Table of Contents

Line Card Groups

Card Types
Common Alarms, Controls, and Indicators

Adapter Cards
ATM Line Module A

ALM/A Features
Connection Types
Traffic Management
ALM/A Operational Parameters
Back Cards for the ALM/A

Universal Voice Module

Introduction to the UVM
List of UVM Features
Modes of UVM Operation
UVM Operation
Channel Pass-Through
UVM Data Operation
Signalling on the UVM
Line Statistics
Loopbacks on the UVM Card Set
UVM Front Card Faceplate
Universal Voice Interface Back Card (BC-UVI-2T1EC)
Universal Voice Interface Back Card (BC-UVI-2E1EC)
Universal Voice Interface Back Card (BC-UVI-2J1EC)
Channel Associated Signalling

CAS Features
CAS Limitations
CAS Switching

Channelized Voice Module (CVM)

CVM Features
Modes of CVM Operation
CVM Voice Operation
CVM Data Operation
Signalling on the CVM
Line Statistics
Loopbacks on the CVM Card Set
CVM Faceplate Description
T1 Interface Back Card (BC-T1)

BC-T1 Faceplate Description

E1 Interface Back Card (BC-E1)
BC-J1 Description
The TDM Transport Feature

Model C Features
Model C Limitations
Inverse Multiplexing

Frame Relay Cards

Introduction
Maximum Connections Per Port With Signalling Protocols
Frame Relay Over T1 and E1 Lines
Universal Frame Module
UFM-C Front Card
UFI-8T1 Back Card
UFI-8E1 Back Cards
UFM-U Front Card

UFM-U Faceplate
The Modes of the Unchannelized Back Cards
Changing the Mode of a UFM-U
Cabling for the UFM-U Back Cards

UFI-12V.35 Back Card
UFI-12X.21 Back Card
UFI-4HSSI Back Card
Frame Relay Module (FRM)

FRM Features and Functions

Frame Relay Card Redundancy
Frame Relay Interface (FRI) V.35 Card

Frame Relay V.35 Port Numbering
FRI-V.35 Data Clocking
Loopbacks
FRM-FRI Compatibility

Frame Relay Interface for X.21

X.21 Data Clocking

Y-Cable Redundancy and Port Modes
Card Self Test
Port Testing (X.21)

Unit Replacement

FRM-2 Interface to the Port Concentrator Shelf

Terminology

Operation and User Interface
PCS-to-IGX Interface

Compatibility

Concentrated Link

Configuration
Activation
upfrport Command
De-activation: the dnfrport Command

PCS Port Configuration

Interface Hardware Configuration
The cnffrport Command

Port Statistics

PCS Monitoring Functions

Collecting the Monitoring Information

Logical Port Speed
Physical Port Speed

PCS Front Panel LEDs
Card Insertion and Removal
PCS Command Summary
PCS Port Failures
Conditioning
PCS General Operation

Firmware Download
Automatic Diagnostics—FRM-2 and FRI-2-X.21 Cards
StrataView Plus Interface
SNMP Manager
User Interface
Concentrated Link Failure

Frame Trunk Module (FTM)

The Cisco Line of Access Devices
The FastPAD Line of Access Devices

Data Cards

High-speed Data Module (HDM)
Synchronous Data Interface Card (SDI)
Low Speed Data Module (LDM)
Low Speed Data Interface Card (LDI)

Optional Peripherals

Line Interface Cards

This chapter describes the hardware and related functions for each line card in an IGX node. The description of each card includes:

• Function
  
  
• System interconnect
  
  
• Faceplate indicators
  
  

A brief description of optional peripherals and third-party equipment appears at the end of the chapter. For system specifications, such as protocols and standards, refer to Appendix A.

For all matters relating to installation, troubleshooting, user-commands, and repair and replacement, refer to the Cisco IGX 8400 Series Installation manual.

Other manuals that relate to IGX operation are:

• The Cisco WAN Switching Command Reference and Cisco WAN Switching SuperUser Command Reference describe standard user commands and superuser commands.
  
  
• The Cisco WAN Switching System Overview contains general Cisco WAN switching network information.
  
  
• The StrataView Plus Operations contains information on Cisco WAN switching network management.
  
  

Line Card Groups

This chapter introduces each of the following line card groups:

• ATM UNI Card Group
  
  
• Voice Card Group
  
  
• Frame Relay Card Group
  
  
• Serial Data Card Group
  
  

Card Types

Table 4-1 lists the front cards that can be used in the IGX. Table 4-2 lists the back cards that can operate in an IGX. In addition to the front and back cards, the IGX uses an Adapter Card Module (ACM). These cards attach to existing IPX 16/32 service module cards and perform the adaptation necessary to allow IPX 16/32 front cards to operate in an IGX 8410, IGX 8420, or IGX 8430 system. Note that this upgrade scheme does not apply to IPX 8-specific cards. Beyond this limitation, IPX back cards can operate in an IGX system without modifications.

Common Alarms, Controls, and Indicators

Front cards and back cards have faceplates with indicator LEDs and, on some front cards, push-button controls. In addition, back card faceplates have the cable connectors. In slots where no back card exists, a blank faceplate must reside to contain Electromagnetic Interference (EMI) and Radio Frequency Interference (RFI) and to ensure correct air flow.

The LED indicators are on the front and back card faceplates. Each plug-in card has both a green ACTIVE LED and a red FAIL LED at the bottom of the faceplate. In general, the meaning of each LED is indicated in Table 4-3. Some other cards have additional indicators, connectors, or controls, which the appropriate sections describe.

Adapter Cards

Cisco  can upgrade IPX service/interface cards for use in an IGX. This allows the IGX to provide all the services of the IPX with cards of proven efficiency, functionality, and reliability. The upgrade is available only as a factory upgrade. The factory upgrade consists of an adding one of three possible Adapter Card Modules (ACM) and possible firmware or hardware modifications. Due to the complexity of the ACM, field upgrades of IPX cards are not possible.

Connecting IPX front cards to their corresponding back cards on the IPX requires the use of a utility or local bus. On upgraded IPX cards (IGX cards), the local or utility bus is not necessary.

The following IPX cards can be adapted for use in the IGX:

• NTC
  
  
• AIT
  
  
• CDP
  
  
• FRP
  
  
• FTC
  
  
• SDP
  
  
• LDP
  
  
• ARC
  
  

ATM Line Module A

The ATM Line Module A (ALM/A) provides an ATM UNI port that allows devices such as routers or ATM switches to support VPCs and VCCs across the IGX. The ALM/A operates as a cell relay interface that supports basic connectivity to IGX interfaces or BPX ASI interfaces through a Cisco Stratacom network. The ALM/A supports ATM connections that terminate at either ATM or frame relay endpoints. With the processing that the IGX provides, cells can cross either an ATM trunk or a FastPacket trunk. This ALM/A description covers the following topics:

Using the addcon command, you can add connections between the following endpoints

Figure 4-1 shows an example of a network using ALMs. The network contains connections from an ALM/A to another ALM/A, a UFM, and an ASI. The dashed lines show virtual connections. The solid lines between the trunk cards and the ATM cloud are trunks.

Figure 4-2 shows the ALM/A faceplate.

ALM/A Features

A general list of ALM/A characteristics follows. (Subsequent sections have more particular information on interworking, traffic management, and back cards.)

• An aggregate throughput of 45 Mbps.
  
  
• PVC rate of 150-96000 cells per second (cps). (Symmetric connections only.)
  
  
• ATM Forum UNI v3.0 and v3.1 headers.
  
  
• CBR, VBR, ATFR, ATFT, and ATFX connection types
  
  
• PLCP DS-3 framing.
  
  
• G.804 E3 framing.
  
  
• Interworking between frame relay and ATM endpoints.
  
  
• Ingress policing.
  
  
• Per VC queuing with an overall capacity of 65535 cells.
  
  
• Up to 1000 connections per card at an aggregate rate up to DS-3.
  
  
• Optional Y-cable redundancy.
  
  
• Support for line and card statistics and alarms.
  
  

Connection Types

The ALM/A supports CBR, VBR, ATFR, ATFT, and ATFX connection types. Thus, the card supports ATM, cell forwarding, and both network and service interworking connections. Table 4-4 shows the card combinations and connections types. The Cisco IGX 8400 Series Installation publication describes where and when to specify the connection type. For more information on the interworking connections, see the Cisco WAN Switching System manual and the addcon description for frame relay and ATM connections in the Cisco WAN Switching Command Reference.

Traffic Management

The ALM/A uses per-VC queue thresholds and credit-based servicing to control the ingress flow of data to the network and guarantee card resources for each connection.

Using addcon for an ATM or cell-forwarding connection, you specify the VC queue depth and the rate for the connection. If you do not specify a queue depth, the system calculates the queue depth for the connection based on the cell rate. The connection rate determines the service credits. If 80% of the VC queue depth is exceeded, cells with CLP=1 are discarded. If a connection's VC queue is full and the cell rate is exceeded long enough, cells are discarded from the affected queue. Therefore, if a high level of bursty traffic expected for given cell rate, specify a deeper queue for a connection.

For cell forwarding connections, you can explicitly assign VC queue depth for each connection. For ATFR connections, you can assign a VC queue depth by choosing or modifying a frame relay class and specifying the queue depth in bytes. Either way, you can over-subscribe the buffer pool on the card. Actually, buffers are dynamically assigned to VC queues as cells arrive and depart. In this scheme, no data is lost unless you over-subscribe and all buffers are occupied. If you do not specify VC queue depth when adding a connection, switch software assigns a percentage of buffers equal to the connection's cell rate divided by the total ALM/A bandwidth. With this approach, over-subscription is not a problem.

