SES PNNI Controller Configuration Guide, Release 1.1
Network Management
Downloads: This chapterpdf (PDF - 247.0KB) The complete bookPDF (PDF - 3.12MB) | Feedback

Network Management

Table Of Contents

Network Management

Minimum System Requirements

Hardware

Software

Installing and Configuring Cisco WAN Manager

Disk Partitioning Requirements

Partitioning One 9-GB Disk

Partitioning Two 9-GB Disks

Modifying the network.conf File for PNNI Networks

Configuring PNNI Topology Discovery

Configuring the SES Controller

Cisco WAN Manager SES Controller PNNI Features

SPVC Overview

WAN CiscoView 3.2

Installing CiscoView

Accessing CiscoView

Navigating in CiscoView

Main Menu Buttons

Status Bar and Buttons

Making Selections and Displaying Menus

Popup Menu Options

Using CiscoView

Preference Setting Options

Device-Specific Buttons within Configure Menu

Integrating New Device Information

Device Support Utility Features

Using the Device Support Utility

Testing Basic Connectivity and Setup

Test the IP Connectivity

Open a Telnet Session to the Device

Verify the CiscoView Preferences

Call Tracing

Connection Trace

Connection Trace Success

Connection Trace Failure

CLI Commands Functionality

conntrace Command

Path Trace

Calling and Called Path Trace Success

Path Trace Failure

SES CLI Pathtrace Commands


Network Management


This chapter describes the following network management tools you can use with SES controllers and PNNI networks:

Cisco WAN Manager SES Controller PNNI Features

WAN CiscoView 3.2

Call Tracing

Call Tracing

Minimum System Requirements

The following sections describe the hardware and software components that make up the Cisco WAN Manager (CWM) network management workstation.

Hardware

This section lists the hardware requirements for a Cisco WAN Manager network management workstation. Table 7-1 lists the minimum workstation requirements. Using a workstation that meets these requirements ensures sufficient performance.

Table 7-1 Minimum CWM Release 10.2 Workstation Requirements

Component
Minimum Requirement

Workstation

Sun Ultra 10

Memory

512 MB

CPU Speed

440 MHz

Hard Disk Drive

9.1 GB

Graphics Card

24-bit

Monitor

19 inch


Three types of machines are supported for WAN Manager 10.2 as standard platforms. They are low-end, mid-range, and high-end platforms. Table 7-2 describes the configuration for each platform.

Table 7-2 Sun Platform Requirements

Platform Type
Machine Type
Number of CPUs
Size of RAM
Hard Disk Drive
Swap Space
Desktops Supported
Connections Supported

Low End

Sun Ultra Enterprise 1 Model 151 or Ultra 10 with SCSI controller

1

512 MB

One 9 GB

1 GB

Less than 5

Less than 5,000

Mid Range option 1

Sun Ultra Enterprise 2 Model 2300 or Ultra 60

2

1 GB

Two 9 GB drives

2 GB

5—10

5,000—50,000

Mid Range option 2

Sun Ultra Enterprise 2 Model 2300 or Ultra 60

2

2 GB

Two 9 GB drives

2 GB

10—20

50,000— 100,000

High End

Sun Enterprise 4000 or Enterprise 450

at least 4

4 GB

Two 9 GB drives or disk array

4 GB

More than 20

More than 100,000



Note The minimum CPU speed requirement for all but the low end platforms is 300 MHz. All platforms require a 24-bit graphics card.


The selection of a proper CWM platform depends on a number of factors, such as, the number of CWM desktops, the number of managed connections, and the number of statistics collected and stored. Table 7-2 lists recommended CWM platforms based on the size of the network.

The following are additional notes for CWM platform requirements:

For every additional CWM desktop application, an additional 8 MB of RAM is needed beyond the standard platform configuration.

You may upgrade the standard configuration such as CPU speed, RAM size, and disk space for future expansion.

The default disk size for the Informix raw database is 900 MB. 2 GB disk space is recommended for the statistics collection process.

If the X server crashes for any reason while CWM is running, CWM should be stopped and restarted.

