmware boots up to the current operational code.

How to Call for Service and Support

If you bought your Network Utility from an IBM Business Partner or Reseller, contact that party to find out how to receive service and support.

If you bought your Network Utility from IBM, the following forms of assistance are available:

• Hardware or code problem service and support

  For telephone support:

  • Inside the U.S. - call 1 800 IBM-SERV (1 800 426-7378).
  • Outside the U.S. - contact your local IBM service representative for the phone number in your country.

  Before you call, have the machine type, model, and serial number from the backplate of the Network Utility available. If you have a software problem, you may need to have a TFTP server and Internet connection available to transfer a memory dump from the Network Utility and send it to IBM support personnel.

  You can also access IBM Service and Support via the World Wide Web at:

    http://www.networking.ibm.com/support/networkutility
  

  Select the Network Utility product to reach product technical hints, tips, FAQs, and code updates. In addition, you can subscribe to receive notification of future code updates.

• Configuration help and how-to questions for initial installations
  • Inside the U.S. - call 1 800 IBM-SERV (1 800 426-7378). This is a free service.
  • Outside the U.S. - contact your local IBM service representative. This service may not be free outside the U.S.

• Service and support contracts for network design, planning, or problem determination
  • Inside the U.S. - call 1-800-IBM-SERV (1 800 426-7378).
  • Outside the U.S. - contact your local IBM service representative.