ALM/A Operational Parameters

The ALM/A has the following operational parameters:

Back Cards for the ALM/A

The ALM/A T3 and E3 back cards are the BC-UAI-1T3 and BC-UAI-1E3, respectively. Each back card has two BNC connectors and six LED indicators. For the BC-UAI-1T3 faceplate items, refer to Figure 4-3 and Table 4-5 . For the BC-UAI-1E3 faceplate items, refer to Figure 4-4 and Table 4-6. As you read from the top-down, the items in the figure and table correlate to each other. For technical specifications on T3 and E3 lines, refer to the appendix titled "System Specifications."

Universal Voice Module

This section introduces the Universal Voice Module (UVM) and its related back cards. The topics covered in this section are:

• A UVM overview and feature list
  
  
• An introduction to the modes of UVM operation
  
  
• A brief description of UVM voice processing
  
  
• A brief description of UVM data processing
  
  
• UVM signalling support
  
  
• Line statistics
  
  
• Loopbacks on the UVM card set
  
  
• Front card faceplate descriptions
  
  
• Back card descriptions
  
  

For setup instructions, see the Cisco IGX 8400 Series Installation publication and the relevant commands in the Cisco WAN Switching Command Reference (voice-specific commands are in the chapter titled "Voice Connections").

Introduction to the UVM

The UVM card set is a multi-purpose module capable of supporting E1 or T1 lines for carrying voice, data, or both types of traffic on a card. The types of voice coding that the UVM supports are:

• Pulse-code Modulation (PCM)
  
  
• Adaptive Differential Pulse Code Modulation (ADPCM)
  
  
• Low Delay Code-Excited Linear Predictive Coding (LDCELP)
  
  

For voice, a UVM can terminate connections to a:

• UVM
  
  
• CVM (except for connections that use LDCELP compression)
  
  
• CDP (except for connections that use LDCELP compression)
  
  

For voice using PCM or ADPCM, the UVM can operate in either 24-channel mode (T1) or 30-channel mode (E1 or J1). If the compression is LDCELP, the UVM supports 16 channels but can pass the remaining DS0s on a T1, E1, or J1 line to another UVM for processing. You assign the circuits on a UVM hop on a per-timeslot basis within a T1, E1, or J1 frame.

The back cards for the UVM are the BC-UVI-2T1EC, BC-UVI-2E1EC, and BC-UVI-2J1EC. Each back card has two ports. You can use one port for communication with user-equipment and optionally use the other port for passing through unprocessed logical channels to another UVM. (See the forthcoming section titled "Channel Pass-Through .")

List of UVM Features

The features of the UVM are:

• Maximum of 24 channels on the two T1 lines, with echo cancellation
  
  
• Maximum of 32 channels on the two E1 or J1 lines, with echo cancellation
  
  
• Packet assembly/disassembly (PAD) for voice or data connections
  
  
• Software configurable ports on T1, E1, or J1 back cards
  
  
• Local and remote loopbacks for port and circuit testing
  
  
• Self-test of all on-board circuits
  
  
• Integral echo cancelling for all connection types except "t"—up to 32 milliseconds per channel
  
  
• A-law or µ-law voice encoding on a per-channel basis
  
  
• Programmable voice circuit gain in the range -8 dB through +6 dB
  
  
• Support for many domestic and international signalling types
  
  
• Flexible signalling-bit conditioning when a circuit alarm occurs
  
  
• Per-channel tone detection to disable compression for modem or FAX circuits
  
  
• Standard rate data connections (single 64-K channels only—no aggregation of DS0s)—main use is for CCS transport
  
  
• PCM at 64 Kbps, on a maximum of 32 channels per card
  
  
• LDCELP at 16 Kbps, on a maximum 16 channels per card
  
  
• CCS signalling on the lines
  
  
• CAS signalling on the lines
  
  
• Voice activity detection (VAD), which also increases compression ratios.
  
  
• Y-cable redundancy
  
  

Modes of UVM Operation

This section describes the operational parameters you can specify for the UVM card set. Table 4-7 is a list of the types of UVM operation. The sections that follow further describe the listed subjects.

The first two entries in Table 4-7 are the two types of 64-Kbps operation. The "p" indicates uncompressed PCM voice. The "t" indicates 64 Kbps clear channel data. The third entry is "v," which indicates VAD. The table then shows voice and data modes, transmit/receive rates, and indications of compression and ZCS (zero code suppression).

A "c" or an "a" preceding a numerical value indicates compression. A "c" indicates compression with voice activity detection (VAD), so "c" does not apply to data connections. An "a" indicates compression without VAD. The "a" can apply to either voice or data. The numerical value following the "a" or "c" is the bit rate. Table 4-7 also explains the significance of the other characters that may follow the bit rate.

Standard-rate (64 Kbps) voice connections terminate on CVM or UVM cards. (The CVM can terminate a connection from a UVM only if the connection does not use LDCELP.) You can configure voice compression, echo cancelling for each channel, and A-law or µ-law encoding (in appropriate circumstances). You can select voice frequency compression of 64 Kbps (no compression), 32 Kbps (2:1 compression), 24 Kbps (3:1 compression), or 16 Kbps (4:1 compression).

UVM Operation

The particular voice features are:

• High-speed modem and FAX circuits.
  
  
• Transparent, 64-Kbps transport of CCS signalling on T1 lines.
  
  
• CAS, by transporting signalling transitions across the network in packets.
  
  
• Optional circuit for the UVM is the on-board Integrated Echo Canceller (IEC): the two models of the IEC are the 24-channel (T1) and the 32-channel (E1) versions.
  
  
• Selectable A-law or µ-law encoding when the circumstances require it. (Normally, T1 automatically uses µ-law, and E1 uses A-law.)
  
  

Channel Pass-Through

The UVM can support mixed connection types. For example, you can add 12 connections that use ADPCM and 6 that use LDCELP. Any combination is possible as long as the configuration does not exceed the capacity of the card. If the number and types of connections exceeds the capacity of the UVM, a second UVM is necessary to take the excessive channels. The name of this approach is pass-through. In this scheme, a primary UVM connects through a cable to a secondary UVM. The primary passes the unprocessed channels to the secondary UVM. The primary application of pass-through is to LDCELP.

The maximum number of channels that can carry LDCELP data is 16, so not all possible channels on a T1 or E1 line are supported if all channels are using LDCELP. To support the other channels, the UVM can pass-through the remaining channels to a second UVM. Pass-through, for example, lets a PBX with a T1, E1, or J1 trunk use all of its bandwidth when the specified compression is LDCELP. In addition to the secondary UVM, you must have an additional cable to connect the two back cards. The command for specifying pass-through is cnflnpass. For a description of how to set up pass-through, refer to the Cisco IGX 8400 Series Installation publication.

Figure 4-5 shows possible arrangements of UVMs with T1 lines. Figure 4-6 shows possible arrangements for E1 or J1 lines. In Figure 4-5, Example A shows UVM 1 set to pass unsupported channels to UVM 2. Example B shows no pass-through, so the UVM is supporting a number of channels (of any combination) that does not exceed its capacity. Example C shows the possibility of two T1 lines to be fully supported on three UVMs (with all channels using LDCELP). In Figure 4-6, Example A shows no pass-through, and Example B shows the pass-though configuration.

UVM Data Operation

A UVM can provide data connections to the network. Data connections that originate on a UVM can terminate on a CVM or a UVM. The only category of data connection the UVM supports is a standard 64-Kbps connection.

Signalling on the UVM

The UVM extracts information from the signalling bits in the T1, E1, or J1 frame. When a signalling bit changes state, the UVM generates signalling packets to the card at the other end of the connection. DPNSS and ISDN signalling are supported through a clear channel (transparent mode). The supported combinations of signalling and line configurations are as follows:

• T1 can use CCS, ABAB, or ABCD signalling bits when the line framing is ESF.
  
  
• T1 can use CCS or AB signalling bits when the line framing is D4.
  
  
• T1 can use CCS in timeslot 24 on applicable PBXs if the application requires CAS to be off.
  
  
• E1 and J1 use either CCS or, if you selected CAS, ABCD signalling bits. For CCS, configure channel 16 as a t-type connection to carry the signalling, making a possible total of 31 channels.
  
  

Use the cnfcln command to configure the signalling and dsplncnf to see the configured signalling.

Up to 23 voice interface types, such as 2-W E&M, FXO/FXS, or DPO/DPS, can be selected from a template to condition the VF signalling. You can also specify customized signal conditioning. Voice channel signalling is programmable for any of the following:

• Robbed bit (either D4 or ESF frame format)—for T1 lines
  
  
• Channel Associated Signalling (CAS)—for E1 lines
  
  
• Transparent CCS (ISDN & DPNSS)—for T1 or E1 lines
  
  
• E&M-to-DC5A and DC5A-to-E&M conversion—for international applications
  
  

Line Statistics

The UVM card set monitors and reports statistics on the following input line conditions:

• Loss of signal
  
  
• Frame sync loss
  
  
• Multi-frame sync loss
  
  
• CRC errors (for E1 only)
  
  
• Frame bit errors
  
  
• Frame slips
  
  
• Bipolar violations
  
  
• Frame bit errors
  
  
• AIS—all 1s in channel 16 (CAS mode)
  
  
• Remote (yellow) alarm
  
  

Loopbacks on the UVM Card Set

You can set up local and remote loopbacks to check UVM-terminated connections. A local loopback functions at the system bus interface. It returns data and supervision back to the local facility to test the local UVM card set and the customer connection. Remote loopbacks extend to the UVM card at the other end and check both transmission directions and much of the far-end UVM.