While CWM is running, if the remote display is killed without properly shutting down the CWM Desktop, then reopening it remotely may not succeed.

CWM must be started from a CDE environment.

For every additional CiscoView instance for BPX/IGX/MGX 8220, you need 7 additional MB of RAM and 4 additional MB of swap space beyond the CWM standard.

Software

This section lists the required software to install on the Cisco WAN Manager Network Management workstation to manage the SES controller and PNNI networks.

Release 10.5 CWM (including CiscoView 5.1)

Solaris 2.7

Informix 9.2

Orbix 3.0

Orbix Web 3.1

WingZ 2.5.5

HPOV 6.1.0

HP OpenView 6.1 is not bundled with CWM CDs. You must order HP OpenView separately.

For HP OpenView installation requirements and procedures, refer to the HP OpenView Network Node Manager Products, Installation Guide (part number J1136-90000 from HP).

BPX 9.2.33

PNNI 1.0

Installing and Configuring Cisco WAN Manager

Refer to the appropriate chapters in the Cisco WAN Manager Installation Guide for Solaris, Release 10.5 for general workstation setup (including disk partitioning) and installation procedures.

Disk Partitioning Requirements

A change in the installation procedure requires that you use the following disk partitioning requirements instead of those found in the CWM 10.5 Installation Guide for Solaris (Doc-7810308=).

Sufficient disk space and proper disk partitioning are essential to achieving the best performance from CWM and your network management workstation.


Note The minimum disk space requirement for CWM 10.5 is one 9-GB disk drive.


Use the following commands to gather some of the required information:

dmesgProvides information about the workstation type, amount of memory and CPU speed.

formatEnables you to determine information about the disk drives on your workstation. Select a disk from the list of those available, and enter the verify command to determine the current partitioning of each disk.

Partitioning One 9-GB Disk

This section describes how to partition a CWM workstation's single 9-GB disk drive. Use the information in Table 7-3 to ensure that all but the final partition will be at a set size.


Note The actual total disk space for a 9-GB disk varies depending on the manufacturer of the disk. One 9-1 GB disk might have 9.05 GB of available space, while another might only have 8.9 GB of space available. Other disk drive flaws also limit disk capacity. Most disks will NOT have a full 9.1 GB, and the slice 2 (s2) total will vary.

Table 7-3 Partitioning a Single 9-GB Disk

Slice
Partition
Space
Comments

s0

/

2000 MB

Allocate third.

s1

swap

1030 MB

Allows for memory upgrade to 1 GB; allocate second.

s2

<overlap>

8996 MB

Total amount of space on the disk; do not attempt to modify.

s3

/opt

500 MB

Allocate fourth.

s4

/var

1000 MB

Allocate fifth.

s5

     

s6

/usr/users

2500 MB

Must be 2000 MB; allocate first.

s7

 

1966

Raw partition; might be less than this amount; allocate last.




Note To check the running total of remaining disk space, click any partition to update the total free value. When the total free value is 0 free and a rounding error of 0 or 1, click OK.


Partitioning Two 9-GB Disks

This section describes how to partition a CWM workstation that has two 9-GB disk drives. When you install Solaris, partition the first disk drive as shown in Table 7-4.


Note Do not partition the second disk. The second disk is automatically partitioned during the CWM software installation.


Table 7-4 Partitioning the First 9-GB Disk

Slice
Partition
Space
Comments

s0

/

2000 MB

Allocate third.

s1

swap

1030 MB

Allows for memory upgrade to 1024 MB; allocate second.

s2

<overlap>

8996 MB

Total amount of space on the disk; do not attempt to modify.

s3

/opt

1000 MB

Allocate fourth.

s4

/var

1000 MB

Remainder of disk; allocate last.

s5

     

s6

/usr/users

2000 MB

Must be 2000 MB; allocate first.

s7

     


Note The total disk space should equal the space shown in s2.


Modifying the network.conf File for PNNI Networks

For PNNI networks using in-band management, provide the following information for your network:

NETWORK:Network2
GATEWAYS:sj234567	
DISCOVERY PROTOCOL:PNNI


Note To save your changes while using the vi editor, remember to press Esc, colon (:), then wq!.


