|
|
This chapter describes how to set up general switch features that apply to multiple switch interfaces, beginning with a configuration quickstart procedure, which introduces the configuration tasks. The following sections provided detailed information on how to complete the configuration tasks.
The quickstart procedure is provided as an overview and as a quick reference for those who have already configured MGX switches.
|
After you assemble a new switch, as described in either the Cisco MGX 8850 Hardware Installation Guide (PXM45/B and PXM1E), you must initialize the switch before you can configure it. Although PXM1E cards ship with the latest version of boot firmware on the front card, the runtime firmware cannot be loaded until both front and back cards have been installed. When you initialize the switch, you are configuring the switch to load a specific runtime firmware version from the PXM1E hard disk.
A new switch must be initialized using a console port management session. As shown in Figure 2-1, a console port management session requires a terminal or workstation with a serial connection to the Console Port (CP) port on the PXMUI-S3 back card.
 |
Note Note that some or all of the commands discussed in this section require service-level or above user privileges. To access these commands, you must have debug (Service or Cisco level) privileges and passwords. Check with TAC for assistance. |
To initialize the switch, use the following procedure.
You can use any personal computer or UNIX workstation with VT-100 emulation software.
|
Note You can connect the terminal to a PXM1E in either slot 7 or slot 8 in the Cisco MGX 8850 (PXM1E), or in slot 1 or slot 2 in the Cisco MGX 8830. |
Step 2 Start the terminal or, if you are using a workstation, start a terminal emulation program and configure it to connect to the switch through the serial port on the workstation. For instructions on configuring the terminal emulation program, refer to the documentation for the emulation program.
The default switch configuration supports the following settings: 9600 bps, 8 data bits, no parity, 1 stop bit, no hardware flow control.
Step 3 At the workstation, enter the command that connects the terminal emulation program to another computer.
Step 4 If the switch power is not on, turn on the switch power as described in the Cisco MGX 8850 Hardware Installation Guide (PXM45/B and PXM1E).
|
Note You can connect the workstation to the switch before or after power is applied. If you start the terminal emulation program before turning on the switch, the terminal emulation program displays the switch startup messages. |
Step 5 If the switch does not display any messages or a prompt, press Return.
When startup is complete for an uninitialized switch, it will display the PXM1E backup boot prompt:
Step 6 Locate and write down the version number for the runtime firmware provided with your switch. You need this version number to complete the next step.
The version number is listed in the Release Notes for Cisco MGX 8850 and MGX 8830 Software Version 3 (PXM45/B and PXM1E). You must use the same format listed in the firmware file name when you enter the number. For example, if the firmware filename is pxm1e_003.000.000.040_mgx.fw, the firmware version number you will enter is 003.000.000.040.
Step 7 When the PXM backup boot prompt appears, define the PXM runtime firmware version by entering the sysVersionSet command as follows:
Replace version with the version number for the runtime firmware. For example:
Step 8 Reboot the switch by entering the reboot command as follows:
During initialization, the switch will appear to boot twice. When the reboot is complete, the switch displays the Login prompt, which indicates that the firmware is loaded and the switch is ready for configuration.
 |
Tip The sysVersionSet command has failed if the switch reboot process stops and displays the message "Can not open file C:/version" or the message "Unable to determine size of C:/FW/filename." If this happens, press Return to display the backup boot prompt, then refer to the "Troubleshooting Upgrade Problems" section in "Downloading and Installing Software Upgrades." |
Step 9 To log in to the switch, enter the login name supplied with your switch, then enter the password for that login name. For example:
|
Note The default configured username and password sets are: user cisco, password cisco; user service, password serviceuser; and user superuser, password superuser. |
|
Note If the switch has not fully started and is operating in init state (which is also called stage 1 CLI mode), an i appears in the switch prompt: unknown.7.PXM.i>. In this mode, you can only log in as user cisco, password cisco, and a limited set of commands are available for troubleshooting. If you log in during init state and the card progresses to the active or standby state, the card will log out the init state user and prompt you to log in again. At this point, you can log in as a configured user with the corresponding password. |
|
Note In a Cisco MGX 8850 (PXM1E), the number 7 in the switch prompt indicates that you are managing the PXM in slot 7. If you are managing the PXM in slot 8, the switch prompt displays the number 8. In an MGX 8850, the number 1 in the switch prompt indicates that you are managing the PXM in slot 1. If you are managing the PXM in slot 2, the switch prompt displays the number 2. |
The switch does not display the password during login. When login is complete, the switch prompt appears.
The switch prompt for PXM1Es and narrow band service modules uses the following format:
Table 2-1 describes the components in the CLI prompt.
Table 2-1 CLI Prompt Components
|
|
Note The prompt for FRSM-2CT3 cards displays VHS2-CT3 as the cardtype, because the FRSM-2CT3 is a VHS card. For example: MGX.1.4.VHS2CT3.a >. FRSM 8T1E1 cards, however, follow the standard naming convention and display FRSM as the cardtype in the switch prompt. |
After initialization, the PXM in the initialized slot becomes active. If a second PXM resides in the other slot, the active PXM initiates a runtime firmware load on the other slot. After the runtime firmware loads on the nonactive PXM, the card enters standby mode, ready to take control if the active card fails.
After you log in, the switch maintains your session for the default period of 10 minutes (600 seconds) after the last keystroke is entered. If the session is idle longer than 600 seconds, the session is terminated.
|
Tip To restart an automatically terminated session, press Return. The switch will prompt you for a login name and password. |
Step 10 To change the session time-out period, enter the timeout command as follows:
Replace seconds with the number of seconds you want the session to remain active before it times out. The maximum value is 600. To disable time-out, enter 0 seconds. The switch uses the new timeout value until you terminate the session. Each time a new session is started, the timout value returns to the default value, 600 seconds.