UVM Front Card Faceplate

Table 4-8 and Figure 4-7 describe the front card faceplate.

Universal Voice Interface Back Card (BC-UVI-2T1EC)

The BC-UVI-2T1EC back card provides a T1 line interface for a UVM card. The BC-UVI-2T1EC plugs into the P2 connector on the UVM front card. The card set can reside in any non-reserved slot. The BC-UVI-2T1EC provides the following features:

The BC-T1 supports two clock modes. The clock modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network.

Figure 4-8 and Table 4-9 provide information on the faceplate of the BC-UVI-2T1EC. When you correlate the descriptions in the table with the callouts in the figure, read from the top of the table to the bottom. The standard port connector is a female DB15.

Universal Voice Interface Back Card (BC-UVI-2E1EC)

The BC-UVI-2E1EC back card provides an E1 circuit line interface for a CVM card. The BC-UVI-2E1EC plugs into the P2 connector on the CVM front card. The card set can reside in any non-reserved slot. The BC-UVI-2E1EC provides the following features:

The BC-UVI-2E1EC supports two clock modes. The clock modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network.

Statistics are kept on most line errors and fault conditions, including the following:

• Loss of signal
  
  
• Frame sync loss
  
  
• Multi-frame sync loss
  
  
• CRC errors
  
  
• Frame slips
  
  
• Bipolar violations
  
  
• Frame bit errors
  
  
• AIS, all-1's in channel 16 (CAS mode)
  
  

Figure 4-9 shows and Table 4-10 lists status LEDs and connections on the BC-UVI-2E1EC faceplate. When you correlate the table and figure items, read from the top to the bottom.

Universal Voice Interface Back Card (BC-UVI-2J1EC)

The BC-UVI-2J1EC back card provides an J1 circuit line interface for a CVM card. The BC-UVI-2J1EC plugs into the P2 connector on the CVM front card. The card set can reside in any non-reserved slot. The BC-UVI-2J1EC provides the following features:

The BC-UVI-2J1EC supports two clock modes. The clock modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network.

Statistics are kept on most line errors and fault conditions, including the following:

• Loss of signal
  
  
• Frame sync loss
  
  
• Multi-frame sync loss
  
  
• CRC errors
  
  
• Frame slips
  
  
• Bipolar violations
  
  
• Frame bit errors
  
  
• AIS, all-1's in channel 16 (CAS mode)
  
  

Figure 4-10 shows and Table 4-11 lists status LEDs and connections on the BC-UVI-2J1EC faceplate. When you correlate the table and figure items, read from the top to the bottom.

Channel Associated Signalling

This section applies only to UVMs that are Model C. Model C provides a service called Channel Associated Signaling (CAS). CAS enables the IGX to convert CAS signalling and DTMF tones to CCS call control messages. CAS routes calls from a PBX over the WAN using SVCs under the control of a VNS. For setup instructions, see the relevant commands in the Cisco WAN Switching Command Reference.

CAS Features

The CAS features are as follows:

• Converts CAS signalling to CCS call control messages.
  
  
• AB bit signalling for 2-wire E&M lines is supported.
  
  
• A-law for T1 lines is supported.
  
  

CAS Limitations

The limitations on CAS switching within Model C firmware are as follows:

• Pass-through configuration is not supported.
  
  
• Y-redundancy on the CAS-switching UVM card that is connected to the PBX cannot provide hot standby.
  
  
• Only one line on the UVM card can be used for CAS-switching.
  
  
• A UVM with CAS switching is compatible only with other UVMs or CVMs that have Model C firmware.
  
  
• All UVMs in the CAS-Switching network must use Model C firmware.
  
  

CAS Switching

CAS switching is designed to enable the IGX to convert CAS signalling and DTMF tone to CCS call control messages.

The converted CCS messages for all channels on the line are carried in one regular t type or special t type PVC connection from the UVM to another UVM/CVM card. The VNS is configured to receive the CCS messages from t-type or td-type PVC connections from the signalling channel of CAS-switching UVM cards on the network.

When a CCS call control message is received, the VNS determines the destination from the message, and issues SNMP commands to add a connection from a channel on the CAS switching line to the destination on the network. Figure 4-11 illustrates CAS switching on the IGX.

Channelized Voice Module (CVM)

This section introduces the Channelized Voice Module (CVM) and its related back cards. The topics covered in this section are as follows:

• A CVM overview and feature list
  
  
• An introduction to the modes of CVM data and voice processing
  
  
• A brief description of CVM voice processing
  
  
• A brief description of CVM data processing
  
  
• CVM signalling support
  
  
• Loopbacks on the CVM card set
  
  
• Front card and back card faceplate descriptions
  
  
• TDM Transport (support for older TDM equipment)
  
  

For setup instructions, see the Cisco IGX 8400 Series Installation and the relevant commands in the Cisco WAN Switching Command Reference (voice-specific commands are in the chapter titled "Voice Connections").

Introduction to the CVM

The CVM is a multi-purpose front card. A CVM circuit line can carry the following types of traffic:

• Voice
  
  
• Data
  
  
• A combination of voice and either low-speed or high-speed data
  
  

The default mode for a channel on a CVM is voice. In addition to other CVMs, a CVM can communicate with a CDP in an IPX or one of the ports on a UVM. Figure 4-12 illustrates different traffic configurations. You can use the CVM in either 24-channel mode (T1) or 30-channel mode (E1). You assign the circuits on a CVM hop on a per-timeslot basis within a T1 or E1 frame.

The CVM cards can reside in any non-reserved slot in an IGX. The front card operates with either a BC-T1, BC-E1, or BC-J1 back card. For details on back cards, see the forthcoming sections titled "T1 Interface Back Card (BC-T1) ," "E1 Interface Back Card (BC-E1)," or "BC-J1 Description ."

CVM Features

The following is a list of CVM features.

• Packet assembly/disassembly (PAD) for voice or data connections
  
  
• Software configurable ports on T1, E1, or J1 back cards
  
  
• Support for Dynamic Network Switching (DNS)
  
  
• Local and remote loopbacks for port and circuit testing
  
  
• Self-test of all on-board circuits, including optional echo cancellers
  
  
• Y-cable redundancy
  
  
• Up to 4:1 voice compression using adaptive differential pulse code modulation (ADPCM) with integral voice activity detection (VAD)
  
  
• Integral, per-channel, echo cancelling (requires optional circuits on the front card)
  
  
• A-law or µ-law voice encoding on a per-channel basis
  
  
• Programmable voice circuit gain in the range -8 dB through +6 dB
  
  
• Support for many domestic and international signalling types
  
  
• Flexible signalling-bit conditioning when a circuit alarm occurs
  
  
• Per-channel tone detection to disable compression for modem or FAX circuits
  
  
• Sub-rate, standard rate, and super-rate data connections
  
  
• Support for transparent TDM channels through a network
  
  
• A transparent data channel for CCS
  
  

Modes of CVM Operation

This section describes the operational parameters you can specify for the CVM card set. Table 4-12 is a list of the types of CVM operation. The sections that follow further describe the listed subjects.

The first two entries in Table 4-12 are the two types of 64-Kbps operation. The "p" indicates uncompressed PCM voice. The "t" indicates 64 Kbps clear channel data. The table then shows voice and data modes, transmit/receive rates, and indications of compression and ZCS (zero code suppression). A "c" or an "a" preceding a numerical value indicates compression. A "c" indicates compression with voice activity detection (VAD), so "c" does not apply to data connections. An "a" indicates compression without VAD. The "a" can apply to either voice or data. The numerical value following the "a" or "c" is the bit rate. Table 4-12 also explains the significance of the other characters that may follow the bit rate.

Standard-rate (64 Kbps) voice connections terminate on CVM or UVM T1 or E1 lines. (The CVM can terminate a connection from a UVM only if the connection does not use LDCELP.) You can configure the voice compression, and echo cancelling for each channel, and-when circumstances make it appropriate-A-law or µ-law encoding. You can select voice frequency compression of 64 Kbps (no compression), 32 Kbps (2:1), 24 Kbps (3:1), or 16 Kbps (4:1). The compression ratios approximately double when you also enable the internal VAD.

CVM Voice Operation

The particular voice features are:

• High-speed modem and FAX circuits.
  
  
• Transparent, 64-Kbps transport of CCS signalling.
  
  
• CAS, by transporting signalling transitions across the network in packets.
  
  
• Optional circuit for the CVM is the on-board Integrated Echo Canceller (IEC): the two models of the IEC are the 24-channel (T1) and the 31-channel (E1) versions.
  
  
• Selectable A-law or µ-law encoding when the circumstances require it. (Normally, T1 automatically uses µ-law, and E1 uses A-law.)
  
  

CVM Data Operation

A CVM can provide data connections to the network. Data connections that originate on a CVM can terminate on either a CVM, UVM, HDM, LDM, CDP (on IPX), SDP (IPX), or LDP (IPX) card set.

Two categories of data connections exist on the CVM. The categories are super-rate and subrate. A super-rate connection is an aggregate of channels that functions as a single, logical connection. A super-rate connection can consist of any combination of 56 or 64 Kbps-connections up to a maximum of 8 connections totalling 512 Kbps. The DS0 timeslots must be contiguous or alternating (not random). All 56-Kbps data channels are bit-stuffed up to 64 Kbps on a circuit line. The CVM removes the bits prior to packetization. Note that super-rate connections carry no supervisory bits.