Configuring PNNI Topology Discovery

Use the Topology Configurator to provide information required to communicate with the nodes. Also use the configurator to specify Network IP.


Note These are nodes that have their SNMP community string for GET operations not set to public, and SNMP community string for SET operations not set to private.


Configuring the SES Controller

To configure a PXM card, telnet to the card and enter the following shellConn commands:

pxm1> shmsimulateresetReason 0
pxm1> deltree "D:/DB2"


Note The above commands are used when inconsistency exists between the database and the image.


pxm1> addpnport 9.1                       assumes a BXM in slot 9
pxm1> cnfpnportsig 9.1 -nniver pnni10

Cisco WAN Manager SES Controller PNNI Features

Cisco WAN Manager (CWM) provides the following features for the SES controller:

SPVC connection management between Release 1.0 SES controllers with BXM cards at each endpoint.

Connection trace

Connection testing

End-to-End connection alarms

End-to-End connection template

Connection database

Java-based CM GUI and service class template GUI

SPVC Overview

Table 7-5 lists the types of SPVCs supported by the SES controller using CWM Release 10.5.

Table 7-5 Supported SPVC Connections

Endpoint 1
Endpoint 2
SPVC Connection
Type
Node Type

BXM

BXM

ATM SPVC

Routing nodes on BPX with feeder node on SES


WAN CiscoView 3.2

The SES controller is managed through WAN CiscoView 3.2. WAN CiscoView 3.2 requires Release 5.1 of the CiscoView Engine, which is included on the CWM Release 10.5 CD.

CiscoView Release 5.1 has been migrated from X/MOTIF to a JAVA based application. WAN CiscoView 3.2 supports line, port and resource partition configuration and real-time counters on SES controllers.

The look and feel of CiscoView 5.1 is slightly different from CiscoView 4.2, but most dialog screens will be familiar to experienced CiscoView users. Rear view selection (normally done by selecting the outer part of the device with the right mouse button) is not available for the SES controller.

Installing CiscoView

During the CWM Release 10.5 installation process, CiscoView 5.1 and BPX-SES device packages are installed for you. This eliminates the need to incrementally select device packages to install.

Accessing CiscoView

Accessing CiscoView is a simple task. From the CWM Release 10.5 Topology Map, a device can be selected for management by CiscoView.

Navigating in CiscoView

When you start CiscoView, the CiscoView main window opens. The following components comprise the CiscoView main window:

Select Device drop-down list box

Device Commands buttons

Main Menu buttons

Graphical Device display window

Select Device Drop-Down List Box

Use the Select Device drop-down list box to select and display a device. Either enter a device name or IP address, or select from the recently displayed devices listed.

Device names and SNMP read and write community strings are preserved when you open new CiscoView sessions.

Device Commands Buttons

Use the device command buttons to activate device commands unique to the displayed device. The device command buttons are described in the online help for each device package.

Main Menu Buttons

Use the Main Menu buttons to perform various CiscoView tasks.

Graphical Device Display Window

Use the Graphical Device Display window to view a graphical display of the device's back or front panel once you select a device. The display shows all device components color-coded according to their current status and refreshed according to your polling frequency. If a hot swap is detected, the device is rediscovered and the display redrawn at the next poll.

Status Bar and Buttons

Use the Status Bar and buttons to display the result of device polling, selections, and so on.

Main Menu Buttons

Table 7-6 describes the Main Menu buttons.

Table 7-6 CiscoView Main Menu Buttons 

Main Menu Button
Description

Telnet

Launches a Telnet command-line session to the managed device.

CCO

Launches a separate browser containing the Cisco Connection Online (CCO) web page.

This feature is not supported in WAN CiscoView 3.2.

Cisco Support

Opens the TAC Mailer dialog box for sending reports to the Cisco Technical Assistance Center (TAC) group. You can describe the problem using the available options and the comment field. When you click Send, your descriptions and information about the runtime device package and operating environment are sent to the specified mail recipients.

This feature is not supported in WAN CiscoView 3.2.