Once you have completed the procedure above, you have established a command line interface (CLI) management session. You can use a CLI management session to configure or monitor the switch.
After initialization, you can terminate and start sessions at any time using the terminal or workstation connection to the CP port, which was described in the previous section.
|
Tip The switch also supports several other types of management connections, including remote connections. For instructions on supporting and starting other types of CLI management sessions, see "Supporting and Using Additional CLI Access Options." |
|
Note Some or all of the commands discussed in this section require service-level or above user privileges. To access these commands, you must have debug (Service or Cisco level) privileges and passwords. Check with TAC for assistance. |
To start a CLI management session at the CP port for switch configuration and monitoring, use the following procedure.
For instructions on preparing the terminal and the connection, refer to the previous section, "Initializing the Switch."
Step 2 If the Login prompt does not appear, press Return. The Login prompt comes from the switch and indicates that the terminal has successfully connected to the switch.
Step 3 When the Login prompt appears, enter the login name supplied with your switch, then enter the password for that login name. For example:
|
Note If the switch has not fully started and is operating in init state (which is also called stage 1 CLI mode), an i appears in the switch prompt: unknown.7.PXM.i>. In this mode, you can only log in as user cisco, password cisco, and a limited set of commands are available for troubleshooting. If you log in during init state and the card progresses to the active or standby state, the card will log out the init state user and prompt you to log in again. At this point, you can log in as a configured user with the corresponding password. |
The switch does not display the password during login. When login is complete, the switch prompt appears.
The switch prompt for PXM1E and service modules uses the following format:
Table 2-1 describes the components in the switch prompt.
|
Note The switch prompt for FRSM-2CT3 cards uses a different card name in the prompt. This is to distinguish FRSM -2CT2 cards from FRSM-8T1 cards. The FRSM-2CT3 cards use the name VHS2CT3 in the place for cardtype. FRSM-8T1 card use the standard naming convention and display FRSM in the place for cardtype. |
After you log in, the switch maintains your session for 10 minutes (600 seconds) after the last keystroke is entered. If the session is idle longer than 600 seconds, the session is terminated.
|
Tip To restart an automatically terminated session, press Return. Depending on the application you use to login to the switch, you may be prompted for a login name and password. |
Step 4 To change the session time-out period, enter the timeout command as follows:
Replace seconds with the number of seconds you want the session to remain active before it times out. The maximum value is 600. To disable timeout, enter 0 seconds. The switch uses the new timeout value until you terminate the session. Each time a new session is started, the timeout value returns to the default value, 600 seconds.
Once you have completed the procedure above, you have established a CLI management session. You can use a CLI management session to configure or monitor the switch.
CLI management sessions automatically terminates after the configured idle time. The default idle time is 600 seconds (10 minutes) and can be changed with the timeout command. To manually end a CLI management session, enter the bye or exit command.
|
Note The bye and exit commands end the CLI session. They do not terminate the terminal session. For instructions on terminating the terminal session, refer to the manuals for your terminal or terminal emulation program. |
To restart the session after entering the bye or exit command, press Return, and the switch will prompt you for a username and password.
The commands in the switch operating system are associated with the cards that are installed in the switch. Before you execute a command, you must select a card that supports the command. The switch displays the currently selected card in the switch prompt. For example, the following switch prompt shows that the PXM card in slot 7 is selected:
To select another card in the switch, enter the cc command:
Replace slotnumber with the slot number of the card you want to manage. Table 2-2 lists the valid slot numbers for each card type.
Table 2-2 Valid Slot Numbers for Each Card Type
|
After you execute the cc command to change cards, verify that you are managing the correct card by viewing the slot number that is shown in the switch prompt. The following example shows the prompt for a CESM card in slot 6 of a Cisco MGX 8850 switch:
If you have trouble entering a command, look at the switch prompt to see if you have selected the correct card and type for the command. The following example shows the response to an unrecognized command:
The dspdate command runs on a PXM1E card. It is not recognized by a CESM card.
|
Tip The command examples in this book include the switch prompt so that you can verify which card types support specific commands. |
The default switch configuration allows you to enter command abbreviations. Because the help command is the only command that begins with he, you can use the abbreviated he command to display help. The following example demonstrates that the switch recognizes your partial entry of the help command because it proceeds to list commands.
|
Tip To disable the command abbreviation feature, enter the cnfcmdabbr command. To display the current setting for this option, enter the dspcmdabbr command. |
Notice the last line of the help command display. Because the help display is too long to appear on one screen, it is displayed in pages. Press Return to display the next page, or type q and press Return to cancel the help display.
The following example demonstrates what can appear when a command is entered at the wrong card prompt.
In the example above, the dspcds command is entered at the FRSM prompt, but this command is not supported on the FRSM card. The dspcds command is only supported on the PXM1E.
The following example demonstrates what can appear when only part of a command is entered at a service module prompt:
In the example above, dspc is entered at the FRSM-2CT3 card prompt. Because there are several possible commands that start with dspc, the switch lists all supported commands that start with those letters.
The following example demonstrates what can happen if you enter only part of a command at the PXM card prompt, but only one command that starts with the letters you entered: the switch defaults to that command.
In the example above, clidb is entered at the PXM card prompt. Because the only command that begins with the letters clidb is clidbxlevel, the switch defaults to that command.
|
Note The example illustrated above is only available on PXM1E. If you enter only part of a command on a
service module, the switch will display the following error message: Unknown Command : |
Whenever the switch displays an error message, be sure to check the spelling of the command, the parameters entered with the command, and the prompt at which the command was entered.