A subrate data connection has a rate less than 64 Kbps and exists within a DS0. Supported rates are 2.4, 4.8, 9.6, and 56 Kbps. The type of subrate data connection that Cisco WAN switching equipment supports is DS0A.

Signalling on the CVM

The CVM extracts information from the signalling bits in the E1 or T1 frame. When a signalling bit changes state, the CVM generates signalling packets to the CVM at the other end of the connection. DPNSS and ISDN signalling are supported through a clear channel (transparent mode). The supported combinations of signalling and line configurations are as follows:

• T1 uses ABAB or ABCD signalling bits when the line framing is ESF.
  
  
• T1 uses AB signalling bits when the line framing is D4.
  
  
• T1 can use CCS in timeslot 24 on applicable PBXs if the application requires CAS to be off.
  
  
• E1 uses either CCS or, if you selected CAS, ABCD signalling bits. For CCS, configure channel 16 as a t-type connection to carry the signalling, making a possible total of 31 channels.
  
  

Up to 23 voice interface types, such as 2-W E&M, FXO/FXS, or DPO/DPS, can be selected from a template to condition the VF signalling. You can also specify customized signal conditioning. Voice channel signalling is programmable for any of the following:

• Robbed bit (either D4 or ESF frame format)—for T1 lines
  
  
• Channel Associated Signalling (CAS)—for E1 lines
  
  
• Transparent CCS (ISDN & DPNSS)—for T1 or E1 lines
  
  
• E&M-to-DC5A and DC5A-to-E&M conversion—for international applications
  
  

Line Statistics

The CVM card set monitors and reports statistics on the following input line conditions:

• Loss of signal
  
  
• Frame sync loss
  
  
• Multi-frame sync loss (for E1 only)
  
  
• CRC errors (for E1 or T1 BSF only)
  
  
• Frame bit errors
  
  
• Frame slips
  
  
• Bipolar violations
  
  
• AIS—all 1s in channel 16 (CAS mode)
  
  
• Remote (yellow) alarm
  
  

Loopbacks on the CVM Card Set

You can set up local and remote loopbacks to check CVM-terminated connections. A local loopback functions at the system bus interface. It returns data and supervision back to the local facility to test the local CVM card set and the customer connection. Remote loopbacks extend to the CVM card at the other end and check both transmission directions and much of the far-end CVM.

CVM Faceplate Description

The CVM faceplate has four LED indicators: ACTIVE, FAIL, MAJOR, and MINOR (see Table 4-13). The label on the faceplate also shows the type of CVM card.

T1 Interface Back Card (BC-T1)

The BC-T1 back card provides a T1 line interface for a CVM card. The BC-T1 plugs into the P2 connector on the CVM front card. The card set can reside in any non-reserved slot. The BC-T1 provides the following features:

The BC-T1 supports two clock modes. The clock modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network.

BC-T1 Faceplate Description

Figure 4-13 and Table 4-14 provide information on the faceplate of the BC-T1. When you correlate the descriptions in the table with the callouts in the figure, read from the top of the table to the bottom. The standard port connector is a female DB15.

E1 Interface Back Card (BC-E1)

The BC-E1 back card provides an E1 circuit line interface for a CVM card. The BC-E1 back card plugs into the P2 connector on the CVM front card. The card set can reside in any non-reserved slot. The BC-E1 provides the following features:

The BC-E1 supports two clock modes. The clock modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network.

Statistics are kept on most line errors and fault conditions, including the following:

• Loss of signal
  
  
• Frame sync loss
  
  
• Multi-frame sync loss
  
  
• CRC errors
  
  
• Frame slips
  
  
• Bipolar violations
  
  
• Frame bit errors
  
  
• AIS, all-1's in channel 16 (CAS mode)
  
  

Figure 4-14 shows and Table 4-15 lists status LEDs and connections on the BC-E1 faceplate. When you correlate the table and figure items, read from the top to the bottom.

BC-J1 Description

The BC-J1 back card provides a Japanese J1 circuit line interface for a CVM card. The BC-J1 back card plugs into the P2 connector on the CVM front card. The card set can reside in any non-reserved slot. The BC-J1 provides the following features:

The BC-J1 supports two clock modes. The modes are normal clocking and loop timing. You select the mode through software control. With normal clocking, the node uses the receive clock from the network for the incoming data and supplies the transmit clock for outgoing data. The node can use the receive clock to synchronize itself with the network. With loop timing, the node uses the receive clock from the network for incoming data and redirects the receive clock to time the transmit data.

Figure 4-15 shows a BC-J1 faceplate. Table 4-16 lists BC-J1 connections and status LEDs.

Figure 4-15: BC-J1 Faceplate
 

The TDM Transport Feature

This section applies only to CVMs (and CDPs in the IPX) that are Model C. Model C provides a service called Time Division Multiplexing Transport (TDM Transport). TDM Transport bundles DS0s to form a single, transparent connection through the network. TDM Transport is most valuable for transporting TDM data from trunks in older, non-Cisco WANs. For setup instructions, see the Cisco IGX 8400 Series Installation and the relevant commands in the Cisco WAN Switching Command Reference.

Model C Features

The Model C firmware features are as follows:

• A collection of properly configured data connections behaves as a single, transparent connection. A single connection can be from 1 to 8 DS0s (64 to 512 Kbps).
  
  
• TDM Transport supports inverse multiplexing support.
  
  
• The maximum rate is 2 Mbps.
  
  
• The supported line coding is 8/8 on the circuit and trunk interfaces.
  
  
• TDM Transport service preserves DS0 alignment within frames.
  
  

Model C Limitations

The limitations on TDM Transport within Model C firmware are as follows:

• TDM Transport supports bundled data connections only—no voice or DS0A.
  
  
• Circuit timing cannot come from user equipment.
  
  
• Pleisochronous clocking is not allowed.
  
  
• Super-frame alignment is not preserved.
  
  
• Framing bits are not transported.
  
  
• A CDP or CVM with Model C firmware is compatible only with other CDPs or CVMs that have Model C firmware.
  
  
• Connections on any CVM or CDP with Rev. C firmware must terminate on only one other CVM or CDP.
  
  
• The parameter "Max. Network Delay of CDP-CDP High Speed Data Connections" must not exceed 45 ms.
  
  
• The difference in delay due to propagation and maximum trunk queuing cannot exceed 25 ms. For example, one connection taking one hop and another taking six hops is not allowed. Likewise, a connection over a satellite link and another over a terrestrial link is not allowed.
  
  
• For loopbacks on a CDP or CVM with Rev. C firmware, the card must have only one connection.
  
  

Inverse Multiplexing

To achieve full T1 and E1 rates on a single CVM or CDP, TDM Transport supports inverse multiplexing. Figure 4-16 shows a simple example. In this example, the three, bundled 512-Kbps data connections symbolized by the arrows add up to a T1 connection.

Frame Relay Cards

This section describes the frame relay service cards. The frame relay cards consist of the UFM (Universal Frame Module) family of front and back cards and the FRM (Frame Relay Module) family of front and back cards. The first part of this section contains information that applies to both the UFM and FRM card sets. Subsequent sections contain information that is particular to individual front and back cards. The frame relay and individual card topics are:

• Introduction to frame relay on the IGX switch
  
  
• UFM front cards
  
  
• UFI back cards, which have the T1, E1, V.35, X.21, and HSSI interfaces
  
  
• FRM front cards, which exists in a channelized and an unchannelized version
  
  
• FRI back cards, which have the T1, E1, V.35, and X.21 interfaces
  
  
• FRM-2 front card and FRI-2 back card, which support the Port Concentrator Shelf (PCS)
  
  
• Frame relay port redundancy
  
  
• Card self-test
  
  
• Port Testing
  
  

Refer to the Cisco IGX 8400 Series Installation publication for installation steps. For technical details on the line types of individual back cards, see the appendix titled "System Specifications." For a description of frame relay on a FastPAD, see the section titled "Data Cards" later in this chapter.

Introduction

This section lists information that applies to frame relay service. The first list describes functions at the node-level and the network interface. The second list contains card-specific information.

An IGX frame relay network features the following:

• A Permanent Virtual Circuit (PVC) service.
  
  
• Three Local Management Interface (LMI) protocols: ANSI Annex D, the ITU-T (CCITT) Annex A, and a StrataCom LMI. LMI protocols operate between the user device and the network. LMI is the only supervisory signalling on a logical port.
  
  
• A frame-forwarding service. This service forwards all valid HDLC frames from one port to a single port in the network without frame relay header processing or LMI control. For example, a workstation that transmits HDLC frames can connect to a single frame relay port on an FRM. To use this feature, the entire physical port must be dedicated to that workstation.
  
  

The FRM and UFM front cards perform the following functions:

  On the receiving end, the UFM, FRM, or FRM-2 card set checks arriving frames against the embedded Frame Check Sequence (FCS) code and discards any non-compliant frames.