Preferences

Opens the Preferences dialog box where you can specify SNMP and community string. The preferences settings are preserved for all new CiscoView 5.1 sessions.

About

Displays the following information:

CiscoView release version and copyright

Active device package, if applicable

All installed device package information

Help

Opens CiscoView 5.1 help if no device is selected.

Opens context help if a device or component is selected.

This feature is not supported in WAN CiscoView 3.2.


Status Bar and Buttons

Table 7-7 describes the options on the status bar and buttons.

Table 7-7 CiscoView Status Bar and Buttons 

Status Bar/Button
Description

Status Bar

Displays the progress and result of device polling, selections, and so on.

System Info Button

Displays system information (name, description, location, contact, and up-time) for a displayed device.

Print Button

Prints the current graphical display.

Color Legend Button

Describes the significance of the colors on the graphical display. Color schemes are listed below:

Blue or Gray—Port is dormant.

Orange—Port is down.

Red—Port failed.

Yellow—Port has a minor failure.

Purple—Port is being tested.

Green—Port is active.


Making Selections and Displaying Menus

When you select a device in CiscoView, a graphical representation of the device is displayed. You view the front device panel and select different components and menu options to configure and monitor status for these devices.

Popup Menu Options

Table 7-8 describes the options on the popup menu.

Table 7-8 Cisco View Popup Menu Options

Popup Menu Option
Description

Configure

Configures device categories, such as Node Management, NNI, and so on.

Monitor

Displays a set of dynamic charts for selected device categories.

Front and Rear

Displays either the front or back device panel. The BPX-SES has only a front panel view.

Resize

Reduces the graphical display down to 90%, 80%, 70%, 60%, or 50%.

You can resize the window back up to 100% after you have reduced it.

Refresh

Triggers component polling and display update.

System Info

Displays system MIB information (name, description, location, contact, and up-time) for a device.


Using CiscoView

Once you install CiscoView and learn to navigate within it, you can perform various tasks.

Starting CiscoView

Depending on the platform, you can start CiscoView.

Within CWM Release 10.5, select a device on the Topology map.

From the CLI, enter the ~svplus/wancv/bin/cvw command.

Selecting a Device

Select a device to view its graphical representation to configure and monitor it. The device names and SNMP read and write community strings are preserved when you open new CiscoView sessions.

Setting Preferences

Use the Set Preferences option to change certain options within CiscoView.

Selecting a Component

Select a component on the graphical device display to configure and monitor it.

Configuring Your Device

Use the Configure menu to configure multiple categories of information, for example, Interface, Management, Physical, and ARP Table, simultaneously.

Different categories of information can be displayed for each device, card, and port. To see the categories of information that can be displayed for each component type, look at the Category pop up menu from the Configuration window.

Monitoring Your Device

Use the Monitor menu to monitor multiple categories of information, for example, Ethernet collisions, Management, Physical, and ARP Table, simultaneously. The Monitoring dialog is non-modal and resizeable.

Preference Setting Options

Setting Community Strings

Use the Preferences Community tab to delete the read and write community strings for the device currently being managed. This lets you enter the read and write community strings for a device after you display the device. If you want to make changes to a device or port setting, but did not specify community string when you first opened the device display, you can enter the community string without exiting and reopening the device window.

If a host's community strings are not already defined within CiscoView, you can add them with the CiscoView Community Strings dialog. Otherwise, CiscoView allows you to enter the correct community strings when you try to access the host.

If you do not enter a host's community strings when accessing the host, CiscoView uses the default read and write community strings of public and private.

Setting SNMP Preferences

Use the Preferences SNMP tab to set polling frequency, SNMP timeout and retries, and default read and write community strings. The recommended values for preferences are as follows:

Polling Frequency (sec.): 60

SNMP Timeout (sec.): 20

SNMP Retry Count: 1

Show MIB Label as: Alias

Use the Default Read and Write Community fields to define the community strings that CiscoView automatically uses for device when you do not specify the device's current community strings.

Device-Specific Buttons within Configure Menu

Table 7-9 describes the Configure menu buttons.