The following sections describe how to display the following types of command help:
The commands you can use to manage the switch are determined by your user name, which is configured for a particular access level. User names and access levels are described in more detail in the "Configuring User Access" section later in this chapter. To display a list of all the commands available to the username you used at log in, enter the help command as follows:
To display a list of commands that include a common set of characters, enter a question mark and the common set of characters, as shown in the following example:
Detailed command lists display the following additional information for each command:
|
Note To display detailed command lists, you must establish a session using a username with SERVICE_GP privileges or higher (access privileges are described later in this chapter in the "Configuring User Access" section). You can also find this information in the Cisco MGX 8850, MGX 8950, and MGX 8830 Command Reference. |
To enable detailed command lists, enter the clidbxlevel command as shown in the following example:
After you enter this command, you can display detailed command lists by entering the help command as shown in the following example:
|
Note After you enter the clidbxlevel command, the help command displays detailed reports for that session only. You can disable detailed reports by entering the clidbxlevel 0 command. Every time you start a new session, detailed command lists are disabled. |
The Access column shows the access level required to enter the command. Access levels are described in the "Configuring User Access" section later in this chapter.
The Card State column identifies the card states during which the command can be executed. Valid card states are active, standby, and init. Cards are labeled as init during switch startup. The options that appear in the Card State column are described in Table 2-3.
If a plus symbol appears in the Log column, each successful execution of the command is logged. If a minus symbol appears in the column, the command is not logged.
To display the syntax of a command, enter the command without any parameters. The following example shows the syntax report provided by the switch entering the addport command.
When a parameter is shown between less-than (<) and greater-than (>) symbols, the parameter represents a variable that must be replaced by a value. The values are described below the command syntax.
When the parameter is shown between brackets ([]), it is an optional parameter. If you omit an optional parameter, most commands will use the last value defined for the option. If no value has been assigned to an option, the default value is used.
|
Note Some commands, such as dspcd and saveallcnf, do not require parameters, so entering the command without parameters executes the command.When you enter the saveallcnf command, which saves the current switch configuration to a file, the switch prompts you to confirm the save before execution begins. Whenever the switch prompts you to confirm a command, the command you are confirming is likely to change the switch configuration, reduce switch performance, or take a long time to execute. |
|
Tip To see the syntax of a command that does not require parameters, enter the command with a parameter you know is incorrect. For example:ERR: Invalid Slot number specified ERR: Syntax: dspcd ["slot_number"] slot number -- optional; |
The usernames and passwords supplied with your switch provide access to all switch features, and they allow you to add and delete users and change user passwords.
When configuring user access for the switch, consider the following recommendations:
The following sections describe how to add users, change passwords for existing users, delete users, and recover the user cisco password.
The Cisco MGX 8850 and the Cisco MGX 8830 switches support up to 50 users. To create a user account, specify the following information:
The user name and password identify the user and determine the user access level for switch management.
An access level must be assigned to a user when the user is added to the switch. The access levels listed in Table 2-4 are used throughout this guide to indicate the level of access required to execute a command or complete a procedure. These access levels are also called access privileges. If a user has access privileges at a lower level than a command requires, the user cannot execute the command. If the user has access privileges at the level required or at a higher level, the user can execute the command.
1. SPVC = soft permanent virtual path
|
Note Earlier releases of the Cisco MGX 8850 software support users at levels Group 2 through Group 5. These user levels have been removed from the software. If you upgrade a switch that has users configured at these levels, the user level for the affected users will change to Group 1 level access during the upgrade. |
To add a user to the switch, use the following procedure.
Step 2 Enter the following command after the switch prompt:
Enter the username using 1 to 12 alphanumeric characters. Specify the access level by entering one of the levels defined in Table 2-4.
|
Note The access levels are case-sensitive and must be entered as shown in Table 2-4. Also, you cannot add users at access levels that are equal to or above your own access level. |
If you enter the command correctly, the switch prompts you for a password.
Step 3 Enter a password, using 5 to 15 characters.
Step 4 When prompted, enter the password a second time to validate the previous entry.
This completes the addition of the new user.
Step 5 To display the new user in a list of all users, enter the dspusers command.
|
Tip To determine which commands are available at a particular access level, log in to the switch as a user at that access level, then enter the help or ? command. |
Step 6 To test the username, enter the bye command, then log in as the new user.
|
Tip If you forget which username you used to log in, enter the whoami command. This command displays the username, access level, and access method (for example, Telnet) for the current session. |
Enter the cnfpasswd command to change your own password.
|
Note The cnfuser command allows you to change another user password if you have the correct access privileges. The next section describes how to use the cnfuser command. |
To change your own password with the cnfpasswd command, use the following procedure.
Step 2 Enter the following command after the switch prompt:
Step 3 When prompted, enter your current password.
Step 4 When prompted, enter a new password, using 5 to 15 characters.
Step 5 When prompted, enter the new password a second time to validate the correct entry.
This completes the change of password.
Step 6 To test the new password, enter the bye command, then log in using the new password.
After you create a user, you can change that user's access level or password using the cnfuser command.
|
Note To change your own user password, enter the cnfpasswd command as described in the preceding section. |
To change the user level or password of a switch user, use the following procedure.
Step 2 Enter the following command after the switch prompt:
Replace username with the name of the user for whom you are making the change.
If you are changing the password, specify the -p option and enter a password containing from 5 to 15 characters. If you are changing the user access level, specify the -l (lowercase L) option and enter the appropriate access level as shown in Table 2-4.
|
Note You can change passwords and access levels only for users who have privileges lower than the username you used to log in. |
Step 3 To test a new password, enter the bye command, then log in using the new password.