Maximum Connections Per Port With Signalling Protocols

For any frame relay card set that has a maximum frame length of 4510 bytes, the type of signalling protocol you may (optionally) specify with cnffrport results in a limit on the number of connections per physical or logical port. The maximum number of connections per port for each protocol is:

• For Annex A: 899
  
  
• For Annex D: 899
  
  
• For StrataLMI: 562
  
  

Neither addcon nor cnffrport prevents you from adding more than the maximum number of connections on a port. (You might, for example, use cnffrport to specify an LMI when too many connections for that particular LMI already exist.) If the number of connections is exceeded for a particular LMI, the LMI does not work on the port, the full status messages that result are discarded, and LMI timeouts occur on the port. A port failure results and also subsequently leads to a-bit failures in other segments of the connection path.

Frame Relay Over T1 and E1 Lines

On an IGX, frame relay over a T1 or E1 line requires one of the following combinations:

• FRM front card with an FRI-T1 or FRI-E1 back card
  
  
• UFM-4C or UFM-8C front card with UFI-8T1-DB15, UFI-8E1-DB15, or UFI-8E1-BNC
  
  

A frame relay T1/E1 connection can terminate on any Frame Relay Interface—V.35, X.21, T1 or E1.

Frame relay over a T1 or E1 interface supports groups of DS0 timeslots in a logical port. A logical port is a single DS0 or a group of contiguous DS0s. Logical ports consisting of multiple DS0s operate at the full rate of 64 Kbps per timeslot. If a logical port consists of a single DS0, you can configure either 56 Kbps or 64 Kbps. See Figure 4-17.

If the rate on a logical port is 56 Kbps, the frame relay interface strips the least significant (signalling) bit from the incoming octet and puts a 1 in the least significant bit of the outgoing octet. The 56-Kbps rate typically applies to a groomed DDS circuit that uses a T1/E1 line.

Multiple and Single DS0s Forming a Logical Port

Universal Frame Module

The Universal Frame Module (UFM) cards support a much higher port density than the FRM cards and support ELMI and frame relay-to-ATM service interworking. The front card (UFM-4C or UFM-8C) can support frame relay traffic on a maximum of 4 or 8 T1 or E1 ports on a back card. The front card (UFM-U) can support frame relay traffic on 12 V.35 ports, 12 X.21 ports, or 4 HSSI ports on a back card.

The next sections describe the UFM-C card sets. Subsequent sections describe the UFM-U card sets.

UFM-C Front Card

The Universal Frame Model (UFM-C) supports either 4 or 8 T1 or E1 ports per back card. The UFM-4C supports 4 ports regardless of the presence of the 8 connectors on the UFI back card. The UFM-8C supports all 8 ports. Both models of the UFM-C support 1 to 24 or 1 to 31 DS0s per physical line.

The UFM-C can also operate unchannelized for E1 only, with 32 DS0s constituting one unchannelized E1. In the unchannelized mode, one logical E1 port maps to one E1 line.

Table 4-17 lists the front and back cards described in this and subsequent sections.

Figure 4-18 shows a UFM-C faceplate. If the label on the faceplate shows "UFM C," the front card is a UFM-8C, which most UFMs are. The UFM-4C label shows "UFM-4C."

UFI-8T1 Back Card

The Universal Frame Interface 8T1 (UFI-8T1-DB15) back card has 8, bi-directional, DB15 connectors. See Figure 4-19. For each port, one tri-color LED displays the status, as the emphasized area of Figure 4-19 shows. The line is inactive if the LED is off. See Table 4-19 .

UFI-8E1 Back Cards

Universal Frame Interface 8E1 cards have connectors for 8 E1 ports. The UFI-8E1-DB15 has 8 bi-directional DB15 connectors. The UFI-8E1-BNC has 16 BNC connectors (2 per port). See Figure 4-20. Each port has a tri-color LED for status display. See Table 4-20 . The line is inactive if the LED is off.

UFM-U Front Card

This section contains the following:

• An outline of the operational features of the UFM-U card set.
  
  
• A table and description of the user-selected modes of the UFM-U card set
  
  
• A table and description of the cables in the UFM-U card set
  
  
• An illustration and explanation of the UFM-U faceplate with its LED indicators
  
  

• Supports 12 V.35 ports, 12 X.21 ports, or 4 HSSI ports on a back card.
  
  
• Maximum port speeds of 16 Mbps on the HSSI and 10 Mbps on the V.35 and X.21 interface.
  
  
• The UFM-U can sustain an aggregate throughput of up to 16 Mbps.
  
  
• The sum of all clock rates can be 24 MHz even though the actual throughput is 16 Mbps.
  
  
• Supports up to 1000 channels. You can allocate the channels across the frame relay data streams.
  
  
• Provides delimiting, alignment, and transparency (bit-stuffing) for HDLC frames.
  
  
• Meets ANSI T1.618 (a 2-octet DLCI Frame Relay Data Transfer Protocol is supported).
  
  
• Supports E-LMI (enhanced LMI), StrataLMI, T1.617 Annex D, and CCITT Q.933 Annex A Frame Relay Signaling protocols.
  
  
• The front card maps, segments and reassembles frame relay data streams to and from FastPackets by using protocols and methods compatible with an FRP or FRM.
  
  
• Supports frame relay UNI or NNI on a per channel basis.
  
  
• The UFM-U generates CLLM messages for congestion notification across NNIs and UNIs.
  
  
• Supports policing on the ingress direction.
  
  
• Supports Explicit Congestion Notification.
  
  
• Applies zero-suppression to the payload space of FastPackets.
  
  
• Collects statistics for ports and connections.
  
  
• Supports internal and external loopback terminations.
  
  
• Supports per PVC loopback termination at the Cellbus.
  
  
• Supports frame relay to ATM service interworking.
  
  
• Detects and discards frames that are corrupted during transmission on the IGX.
  
  
• Supports frame forwarding.
  
  
• Supports CIR=0 as well as a CIR of less than 56 Kbps.
  
  
• Supports looped clocks (V.35 only).
  
  
• Supports Y-cable redundancy on all ports for V.35 and X.21, one port on HSSI.
  
  
• The UFM-U monitors port speed. It allows up to 2% over-speed for data rates above 1 Mbps and 5% over-speed for data rates below 1Mbps. Over-speed causes a minor alarm.
  
  

UFM-U Faceplate

Figure 4-21 shows the faceplate of the UFM-U. Table 4-21 describes its LED indicators. The first two columns of Table 4-21 shows the names of the LEDs and possible combinations of on and off states. The "Description" column describes the meaning of the LED states.

The Modes of the Unchannelized Back Cards

This section describes how to specify active ports and the maximum speed allowed on each active port. Specifying the maximum speed for active ports requires careful planning, and this section provides the information for understanding and planning the port specification.

The hardware does not permit random combinations of speeds across the ports. Speeds are constrained to certain combinations of maximum rates and ports. The combination of maximum speeds and active port number is called the mode of the card. Table 4-22 shows the maximum bit rate on each port for the modes and additionally groups the ports under a letter. If you need to change the mode of a card after connections have been added, you must consider these port groups. The forthcoming section titled "Changing the Mode of a UFM-U " describes the application of port groups.

In the port column of Table 4-22, the numeric value is the maximum bit rate for that port. You specify the actual bit rate for a port with either the cnfport or cnffrport command. In Table 4-22:

• 3 means 3.072 Mbps.
  
  
• 8 means 8.192 Mbps.
  
  
• 10 means 10.240 Mbps.
  
  
• - Indicates the port is not usable for this mode.
  
  

Changing the Mode of a UFM-U

Initially, an unchannelized UFM card set comes up in mode 1. To specify another mode, use either cnfmode or cnfufmumode. The preferable time to specify a mode is before you add connections. After connections exist on the card, you must delete some or all connections and down ports before you change the mode. To change the mode of a card set, you must first delete all the connections in a group and down all the active ports in a group if any maximum port speed changes in a group as a result of the mode change. This requirement means you may or may not have to delete connections in a particular group. Use Table 4-22 to follow two examples that illustrate what you must do before you change modes.

1. If you change from mode 1 to mode 4 and connections exist on ports 1, 3, and 9-12, you must delete connections on 1 and 3 only and down these ports (no change in maximum speed on 9-12).

2. If you change from mode 2 to 9 and connections exist on all active ports, you must remove connections from ports 1 and 3 then down these ports (and relocate the connections from port 3 to port 1).

The steps for changing the mode of a UFM-U when connections exist on the card is as follows:

Step 1   Use the delcon command to remove all connections in any group where any maximum port speed changes because of the mode change.

Step 2   Use the dnport command to deactivate all ports in any group where any maximum port speed changes because of the mode change.

Step 3   Use the cnffrport command to configure new port speeds (at or below the new maximum resulting from the mode change).

Step 4   Use the cnfmode command to change the mode.

Step 5   Use the upport command to activate the inactive ports that are operational under the new mode.

Step 6   Use addcon to add connections as needed.

Cabling for the UFM-U Back Cards

This section describes the standard cabling and the Y-cabling scheme for the unchannelized back cards. The back cards are the UFI-12V.35, UFI-12X.21, and UFI-4HSSI. Because of the port variety and because the cables are either DCE or DTE, a significant number of cable configurations exist. Table 4-23 shows cable names, part numbers, and descriptions.

The connectors on the UFI cards are high density. Each UFI-12V.35 and UFI-12X.21card has 6, 60-pin connectors. Each of these connectors has two ports. If your specification dictates that both ports on a connector are active, you must have a cable with wiring for two ports. If a part number in Table 4-23 has the form "CAB-2...," the cable has wiring for two ports. Note that if both ports on a connector are active, both ports must be either DCE or DTE because the cable itself is either DCE or DTE. If a connector has only one active port, the cable can be the less expensive, single-port version. A HSSI connector has 50 pins and supports one port.