Table 7-9 Configure Menu Buttons (Device Specific)

Device-Specific Buttons
Description

OK

Writes modification of all categories to managed device then closes the dialog box.

Apply

Writes modification of the current category to managed device, leaving the dialog box open.

Cancel

Aborts changes and closes the catalog list.

Print

Prints the current category.

Help

Launches device-specific help.

Create

Launches a table row creation dialog box.

Delete

Deletes a selected row from the table.


Integrating New Device Information

Use the Device Support Utility to integrate new Cisco device information asynchronously with the CiscoView engine, uninstall device packages, install new device packages, or upgrade existing installed packages.

The Device Support Utility operates in one of two modes: Interactive mode or Command Line mode. The functionality of both modes is similar; the only difference between the two is that Interactive mode provides a Graphical User Interface (GUI). Each mode allows the user to display a list of currently installed device packages and their versions, uninstall one or more packages, and automate device package installations and upgrades.

Device Support Utility Features

Use the Device Support Utility to perform the following tasks:

Install and uninstall device packages.

Upgrade device packages.

View a list of currently installed device packages and their versions.

Using the Device Support Utility

Starting the Device Support Utility

From a UNIX platform, you can start the Device Support Utility by running the script "xdsu" from the ~svplus/wancv/bin directory.

Installing Device Packages

In Interactive mode, the Install Device Packages dialog box installs new device packages or upgrades existing packages. The Device Support Utility will not allow you to select a package whose superseding version has already been installed in the package repository.

Uninstalling Device Packages

In Interactive mode, the Device Support Utility dialog box shows a list of the device packages that are already installed. It also acts as a launch point for uninstalling device packages.

Testing Basic Connectivity and Setup

The following sections describe how to test the basic connectivity and setup for CiscoView. Perform the following steps when you have a CiscoView-related problem:


Step 1 Test IP the Connectivity

Step 2 Open a Telnet Session to the Device

Step 3 Verify the CiscoView Preferences


Test the IP Connectivity

From the UNIX workstation, try to ping the router's IP address. If the ping is unsuccessful, make sure that IP routing is properly enabled and is functioning. Use "ping -s" to check for slow IP response. Ping the device by its Network IP as well as by its LAN IP address.

If you can ping the device by its LAN IP address but not its Network IP address, there is a Network IP problem. Consult your system administrator for assistance in resolving this problem.

Open a Telnet Session to the Device

Enter the dspsnmp command to view the SNMP configuration and verify the community strings. If the strings are not correct, configure the device with the cnfsnmp command.

Verify the CiscoView Preferences

Use the Preferences SNMP tab to set polling frequency, SNMP timeout and retries, and default read and write community strings. The recommended values for preferences are as follows:

Polling Frequency (sec.): 60

SNMP Timeout (sec.): 20

SNMP Retry Count: 1

Show MIB Label as: Alias

Call Tracing

The SES controller supports the Call Trace feature as two distinct facilities, as defined by ATM Forum PNNI v2.0 Living List, July `98.

Connection Trace—Allows you to trace an existing connection.

Path Trace—Allows you to trace the calls in real time.

Both these facilities can be used to trace a call in the Control Plane.

Connection Trace

The Connection Trace facility can be used to determine the path taken by an existing connection from any node in the network to the destination node of the call. You initiate the trace of a connection through the SES CLI by providing the ingress interface, the call reference for Call Ref or VPI/VCI. This generates a TRACE CONNECTION message, which includes Trace Transit List (TTL) IE. This message will travel towards the node which has the called address.

Each node fills up TTL IE with node ID and Egress logical port ID and passes the message on. The egress logical port ID is obtained from the call record of the existing connection, using CallRef and VPI/VCI. The destination node makes the portID zero.

This feature is not supported on an IISP/AINI interface, neither for the transporting of IEs nor for the processing of the IEs.

Connection Trace Success

The destination node copies the TTL IE as is into TRACE CONNECTION ACK message, and sends it back to the source node with the following status:

trace completed normally

Use the Trace Transit List to find the path taken by an existing connection.

Connection Trace Failure

A connection trace may fail at any node for the following reasons:

1. Message is dropped as it is not supported.

2. Trace can not be progressed to the destination node due to some failure (say the call is already released, or call release is in progress).