Step 4 To verify a user access level change, enter the dspusers command.
The dspusers command displays all the usernames and the access level for each user as shown in the following example:
To delete a user, use the following procedure.
Step 2 Enter the following command after the switch prompt:
Enter the username using from 1 to 12 alphanumeric characters.
This completes the deletion of a user.
Step 3 To verify the user has been deleted, enter the dspusers command.
If you lose or forget your password for switch access, you should ask a user with a higher access level to reset your password using the cnfuser command. If you do not have any passwords for any access levels, you can use the following password recovery procedure to reset the password for user cisco. This procedure resets the user cisco password to cisco and leaves all other passwords unchanged. (You can change the other passwords with the cnfuser command after logging in as user cisco.)
|
Note This feature can be disabled using the cnfpswdreset command as described in the next section. You can determine if this feature is enabled or disabled by logging in as a user at any level and entering the dsppswdreset command. |
 |
Caution Anyone with physical access to the switch CP can reset the password, deny access to other users, and reconfigure the switch. To prevent unauthorized switch access and configuration, the switch should be installed in a secure area. |
Step 2 When the login prompt appears, press ESC, CTRL-Y to reset the password.
Step 3 Log in using username cisco and password cisco.
Step 4 To maintain switch security after resetting the cisco user password, change the password using the cnfpasswd command.
If the switch you are managing is in an insecure area, you might want to disable the user cisco password reset feature. Otherwise, anyone with physical access to the switch CP can reset the password, deny access to other users, and reconfigure the switch. This feature can be enabled again at a later date if you know the user name and password for a user at the SERVICE_GP privilege level or higher.
To enable or disable the password reset feature, use the following procedure.
Step 2 To disable password reset, enter the cnfpswdreset off command.
Step 3 To enable password reset, enter the cnfpswdreset on command.
Step 4 To view the status of this feature, enter the dsppswdreset command.
The switch name identifies the switch you are working on, which is important when you are managing multiple switches. The current switch name appears in the CLI prompt when you are managing a PXM cards and service modules. To change the switch name, use the following procedure.
Step 2 Enter the following command after the switch prompt:
Enter up to 32 characters for the new node name, and since the node name is case-sensitive, be sure to use the correct case. For example:
|
Note The node name cannot contain any spaces or special characters. |
The new name appears immediately in the next CLI prompt.
The switch date and time is appended to event messages and logs. To assure that events are properly time stamped, use the following procedure to view and change the date and time.
Step 2 To view the current switch date and time, enter the following command after the switch prompt:
Step 3 To change the switch date, enter the following command:
Step 4 To change the time zone, enter the following command:
Replace timezone with one of the parameter values listed in Table 2-5. If your switch is located outside the Western Hemisphere, select GMT and use the next step to specify an offset from GMT. If your switch is located in the Western Hemisphere choose the appropriate option from Table 2-5. Daylight times are adjusted by one hour in the Fall and Spring for daylight savings. Standard times are not adjusted.
Step 5 To configure an offset from GMT, enter the following command:
Replace <timeoffsetGMT> with the offset in hours from GMT. Enter a number from -12 to +12.
Step 6 To change the switch time, enter the following command:
Replace <hh> with the hour of the day (0 to 23), mm with the minute of the hour (0 to 59), and ss with the number of seconds in the minute (0 to 59).
Step 7 To verify the new date and time settings, enter the dspdate command.
The MGX switches support many PNNI configuration commands. This section describes how to configure the basic PNNI configuration parameters for the switch. "Managing PNNI Nodes and PNNI Routing," describes how to manage PNNI after you have brought up the PNNI node.
|
Caution It is important to configure the PNNI node parameters before you start creating SPVCs as described in "Provisioning PXM1E Communication Links." If you create SPVCs using the default PNNI node parameters and later change those parameters, the node will advertise the old ATM address information for the older SPVCs as well as the new ATM address information. To keep PNNI running at maximum efficiency, set the PNNI node parameters to the proper values before creating SPVCs, or delete and recreate old SPVCs after making PNNI node parameter updates. |
The PNNI controller simplifies switch configuration by using PNNI protocol to discover call routes in an ATM network. Without the PNNI controller, each route through the network would have to be defined manually. "Managing PNNI Nodes and PNNI Routing," provides more information on PNNI. This section describes how to enable and configure the PNNI controller for the switch.
|
Note Before entering the following command, you must log in as a user with SUPER_GP privileges or higher. |
To enable and configure the PNNI controller, enter the following command:
Table 2-6 describes the parameters for the addcontroller command.
|
Tip Remember to include the i option, which identifies the controller as an internal controller. |
Table 2-6 Parameter Descriptions for the addcontroller Command
|
To display the PNNI controller configuration, enter the dspcontrollers command:
The Cisco MGX and SES PNNI Network Planning Guide provides guidelines for selecting a PNNI level and peer group ID. To set these parameters in the switch, use the following procedure.
Step 2 Disable PNNI node operation by entering the following command:
The node-index uniquely defines a logical PNNI node within the switch. Initially, there is just one logical PNNI node at the lowest PNNI level, and its index number is 1. If you add a higher level logical node to the physical node, the first higher level will be numbered two, and the next higher level will be number three. Additional levels receive sequentially higher node index numbers.