Port redundancy through the use of a Y-cable requires two card sets, at least one Y-cable, and the standard cabling. On a UFI-4HSSI, Port 1 is the only port that supports Y-cable redundancy. On the V.35 and X.21 UFIs, redundancy at more than one connector requires a corresponding number of Y-cables and standard cables. The standard cables are either single-port for HSSI, V.35, or X.21 or dual-port for the X.21 and V.35 UFIs.

To set up the Y-cabling, you attach a branch of the "Y" at the primary and redundant connectors then attach the near end of the standard cable to the base of the "Y." Figure 4-22 illustrates single and dual-cabling in a Y-cable setup. The single-cable arrangement applies to HSSI, V.35, and X.21. The dual-port redundancy setup applies to a UFI-12X.21 or UFI-12V.35, where both ports within the UFI connector are used in a Y-cable redundancy scheme.

UFI-12V.35 Back Card

The UFI-12V.35 has six connectors. Each connector contains two V.35 ports. Each port has a tri-color LED to indicate its status. Because each connector supports two ports, each connector has two associated LEDs. Figure 4-23 shows the faceplate of the UFI-12V.35, and Table 4-24 describes the significance each color of the LED.

You can configure each port on the UFI-12V.35 to use normal clocking or a looped clocking, then configure the port to run at one of the following speeds:

• 56 Kbps
  
  
• n x 64 Kbps (where n = 1 ... 32)
  
  
• n x 1.024 Mbps (where n = 2 ... 10)
  
  

UFI-12X.21 Back Card

The UFI-12X.21 has six connectors. Each connector contains two X.21 ports. Each port has a tri-color LED to indicate its status. Because each connector supports two ports, each connector has two associated LEDs. Figure 4-24 shows the faceplate of the UFI-12X.21, and Table 4-25 describes the significance each color of the LED.

You can configure each port on the UFI-12X.21 to use normal clocking or a looped clocking, then configure the port to run at one of the following speeds:

• 56 Kbps
  
  
• n x 64 Kbps (where n = 1 ... 32)
  
  
• n x 1.024 Mbps (where n = 2 ... 10)
  
  

UFI-4HSSI Back Card

The UFI-4HSSI has four connectors. Each connector has a tri-color LED for status. Figure 4-25 shows the UFI-4HSSI faceplate, and Table 4-26 describes the significance each color of the LED.

Do not connect two native DCE ports using a HSSI standard straight cable. Connecting two native DCE ports with a HSSI standard straight cable could permanently damage the interface. Interfaces on the UFI-4HSSI back card are already in DCE mode (default) so you can directly connect any DTE interface to the back card using a straight pin-to-pin HSSI standard cable.

---

Note The UFI-4HSSI back card can be configured in DTE mode by using the HSSI-DTE cable to connect back cards in DTE mode to DCE interfaces.

---

Using cnffrport, you can configure each port with a speed that is a multiple of 1 Mbps up to the maximum for the mode of the card (see "The Modes of the Unchannelized Back Cards").

Figure 4-25: UFI-4HSSI Faceplate

Table 4-26: UFI-4HSSI Faceplate Indicators

Green	Indicates line is active and functional.
Red	Indicates line is active but either has no cable or the wrong cable or is running over-speed.
Yellow	Indicates the line is active and in loopback mode.
Off	Indicates that either the port is inactive or has no power.

Frame Relay Module (FRM)

The Frame Relay Module (FRM) front card supports 1 to 4 data ports and, in single-port mode, operates at up to 2.048 Mbps.

---

Note FRM front cards exist in two forms. One uses an ACM1 adaptor. The other is a single-card or "native" version. Functionally they are identical. For the single-card version, the minimum firmware revision is "V."

---

Table 4-27: Frame Relay Module (FRM) Card Sets

Front Card	Model	Back Card	Model
FRM Frame Relay Module, unchannelized	D	FRI-V.35 FRI-X.21	B A
FRM Frame Relay Module, channelized (with 31 channels)	E	FRI-T1 or FRI-E1	A
FRM2 Frame Relay Module, Port Concentrator Shelf	F	FRI2-X.21	A

FRM Features and Functions

The FRM supports a maximum port speed of 2.048 Mbps plus the following features:

• Bundled connections
  
  
• Frame forwarding
  
  
• GMT request/response
  
  
• Explicit Congestion Notification (ECN)
  
  
• Multicast services
  
  
• ForeSight dynamic congestion avoidance
  
  
• AIP applications
  
  
• Enhanced V.35 loopback test
  
  
• T1 and E1 frame relay port interfaces
  
  
• Network-to-Network Interface (NNI) ports for V.35 and X.21 frame relay connections that extend across a Cisco WAN switching network to a foreign network
  
  
• Network-to-Network Interface (NNI) ports for T1 and E1 frame relay connections
  
  

The FRM can support a maximum of 252 virtual circuits (PVCs). PVC distribution can cross all four ports if they do not exceed the 252 PVC limit and the limit of 2.048 Mbps per FRM.

Bundled and grouped connections are software groupings of multiple virtual circuits within a single routing connection. Grouping allows a node to support up to 1024 virtual circuits, which is the logical equivalent of four FRMs.

Frame Relay Card Redundancy

Frame relay card redundancy can be provided through a second card set in adjacent slots and a Y-cable between each port that connects to the user-equipment. See Figure 4-26 for an illustration. The hardware kits for this feature contain a second frame relay card set, a set of Y-cables to interconnect the two card sets, and any other pieces that apply to the card types. Y-cable redundancy is not possible using back cards with different interfaces, such as an FRI T1 and FRI V.35.

Figure 4-26: Frame Relay Port Redundancy

Frame Relay Interface (FRI) V.35 Card

The Frame Relay Interface V.35 (FRI-V.35) is a four-port back card to the FRM card. Two models of the FRI-V.35 can support the FRM:

• FRI-V.35 Model A—supports composite data rates up to 1.024 Mbps and provides a CCITT V.35 interface for each port.
  
  
• FRI-V.35 Model B—supports composite data rates to 2.048 Mbps with V.35 interface.
  
  

The FRI-V.35 has the following functions and features:

• Interfaces 1 to 4 frame relay ports via 34-pin MRAC connector (Winchester, female)
  
  
• RTS, CTS, DSR, DTR, DCD, LLB, RLB, and TM control leads supported
  
  
• Handles up to 252 virtual circuits per card
  
  
• Hardware jumpers to configure the interface as DCE or DTE
  
  
• Supports normal and looped clocking
  
  

Figure 4-27: Frame Relay V.35 Connectors and Indicators

Table 4-28 shows the relationship between the number of ports used on the FRI and maximum operating speed for each port. Model A FRM and FRI cards are included for early users who may not have updated the cards. Note that the port numbers start at the top on the FRI faceplate.

Table 4-28: Port Speed Combinations

A/A	£ 256	£ 256	£ 256	£ 256
	1024	0	0	0
B/A	£ 256	£ 256	£ 256	£ 256
	£ 336	£ 336	£ 336	0
	£ 512	£ 512	0	0
	1024	0	0	0
C/A	£ 256	£ 256	£ 256	£ 256
	£ 336	£ 336	£ 336	0
	£ 512	£ 512	0	0
	1024	0	0	0
C/B D/B	One port can operate at 2048 or 1920. Combinations of ports can add up to 2048 Kbps. If one port is 1920 or 2048, no other port can be active.

Frame Relay V.35 Port Numbering

Each frame relay logical channel has a number in the form:

  SLOT.PORT.DLCI

where

• SLOT is the number of the front and back slots where the FRM and FRI-V.35 reside.
  
  
• PORT is the back card port number in the range 1 to 4.
  
  
• DLCI is the identification number for the PVC. The usable range of DLCIs is 16 to 1007. The reserved DLCIs are 0 through 15 and 1008 through 1023.
  
  

FRI-V.35 Data Clocking

The two clocking modes that the FRI supports are normal and looped. Figure 4-28 illustrates the two modes. Note that the direction for the clock and data is reversed for the two FRI mode configurations (DCE or DTE), as follows:

• If the FRI is DCE with normal clocking, it provides both transmit and receive clock to the user device.
  
  
• If the FRI is DTE with normal timing, the user device provides both the transmit and the receive clock.
  
  
• If the FRI is a DCE with looped timing, the user device provides the transmit clock on the EXT XMT CLK line, and the FRI provides the receive clock to the user device.
  
  
• If the FRI is DTE with looped timing, it provides the transmit clock on the EXT XMT CLK line, and the user device provides the receive clock.
  
  

Figure 4-28: Frame Relay Data Clocking Modes

---

Note Clock is looped by the FRI instead of the user device.

---

Figure 4-29 illustrates the FRI-T1 and FRI-E1 back cards.

Figure 4-29: Frame Relay T1/E1 Back Cards

Loopbacks

The IGX does not support port loopbacks (tstport and addextlp command) towards the IGX. In contrast, for V.35 and X.21 interfaces only, you can set up connection loopbacks towards the facility by using the addloclp and addrmtlp commands.

FRM-FRI Compatibility

The firmware on the front card must match the type of interface on the back card. Rev D firmware on the FRM supports X.21 and V.35 protocols. Rev E firmware on the FRM supports T1 and E1 protocols. The Display Card (dspcd) command indicates the type of back card that the FRM firmware supports and reports any mismatch.