You are then informed of these failure events, with appropriate cause values along with the TTL up to the last node which sent the failure message back.

CLI Commands Functionality

Refer to the Cisco SES PNNI Controller Software Command Reference, Release 1.1 for command syntax details.

conntrace Command

Refer to the Cisco SES PNNI Controller Software Command Reference, Release 1.1, for the full syntax of the conntrace command

Figure 7-1 shows the section within which a call can be traced. For instance, you could start the trace for a call setup between SRC and DST. The Trace Connection message would terminate at the IISP interface and an ACK/NACK would be sent back. This would enable the user to complete a partial trace.

Figure 7-1 Connection Trace in PNNI and IISP Network

Path Trace

Path Trace facility allows you to trace calls in real time. You enable or disable this feature node wide, on a per UNI interface basis or based on called party/calling party number. When enabled, the source node adds a TTL IE as part of the Setup/AddParty message and subsequent nodes supporting this feature add their own TTL IE. Various flags can be turned ON/OFF as part of the TTL on the source node, enabling a user to filter details on the trace. These flags are

1. Hierarchy (H)—Information from all the DTLs in the hierarchy are added if this flag is enabled, as defined in the ATM Forum specifications.

2. Crankback (CB)—If a call fails with a Crankback, the cause value is inserted in the TTL IE.


Note This feature is not a safeguard against mis-configuration, to prevent the calls from being released inadvertently.


3. VPI/VCI values (V): If enabled, VPI/VCI values of the egress port are filled up in the TTL IE at every node.

4. Call Reference values (CR): If enabled, Call Reference values of all the egress ports are filled up in the TTL IE.

Path trace information for all the traced calls are stored in the trace log file, where they can be display information on the CLI screen.

Calling and Called Path Trace Success

Use the dsplog command to display the information from the response message. This information includes the path and result of the return code.

Path Trace Failure

Path trace fails due to similar reasons specified above for Connection Trace Fail. However, since this is a real-time trace, the node detecting the failure fills in the proper information in TTL IE, apart from filling up the Release Message.

An IISP interface is treated similarly to a UNI interface and the trace is terminated.

SES CLI Pathtrace Commands

The Pathtrace commands are as follows:

pathtracenode

pathtraceport

pathtraceie

Their syntax is described in the Cisco SES PNNI Controller Software Command Reference, Release 1.1 , Chapter 2, "SVC, SPVC, and PNNI Commands."

pathtracenode

If path trace is enabled, the source node adds the TTL IE in the Setups depending on the flags set. The via node processes this IE and adds the relevant octets into the IE.

The signaling stack checks this flag before decoding TTL IE in an incoming setup message. It then checks this flag before starting TTL IE encoding procedures for an outgoing call.

pathtraceport

The pathtraceport command is for incoming calls on a source node. Signaling checks these flags/parameters before generating/decoding the TTL IE. If the user selects disable, the rest of the parameters are ignored.

pathtraceie

The pathtraceie command is used for interoperability when the SES controller is connected to equipment that does not support path trace. It takes the following two option:.

rmv

This command option allows you to control/handle Path trace on a interface level, when the PNNI network is connected to another through an IISP link, or to another PNNI network that does not support this facility. When this command is issued, TTL IE would be removed from all subsequent Setup messages going out on that interface, and would be reinserted to Connect/Release/Rls_Comp messages coming back for the same call. This ensures interoperability with other vendors' switches. All other flags would be ignored when issued with the removeIE flag. It would also remove the TTL IE from the connect message going out on that interface.

ins

This option inserts the TTL IE to any incoming Setup/Connect message. The direction of this option is exactly opposite of removeIE option. Both these options can be applied on any interface on any via node.

Figure 7-2 illustrates a sample network with pathtrace insert and remove IEs.

Figure 7-2 Insert and Remove IEs

The SES controller supports Connection Trace and Path trace facility for maximum 10-hops network and maximum 5 traces are allowed to be in progress at a time. The trace log file is recycled every time it reaches 1 MB in size.