During this general node configuration, you are setting the PNNI level and peer group ID for the lowest PNNI level, so replace node-index with 1.
|
Note For instructions on creating logical nodes above the lowest PNNI level, see "Managing PNNI Nodes and PNNI Routing." |
Step 3 Change the PNNI address with the cnfpnni-node command as follows:
To configure the lowest PNNI level, replace <node-index> with 1. Replace level with the PNNI level you want to use, and replace peerGroupID with the 13-byte peer group ID you want to use. For example:
Step 4 Enable PNNI node operation by entering the following command:
Replace node-index with the value you used when disabling and reconfiguring the PNNI node.
Step 5 To display the PNNI node configuration, enter the following command:
The switch displays a report similar to the following example:
The Cisco MGX and SES PNNI Network Planning Guide provides guidelines for setting the PNNI node address, which is identical to the switch ATM address. To set the PNNI node address, use the following procedure.
|
Caution When installing new switches, you can assume that each default node address will be unique. When PXM cards are repaired or moved between switches, however, it is possible that two switches will start using the same node address. To prevent duplicate node addresses, use your own address plan, and check the node address whenever a PXM card is replaced or moved from one switch to another. |
Step 2 Disable PNNI node operation by entering the following command:
The node-index uniquely defines a logical PNNI node within the switch. Initially, there is just one logical PNNI node at the lowest PNNI level, and its index number is 1. If you add a higher level logical node to the physical node, the first higher level will be numbered two, and the next higher level will be number three. The node index is a reference to particular logical PNNI process in the node.
The PNNI address is configured at the lowest PNNI level, so replace <node-index> with 1.
|
Note The PNNI address you enter at the lowest level is used for all levels. PNNI increments the selector byte (which is the last byte) of the ATM address to represent logical nodes at higher PNNI levels. |
Step 3 Change the PNNI address with the cnfpnni-node command as follows:
To modify the PNNI address at the lowest level, replace <node-index> with 1, and replace atm-address with the 20-byte ATM address you want to use. For example:
|
Note The ATM address in the example above shares the same seven most-significant bytes (level 56 peer groups use the first 7 bytes) as the peer group ID example in the previous section, so PNNI can advertise only the peer group ID outside of the peer group. If the ATM address and peer group ID used different prefixes, PNNI would have to advertise the node ATM address and the peer group ID. The ATM address should conform to your ATM address plan. For more information, refer to the Cisco MGX and SES PNNI Network Planning Guide. |
|
Tip Use the Copy and Paste functions of terminal session software to copy an existing ATM address into the command line. Then you can use your editing keys to make changes to the address before pressing Enter to execute the command. |
Step 4 Enable PNNI node operation by entering the following command:
Replace <node-index> with the value you used when disabling and reconfiguring the PNNI node.
Step 5 To display the PNNI node configuration, enter the command:
The switch displays a report similar to the following example:
The PNNI node ID appears in many CLI displays, including the dsppnni-node command display. The default node ID is PNNIlevel:160:defaultATMaddress. If you change the PNNI level or the node ATM address, you should also change the node ID so that the node ID represents the correct PNNI level and ATM address. This will make it easier to identify the node when using CLI commands because most CLI commands reference the node ID, not the node ATM address. For example:
In the example above, there is no reference to the ATM address for the remote switch named 8850_SF. However, if the node ID is set to match the ATM address, it will be easy to determine the ATM address of a remote switch.
To set the PNNI node ID, use the following procedure.
Step 2 Disable PNNI node operation by entering the following command:
The node-index uniquely defines a logical PNNI node within the switch. Initially, there is just one logical PNNI node at the lowest PNNI level, and its index number is 1. If you add a higher level logical node to the physical node, the first higher level will be numbered two, and the next higher level will be number three. The node index is a reference to particular logical PNNI process in the node.
The PNNI node ID is configured at the lowest PNNI level, so replace <node-index> with 1.
|
Note The node ID you enter at the lowest level is used for all levels. PNNI uses a modified version of the lowest level node ID for upper level nodes. |
Step 3 Change the PNNI node ID with the cnfpnni-node command as follows:
To configure the lowest PNNI level, replace <node-index> with 1. Replace PNNIlevel with the lowest PNNI level, and replace atm-address with the 20-byte ATM address you want to use. For example:
Step 4 Enable PNNI node operation by entering the following command:
Replace <node-index> with the value you used when disabling and reconfiguring the PNNI node.
Step 5 To display the PNNI node configuration, enter the command:
The switch displays a report similar to the following example:
The Cisco MGX and SES PNNI Network Planning Guide provides guidelines for selecting the SPVC prefix. The SPVC prefix is the ATM prefix that PNNI advertises for all SPVCs and Soft Permanent Virtual Paths (SPVP) on this node. The ATM address for each SPVC and SPVP is the combination of the SPVC prefix and a port identification number.
You can configure one SPVC node prefix per node. To set the SPVC prefix, use the following procedure.
|
Note Although the SPVC prefix is set to match the first 13 bytes of the PNNI node address by default, changing either the PNNI node address or the SPVC prefix has no effect on the other setting. If the PNNI node ATM address and the SPVC prefix do not match, the switch advertises both prefixes instead of just one, and this advertising takes additional bandwidth. |
|
Note You can change the SPVC prefix only when no SPVCs or SPVPs have been defined. Once an SPVC has been defined, you must delete all SPVCs before you can change the SPVC prefix. For information on deleting SPVCs, see the "Deleting SPVCs and SPVPs" section in "Provisioning PXM1E Communication Links." |
Step 2 Use the following command to display the current SPVC prefix:
The switch response is similar to the following example:
|
Tip If the SPVC prefix begins with 47.009181000000, the SPVC prefix is probably set to the default value. To display the current PNNI node address, enter the dsppnni-node command. |
Step 3 To change the SPVC prefix, enter the following command:
Replace prefix with the 13-byte prefix you want to use. For example:
|
Note The SPVC prefix in the example above matches the first 13 bytes of the node PNNI address example presented in the previous section, so PNNI can advertise one prefix to support both SVC connections through the node and SPVCs. If the SPVC prefix does not match the corresponding bytes in the ATM address, PNNI advertises two prefixes instead of one. The SPVC prefix should conform to your ATM address plan. For more information, refer to the Cisco MGX and SES PNNI Network Planning Guide. |
Step 4 Verify the correct entry of the prefix by entering the dspspvcprfx command.