Frame Relay Interface for X.21

The FRI-X.21 back card provides an X.21 interface to the user equipment. The two model D FRI cards are the FRI-V.35 and FRI-X.21. They differ only in the physical connectors (see Table 4-29). The operating rates of each port and the composite data rate supported by the FRI-X.21 card is the same as the FRI-V.35. You can configure each port as either a DCE or a DTE. Another FRI card is the FRI-2-X.21. This is the back card that provides the interface between the Port Concentrator Shelf (PCS) and the FRM-2. See the section titled "FRM-2 Interface to the Port Concentrator Shelf ."

The FRI-X.21 uses leased line service for international networks. The V.35 version is for domestic (U.S.) use and also uses a leased line service for its connections. The FRI-X.21 back card features:

• Four frame relay data ports with CCITT X.21 interface through DB15 connectors.
  
  
• Support for all standard X.21 data rates up to 2.048 Mbps.
  
  
• Support for C (control) and I (indication) control leads.
  
  
• Daughter board used to configure FRI as DCE or DTE.
  
  
• Card redundancy option provided by Y-cable and standby card pair.
  
  

FRI configuration supports one to four ports. The configuration depends on the maximum speed requirement (the card itself has a maximum composite speed). Figure 4-30 shows the FRI faceplate. Table 4-30 lists the available port operating speeds.

Any one port can operate at 2048 Kbps. Any combination of ports can equal 2048 Kbps. If a port is operating at 2048 Kbps, it must be port 1, and no other port can be active. Numbering of the 4 DB15 connectors starts at the top of the faceplate. Table 4-31 lists the cable and pinouts for an X.21 port.

Table 4-29: FRI Card Types

FRI-X.21	X.21	4-ports	DB15 Sub miniature, female
FRI-V.35	V.35	4-ports	34-pin MRAC type (Winchester), female

Table 4-30: Frame Relay Port Speeds Available

Any multiple of 56 Kbps	Up to 1.792 Mbps
Any multiple of 64 Kbps	Up to 2.048 Mbps

Table 4-31: FRI-X.21 Port Pin Assignments (DB 15-connector)

1	Gnd.	both	Protective (Shield) ground
8	Sig. Gnd.	both	Signal ground
3/10	C	both	Control
5/12	I	both	Indication
2/9	TxD	DTE	Transmit Data from DTE
4/11	RxD	DCE	Receive Data to DTE
6/13	S	DCE	Clock Out (DCE), Clock In (DTE)

Figure 4-30: Frame Relay X.21 Connectors and Indicators

X.21 Data Clocking

The FRI-X.21 supports only normal clock mode. The direction of the clock and data lines is reversed if the FRI is configured as a DCE or as a DTE as follows (see Figure 4-31):

• If the FRI is configured as a DCE, it provides a clock signal to the user device (DTE) on the S (clock) lead. This clock is synchronous with the node timing.
  
  
• If the FRI is configured as a DTE, the FRI-X.21 receives clock from the user device (DCE) on the S (clock) lead.
  
  

Figure 4-31: Frame Relay Data Clocking Modes

Y-Cable Redundancy and Port Modes

The Y-cable redundancy kits for the FRI-X.21 and FRI-V.35 contain four extra daughter cards for specifying individual ports as either DCE or DTE. The extra daughter cards are 200-Ohm versions for the FRI already installed. The higher impedance cards are necessary because of circuit behavior at higher speeds when the two interfaces are in parallel (by way of the Y-cable).

Card Self Test

As with all IGX cards, the FRI-X.21 includes internal diagnostic routines that periodically test the card's performance. These self-test diagnostics automatically start and run in background. They do not disrupt normal traffic. If a failure is detected during the self test, the faceplate red FAIL LED is turned on. The operator can also view the status at the control terminal by executing the Display Card (dspcd) command.

A report of a card failure remains until cleared. A card failure is cleared by the Reset Card (resetcd) command. The two types of reset that resetcd can do are hardware and failure. The failure reset clears the event log of any failure detected by the card self-test but does not disrupt operation of the card. A reset of the card firmware is done by specifying a hardware reset. This reboots the firmware and momentarily disables the card. If a redundant card is available, the hardware reset causes a switch over to the standby card.

Port Testing (X.21)

The X.21 frame relay ports and any associated external modems, CSUs, or NTUs can be tested using data loopback points in the circuit path. The three possible loopbacks for X.21 frame relay ports are:

• An internal loopback of the port
  
  
• A loopback of the near end (local) modem
  
  
• A loopback of the other end (remote) modem
  
  

The modems must be compatible with the Cisco loopback protocol. For information on supported modems and protocols, refer to the appendix titled "Peripherals Specifications" in the Cisco IGX 8400 Series Installation publication. Also, refer to the Cisco WAN Switching Command Reference for protocol requirements for the addextlp, addloclp, and addrmtlp commands. For information that applies to loopbacks on the FRM-2/FRI-2 and PCS, see the descriptions of these loopback commands in the Cisco WAN Switching Command Reference. If these sources fail to clarify a particular situation, contact Cisco Customer Service.

All three loopbacks are set up using the tstport command. Only one port at a time can be in loopback mode for testing.

The internal loopback point is established inside the FRI card. Figure 4-32 illustrates the setup. The FRM generates a test pattern, sends it out on the transmit circuitry, and detects this pattern on the receive circuitry. This test takes several seconds and momentarily interrupts traffic on the port. The test runs on a port that is in either DCE or DTE mode.

Figure 4-32: Frame Relay Loopback Modes

For ports configured for DTE, the two additional tests (local loopback and remote loopback) are available. For these ports there are two methods of loopback testing:

• A loopback data pattern (Test Mode) is transmitted to initiate a loopback. The modems or NTUs may or may not recognize these codes to perform the loopback. The FRI does not care and waits a programmable period (default = 10 seconds) of time to send the test pattern. After the test is completed, transmission of the codes is terminated and the circuit returns to normal operation. A Model D FRM is required for this testing. The test result is displayed on the IGX control terminal tstport screen.
  
  
• Some external equipment supports loopback testing but does not recognize the test pattern (Test Mode) in the data stream. In these cases, the FRM/FRI toggles the V.35 LLB (local loopback) and the RLB (remote loopback) leads then runs the test pattern. The FRM/FRI still waits the user-programmable time period (default = 10 secs) before running the data test pattern. A Model D FRM is required for this testing. The test result is displayed on the IGX control terminal tstport screen.
  
  

Unit Replacement

FRI card replacement is the same as other back card replacement. For back card replacement procedures, refer to the Cisco IGX 8400 Series Installation publication.

FRM-2 Interface to the Port Concentrator Shelf

This section introduces the Port Concentrator Shelf (PCS) for frame relay traffic. The PCS is an external device that expands the capacity of a Frame Relay Module (FRM) to 44 low-speed ports. This ability to increase the port density of an IGX switch supports more efficient usage for the common switch equipment. The port parameters are as follows:

• The per-port range is 9.6 Kbps through 384 Kbps.
  
  
• The typical configured rate is 64 Kbps per port or less.
  
  
• The maximum total throughput for the 44 ports is 1792 Kbps.
  
  

The PCS requires a version of the FRM/FRI card set that is exclusively dedicated to the PCS. The front card is the FRM-2. The back card that interfaces the FRM-2 to the PCS is the FRI-2-X.21. Figure 4-33 illustrates the relation ship between the FRM/FRI card set and the PCS. It provides one or more X.21 links. Each X.21 link is called a concentrated link. In a full configuration, each concentrated link services one of four 11-port modules in the PCS. This makes a total of 44 ports on the FRM-2.

---

Note FRM-2 front cards exist in two forms. One uses an ACM1 adaptor. The other is a single-card or "native" version. Functionally they are identical, and the firmware is the same.

---

For more detailed information on the PCS, refer to the section on the Port Concentrator Shelf in the Cisco WAN Switching System Overview. For detailed installation instructions, refer to the Port Concentrator Installation document that comes with each unit. Cabling information for the PCS appears in the cabling appendix and in the Port Concentrator Installation document.

Figure 4-33: Port Concentrator Shelf Components

Terminology

The following terms are used to identify PCS components:

• "FRC" (Frame Relay Concentrator) ports—FRM-2-X.21 ports connected to the PCS, used in commands such as dspfrcport. These are also known as PCS "physical" ports.
  
  
• Frame relay logical ports—each of the 44 PCS ports, from the perspective of the IGX interface, is a logical port. This distinguishes it from a physical frame relay port.
  
  

Operation and User Interface

Other than front panel LEDs, the PCS has no user interface because the PCS functions as an extension of the FRM-2. PCS ports are operated and maintained from the IGX user interface.

The PCS operates within the IGX environment as a frame relay card with 44 ports. Existing frame relay commands (cnffrport, upfrport, dnfrport, addcon, delcon, etc.) are the same in syntax and function. The difference is that a range of 44 ports can be specified instead of 4. The configuration of each of the PCS logical ports is similar to that of non-PCS frame relay ports. A frame relay card connected to a PCS notifies the system database and permits the additional ports to be specified.

PCS-to-IGX Interface

Compatibility

The PCS requires the following:

• System software version 8.1 or later
  
  
• FRM-2 front card and FRI-2-X.21 back card
  
  

---

Note Use of the PCS with the FRM-2 is exclusive: the PCS does not work with any other model of the FRM, and only a PCS works with the FRM-2. For example, if fewer than four PCS modules are used, unused frame relay ports cannot connect to something other than the PCS.

---

Concentrated Link

The concentrated link refers to the connection between port concentrator and frame relay card. Each module supporting 11 external ports is connected to 1 of the 4 ports on the FRI-2-X.21 back card.