After you configure the PNNI level, peer group ID, ATM address, and SPVC prefix, review the summary addresses the node will advertise. If all PNNI parameters are properly coordinated, the node should display a single summary address that represents all PNNI destinations in that node. To display the summary addresses, enter the dsppnni-summary-addr command as shown in the following example:
The example above is coordinated with the examples in the previous sections, so just one PNNI summary address is broadcast to the peer group. The following example demonstrates what happens when the node ATM address and the SPVC prefix are not coordinated:
In the example above, the node ATM address does not conform to the peer group ID or the SPVC prefix, so it must be advertised in addition to the SPVC prefix.
The "Network Clock Source Plan" section in "Preparing for Configuration," introduces Building Integrated Timing System (BITS) clock sources and provides guidelines for developing a network clock source plan. When the network clock source plan requires BITS clock sources on the switch, you can use the procedure in this section to configure the BITS clock connections.
Figure 2-2 shows how BITS clock sources connect to the PXM1E-UI-S3 back card.
The PXM1E-UI-S3 clock source ports can be used to receive clock signals from either T1 or E1 lines; the card does not support both line types simultaneously. These clock ports support stratum levels 1 to 3.
|
Note When using an external clock source and redundant PXM1E cards, use a Y-cable to connect that clock source to the same clock port on both PXM1E cards. Otherwise, the clock source is available to only one of the PXM1E cards. |
Release 3 of the MGX switches supports two forms of network clock source configuration:
Both types of network clock configuration are described in the sections that follow.
|
Note When NCDP is enabled, your manual configuration is disabled, and vice versa. When you disable NCDP, your node reverts back to any manual clock configuration that was previously done on the node. If you re-enable NCDP after disabling it, your switch will remember your last NCDP configuration and use that unless you change it. |
The following procedure describes how to configure the switch to use clock sources on the BITS ports.
|
Note For instructions on configuring the switch to use a clock source on a PXM1E line, refer to the "Configuring PXM1E Line Clock Sources" section in "Provisioning PXM1E Communication Links." |
Step 2 To configure a primary or secondary BITS clock source, enter the cnfclksrc command:
Table 2-7 describes the parameters for this command.
Table 2-7 Parameter Descriptions for cnfclksrc Command on the PXM1E
|
Step 3 To configure an additional BITS clock source, repeat Step 2 using the correct parameters for the additional source.
Step 4 To display the clock source configuration, enter the dspclksrcs command.
The dspclksrcs command is described in the "Managing Manually Configured Clocks Sources," in "Switch Operating Procedures.".
|
Note Manual clock distribution provides a revertive function that can apply when the primary clock source fails and is subsequently restored. A failure is a loss of the primary clock source after the switch has locked on to that clock source. If the primary clock source recovers and revertive mode is enabled, the switch automatically reverts to the primary source |
The following command example shows how to configure a primary E1 external clock source at the upper connector of the PXM1E-UI-S3. Note the command punctuation.
The next example configures a primary network clock source and enables the revertive option.
The last example disables the revertive function for an E1 BITS clock.
Use the following procedure to enable NCDP on each node in your network.
Table 2-8 describes the options available for the cnfncdp command.
Table 2-8 cnfncdp Command Parameters
|
Step 2 Enter the dspncdp command to verify that the NCDP parameters were set properly.
Once NCDP is enabled on your node, the blest clock source and second best clock source are automatically selected and distributed to all nodes in the network with NCDP enabled. If no previous NCDP clock configuration has been done, NCDP selects a root clock source will come from an internal oscillator. If you want to specify the root clock source to come from an external source, use the cnfncdpclksrc command as described in the "Configuring an NCDP Clock Source" section in "Switch Operating Procedures."
|
Note Cisco recommends using an external clock source instead of the internal oscillator. |
|
Caution If you want to specify the root clock source to come from an external source before you enable NCDP, use the cnfncdpclksrc <portid> 0 command as described in the "Configuring an NCDP Clock Source" section in "Switch Operating Procedures." If you run cnfncdpclksrc <portid> 0 before you enable NCDP with the cnfncdp command, the root clock source will be the external clock you configured, instead of the internal oscillator. |
If you wish to change the BITS clock selected by NCDP, enter the cnfncdpclksrc command, as described in the "Configuring an NCDP Clock Source" section in "Switch Operating Procedures."
The switch uses two types of IP addresses for Ethernet LAN access:
The following sections describe how to set these addresses. For information on how the switch uses these addresses and how to choose the addresses, see the "Guidelines for Creating an IP Address Plan" section in "Preparing for Configuration."
|
Note The switch also supports IP addresses for dial-in and ATM inband access. For more information on these access options, see "Supporting and Using Additional CLI Access Options." |
The boot IP address is the LAN port IP address that a PXM card uses when it first starts up. If the switch cannot fully start, this IP address can be used to access the switch in boot mode. When the switch is properly configured (with different addresses set for the boot IP and LAN IP addresses), the boot IP address can also be used to access the standby PXM card directly, while the disk IP address can be used to access the active PXM.
|
Note Because the LAN IP address is stored on the PXM hard disk and is not used until after the runtime software loads, Cisco recommends that the boot IP address be set in every switch. This enables switch management over Ethernet when the boot software has loaded. |
To set the boot IP address, use the bootChange command, which also allows you to define a remote boot location, a default gateway IP address, and a username and password for the remote boot location.
Step 2 Enter the bootChange command as shown in the following example.
In this example, the switch is waiting for you to take action on the boot device option. Enter a period <.> to clear the current value (lnPci), enter minus <-> to go back to the previous field (although this is the first of 14 fields), or press Return to accept the current value and display the next option. The following example shows all options.
|
Note The only two options that must be set to support the boot IP address are inet on ethernet (e) and gateway inet. The bootchange command operates only on the active card. If you are having trouble bringing up a standby card, you can set the boot IP address with the sysChangeEnet command as described in the "Troubleshooting Upgrade Problems" section in "Downloading and Installing Software Upgrades," If you set the boot IP address on the standby card with the sysChangeEnet command and it is different from the IP address set with the bootchange command on the active card, the standby card will start using the boot IP address set with the bootchange command when it reaches standby mode. |
Step 3 Accept, clear, or change option values as necessary until the inet on ethernet option appears. Table 2-9 defines the options that you can change.
Table 2-9 bootChange Command Option Descriptions
|
Step 4 Set the inet on ethernet (e) option to the boot IP address value you want to use. The following example shows how the command appears when a new value has been entered:
The 172.29.52.88 address appeared as part of the prompt. If no address had been previously defined, no text would appear after the colon. In this example, 172.29.52.108 is the new boot IP address, and 255.255.255.0 is the new network mask.
Step 5 Set the gateway inet option to the IP address for the default gateway on the subnet that hosts the switch.
Step 6 Accept, clear, or change values as necessary until the switch prompt reappears.
Step 7 To verify the new values you have set, enter the bootChange command and press Return for each of the 14 values.
A local LAN connection extends switch management to all workstations that have connectivity to the LAN to which the switch is connected. Figure 2-3 shows the hardware required for a local LAN connection.
|
Note The PXM-UI-S3 card shown in Figure 2-3 has two LAN ports. In the current release, only the LAN 1 connector is enabled for communications. Communication through the LAN 2 connector is disabled. |
Before you can manage the switch through the PXM LAN port, you must first assign an IP address to the LAN port. The LAN or disk IP address is the IP address for the Ethernet LAN port on the active PXM. The LAN IP address is also called the disk IP address because it is stored on the PXM hard disk. However, the IP address for the Maintenance Port is also stored on the hard disk and must be different, so this section refers to this address as the LAN IP address.
|
Tip The significance of the disk IP address for the LAN Port is that it is stored on the hard disk and is not available until the runtime software is loaded on the PXM card and the card is active. To access the LAN port over Ethernet when a PXM is operating in boot or standby mode, you must use the Boot IP address. |
The LAN IP address can be set to match the boot IP address when only one IP address is available, or it can be set to a unique address to support access to the standby PXM during regular operation. For more information on how the boot and LAN IP addresses are used, see "Guidelines for Creating an IP Address Plan."
To set the IP address, enter the ipifconfig command as described in the following procedure.
Step 2 Verify that the IP address is not already configured by entering the dspipif command:
|
Note If you omit the lnPci0 option, the switch displays the configuration for all switch IP interfaces: the ATM interface (atm0), the PXM LAN port interface (lnPci0), and the PXM maintenance port interface (sl0). Note that the address for each interface must be unique. |
In the IP Interface Configuration Table, look for an Internet address entry under the lnPci entry. If an IP address is configured, you can use that address and skip the rest of this procedure. However, if the address has not been entered or is incompatible with your network, you must configure a valid IP address as described in the next step.
|
Note If you are using CWM to manage your network, the IP address 10.0.XX cannot be used as the LAN address for the switch. |
Step 3 To set the IP address for the LAN port, enter the ipifconfig command using the following format:
Replace <IP_Addr> with the IP address you want this port to use, and replace <Mask> with the network mask used on this network.
|
Note There are other options for the ipifconfig command, and you can set one or more options simultaneously. Any options you do not define in a command remain unchanged. For more information on this command, refer to Cisco MGX 8850, MGX 8950, and MGX 8830 Command Reference. |
Step 4 Verify that the IP address changes by entering the dspipif command. For example:
|
Tip You can view the IP routing table for the switch by entering the routeShow command. To manage routes in the routing table, you can enter the following commands: routeAdd, routeDelete, routeNetAdd, and routestatShow. |
The switch includes a Telnet server process that you can use to connect to and manage the switch. Before you can establish a CLI Telnet session, you must set up the hardware for your access method and assign the appropriate boot and LAN IP addresses.
After the LAN IP interface has been configured and a physical path established to the Cisco MGX 8850, you can start a CLI session using a workstation with a Telnet client program. To establish a CLI management session, use the following procedure.
Replace ipaddress with the appropriate LAN IP address as follows:
|
Note The Telnet program on your workstation may require a different start up and connection procedure. For instructions on operating your Telnet program, refer to the documentation for that product. |
|
Tip If you have trouble accessing the switch from a workstation, use the PING program at the workstation to test communications. For example: ping 10.10.10.1. You can also view the IP routing table for the switch by entering the routeShow command. To manage routes in the routing table, you can enter the following commands: routeAdd, routeDelete, routeNetAdd, and routestatShow. |
Step 2 If the Login prompt does not appear, press Enter.
The Login prompt comes from the switch and indicates that the workstation has connected successfully to the switch.
Step 3 When the Login prompt appears, enter the user name provided with your switch and press Enter.
Step 4 When the password prompt appears, enter the password provided with your switch and press Enter.
After you successfully log in, a prompt appears that is similar to the following:
The Cisco MGX 8850 and the Cisco MGX 8830 switches include a Simple Network Management Protocol (SNMP) agent that you can configure for communications with a network management station such as Cisco WAN Manager (CWM) or a third-party SNMP manager. When configured for SNMP management, the switch accepts configuration commands from management stations and sends status and error messages to the management station.
Typically, CWM operates on a workstation that is connected to an IP network, and CWM uses IP over ATM connections to connect to Cisco MGX 8850 and Cisco MGX 8830 switches. For information on establishing this type of access, see the "Setting Up ATM WAN Connections" section in "Supporting and Using Additional CLI Access Options."
To support the auto-discovery feature of CWM, ILMI should be brought up on all links between the CWM workstation and the switches it will manage. For information on bringing up ILMI, see the "Configuring ILMI on a Port" section in "Provisioning PXM1E Communication Links."
The following tasks are described in this section:
The SNMP trap source IP address is sent to SNMP managers, such as CWM, in the SNMP trap Packet Data Unit (PDU). This IP address identifies the source of the trap and can be used by the SNMP manager to access the remote SNMP agent. This address must be configured to enable communications with an SNMP manager.
|
Note If the trap manager IP address is not set, CWM will reject traps from the switch. |
The switch can communicate with an SNMP manager over the LAN or ATM IP interfaces. In some installations, the LAN IP interface will be used for CLI management and the ATM IP interface will be used for SNMP management. When you select the SNMP trap manager IP address, you must select the correct interface address.
To define the SNMP trap manager IP address, enter the cnftrapip command as follows:
The IP address should match the LAN IP address or the ATM interface IP address. For information on setting and viewing the LAN IP address, see the "Setting the LAN IP Addresses" section earlier in this chapter. For information on setting and viewing the ATM interface IP address, see the "Setting Up ATM WAN Connections" section in "Supporting and Using Additional CLI Access Options."
The SNMP Manager destination IP address identifies the IP address of an SNMP manager, such as CWM, to which the switch sends SNMP traps. If you are using CWM to manage the switch, CWM will automatically configure the destination IP address on the switch. If you are using another SNMP manager, you can configure the destination IP address with the addtrapmgr command as follows:
Replace ipaddress with the IP address of the SNMP manager, and replace port with the UDP port number assigned to that manager. For more information on the SNMP manager IP address, refer to the SNMP manager documentation.
To configure information about a switch in the local SNMP agent, use the following procedure.
Step 2 To define the SNMP password for network management, enter the following command:
If the password parameter is not specified, the password becomes <private>.
Step 3 To define a text string that identifies the location of the switch to the management station, enter the following command:
If the location parameter is not specified, the location is set to null (no text). The location value is sent to SNMP managers when information is requested about the sysLocation MIB object.
Step 4 To define a text string that identifies a person to contact regarding issues with this switch, enter the following command:
If the contact parameter is not specified, the location is set to null (no text). The contact value is sent to SNMP managers when information is requested about the sysContact MIB object.
Step 5 To display the SNMP agent configuration, enter the dspsnmp command. The command display appears similar to the following example:
Before you can configure your switch, you need to collect information about the cards and software installed on the switch. You need to enter this information during the various configuration tasks. Table 2-10 shows the information you need and serves as a worksheet where you can enter this information.
The following procedure describes how to display the configuration information you need to enter in this table. It also describes how to verify that the correct upper and lower back cards are installed for each front card.
Step 2 To display a list of all the cards installed in the switch, enter the dspcds command after the switch prompt:
A Cisco MGX 8830 switch displays a report similar to the following example:
A Cisco MGX 8850 switch displays a report similar to the following example:
Step 3 In the worksheet in Table 2-10, write down the following information for each card:
Table 2-10 Hardware Configuration Worksheet
|
Step 4 For each slot in which a card is installed, complete the following tasks:
The dspcd command displays information that is unique to a particular card. For PXM1E cards, the switch displays a report similar to the following example:
|
Note The dspcd and dspcds commands are very similar, but they produce different reports. The dspcd command displays information about a specific card. The dspcds command displays summary information for all cards in the switch. |
For narrowband service modules, the switch displays a report similar to the report displayed on the PXM1E card. The following example shows the dspcd report for a CESM8T1 card:
For SRM cards, the switch displays a report similar to the following example:
|
Note You can not run the dspcd command on the SRM itself, because all SRM card configuration is done from the PXM card. Enter dspcd <SRM_slot_number> at the PXM to display information about the SRM cards in your switch. |
b. In the worksheet in Table 2-10, write down the following information for each card:
|
Tip Another way to display a detailed report on a card is to enter the cc command to select the card, then use the dspcd command without a slot number. However, the preferred method is to use the dspcd command with a slot number because this method can display information on a card when card errors prevent access through the cc command. |
Step 5 After you enter the required information for all cards in Table 2-10, use Table 2-11 to verify that each card is installed in a slot that supports that card type. You also need to verify that the correct back cards are installed for the corresponding front cards.
|
Note The locations where the upper and lower back cards are installed are also called bays. Each slot has an upper and a lower bay for back cards. |
Table 2-11 Valid Card Installation Options
|
If any of the cards are installed incorrectly, refer to the Cisco MGX 8850 Hardware Installation Guide (PXM45/B and PXM1E) for instructions on installing the cards correctly.
Posted: Fri Jan 23 20:44:36 PST 2004
All contents are Copyright © 1992--2004 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.