Configuration

Each of the four composite links between PCS and IGX has a fixed configuration of:

• Speed: 512 Kbps
  
  
• FRM-2 physical port: FRI-2-X.21 DCE
  
  
• PCS composite port: X.24/V.11 DTE
  
  

The PCS Concentrated Link cable is illustrated in the cabling appendix. Its maximum length is 25 feet. You cannot use a modem to extend this distance.

Activation

FRM-2 firmware interacts with the PCS over composite links only if the FRM-2 is active. The FRM-2 changes from standby to active state when its first logical port is activated.

upfrport Command

A PCS logical port associated with an FRM-2 card is activated with the upfrport command. The upfrport command requires the slot number of the FRM-2 to which the PCS is connected. Enter the slot number and a logical port in the range of 1-44 (assuming a connection exists for all 4 composites between the PCS and IGX).

  Example: upfrport 4.1

This example indicates that the FRM-2 in slot 4 and concentrated link 1 are connected.

Entering the upfrport command for one port activates all four ports. The following events are generated by successful activation of four concentrated links. The display is from the example "upfrport 4.1:"

  "Info FRM 4 Activated"
"Info FRM Port 4.1 Activated
"Info FRM Concentrated Link 4.1 Failure Cleared"
"Info FRM Concentrated Link 4.2 Failure Cleared"
"Info FRM Concentrated Link 4.3 Failure Cleared"
"Info FRM Concentrated Link 4.4 Failure Cleared"

A noticeable delay occurs after upfrport begins executing on the first port. During initial upfrport execution, the FRM-2 performs first-time configuration, diagnostic, and up/download functions.

If a concentrated link is not connected or fails to come up, the logical port remains in a failed state until either the link comes up or the port is deactivated with the dnfrport command.

De-activation: the dnfrport Command

The FRM-2 card set returns to the standby state after you de-activate all 44 logical ports by executing the dnfrport command.

PCS Port Configuration

When the frame relay ports are activated, the IGX recognizes them as PCS-connected ports. Subsequently, all applicable frame relay port management commands accept logical port numbers in the form slot.port. The range for port is 1 to 44. Table 4-32 shows the logical ports the PCS supports. In Table 4-32 , slot is the IGX slot in which the FRM-2 resides.

Table 4-32: PCS Logical Ports

	PCS Ports
Composite L1	slot.1—slot.11
Composite L2	slot.12—slot.22
Composite L3	slot.23—slot.33
Composite L4	slot.34—slot.44

Interface Hardware Configuration

The interface and clocking characteristics for each PCS port is independently configured to be V.11 (X.21), V.35, or V.28 by inserting the required interface card (or "ICARD") into the associated slot in the PCS. For detailed information on PCS hardware interfaces, refer to the Port Concentrator Installation document. This document comes in the PCS shipping container.

The IGX does not have the capacity to read the type of interface present for the PCS port. The values you enter under Interface Type with the cnffrport command appear in the display only and cannot be checked against the hardware.

The cnffrport Command

You configure a PCS with the cnffrport command. With the following limitations, all parameters for the cnffrport command are supported:

1. For each group of 11 logical ports, the total speed must be 384 Kbps or less. The remaining 64 Kbps of composite link speed is reserved for control information. This total consists of only active ports.

2. The default port speed is 38.4 Kbps (instead of 256 Kbps for non-PCS ports).

3. Each PCS logical port supports speeds of 9.6, 14.4, 16, 19.2, 32, 38.4, or 48 Kbps. Higher speeds (56 to 384 Kbps) are valid as long as your configuration stays within limitation 1, above.

4. Only the active Interface Control Template is supported by the PCS.

Port Statistics

All frame relay summary and interval statistics are kept for PCS ports. The PCS and FRM-2 share responsibility for statistics collection on PCS ports. The PCS maintain counters for:

• CRC errors on received frames
  
  
• Aborted/underrun received frames
  
  
• Transmit Frames Discards
  
  
• Transmit Bytes Discards
  
  

All remaining statistics counters are collected by the FRM-2. Although the IGX supports ForeSight for PCS connections, CLLM (ForeSight) statistics are not available in Release 8.1. These fields are present but not valid.

PCS Monitoring Functions

Monitoring functions generally apply to the PCS except that you can specify up to 44 logical ports for a FRM-2 slot. For descriptions of the monitoring commands, refer to the "Troubleshooting" chapter of the Cisco WAN Switching Command Reference. Note that commands dspchcnf, dspchstats, dspportstats, and dspbob fail when the required concentrated link is down. Trying to execute one of these commands on a concentrated link that is down causes an error message to appear.

Collecting the Monitoring Information

Logical Port Speed

The PCS measures the speed of receive data on logical ports if the port is configured as a DTE interface. To see the measured speed, use the dspbob command. The PCS measures port speed after any of the following occurs:

• When the PCS is first powered up.
  
  
• When a port is configured with the cnffrport command. After the port is reconfigured, speed of incoming data is measured.
  
  
• When signals are displayed for the port with the dspbob command, you are given the option of measuring port speed at that time.
  
  
• When the PCS is reset or power is interrupted for any reason.
  
  

The process of measuring port speed sends out two 1-byte frames with no CRC on the port.

Physical Port Speed

The IGX measures the physical port speed for FRI-2-X.21 ports once per minute. The current measured speed is displayed with the dspfrcport command and should always read 512 Kbps when the port is active.

PCS Front Panel LEDs

Table 4-33: PCS Module Front Panel LEDs

LED Name	Function
State	OFF = No power. RED = power applied but software not running. GREEN = Operating software running.
Rx	Flashes for frames received on concentrated link. With no activity on link, Link Rx blinks once per second. With steady activity on link, Link Rx is on continuously.
Tx	Flashes for frames transmitted on concentrated link With no activity on link, Link Tx blinks once per second. With steady activity on link, Link Tx is on continuously.
CPU A	On solid when operating software on CPU A is running. Should always be On during operation.
CPU B	On solid when operating software on CPU B is running. Should always be On during operation.
Port 1	On for module connected to FRM-2 Port 1. Should always be On for Module 1.
Port 2	On for module connected to FRM-2 Port 2. Should always be On for Module 2.
Port 3	On for module connected to FRM-2 Port 3. Should always be On for Module 3.
Port 4	On for module connected to FRM-2 Port 4. Should always be On for Module 4.
Download	Flashes when frames are received during software download from FRM-2.

Card Insertion and Removal

If the FRM-2 or FRI-2.X21 card is removed for any reason, be sure to maintain card compatibility upon card replacement: the FRM-2 card is compatible with only the FRI-2-X.21 back card. The IGX declares a mismatch state for any other back card inserted into an active FRM-2 slot. Inserting compatible hardware is the only way to clear the mismatch. Similarly, once an IGX slot is active with an FRM-2, a mismatch is declared if any other front card is inserted into this slot. Before the slot can be used for any other type of card, the slot must be de-activated as a PCS-capable frame relay card.

PCS Command Summary

The commands in the list that follows apply to PCS frame relay ports. Most commands have the syntax described in the Cisco WAN Switching Command Reference with the exception that system software recognizes 44 ports per FRM-2 instead of 4. Some commands are PCS-specific.

addcon	used to add frame relay connections.
addloclp	test command for creating local loopback.
addrmtlp	test command for creating remote loopback.
cnfcondsc	configures frame relay connection description.
cnfchpri	configures frame relay channel priority.
cnfchutl	optimizes frame relay channel utilization.
cnffrcls	configures frame relay class of service (also cnfcos command for connection)
cnffrcon	configures frame relay connection parameters.
cnffrcport	configures the concentrated link between the FRM-2 cards and the PCS.
cnffrport	configures frame relay port parameters.
cnfpref	configures preferred route for frame relay connection.
clrchstats	clears PCS frame relay channel statistics.
clrfrcportstats	PCS-specific command for clearing FRI-2-X.21 physical port statistics.
clrportstats	clears PCS frame relay port statistics.
delcon	deletes a frame relay connection.
delcongrp	deletes a frame relay connection group.
dellp	deletes a frame relay loopback condition.
dncon	deactivates a frame relay connection.
dspchcnf	displays channel configuration.
dspchstats	displays PCS frame relay channel statistics.
dspfrcportstats	PCS-specific command to display FRI-2-X.21 physical port statistics.
dspfrcbob	PCS-specific command for displaying FRI-2-X.21 physical port signals.
dspfrport	displays PCS logical port.
dspfrcport	displays FRI-2-X.21 physical port.
dsppcs	PCS-specific command to display PCS unit information.
dspportstats	displays PCS frame relay port statistics.
dspcon	displays frame relay connections.
prtcons	prints frame relay connections.
prtchcnf	prints frame relay channel configuration.
tstcon	tests frame relay connection.
tstdelay	tests frame relay connection delay.
tstpc	PCS-specific command to test the PCS.
resetpc	PCS-specific command to reset the PCS.

PCS Port Failures

A PCS logical port failure is defined as a minor alarm. The FTC/FRP Port Comm Failure icon appears in the dspalms screen. Any connection that terminates on a failed port is also failed. Three causes of a port failure are defined, as described under Alarms and Events.

Conditioning

A failed connection on a PCS logical port is conditioned in the same manner as a failed connection on a non-PCS FRP port. Only the active control template is supported on PCS ports. The "conditioned" control template should not be used for PCS logical ports.

PCS General Operation

Firmware Download

When it detects a Port Concentrator on one of its links, the FRM-2 checks for a compatible firmware revision on the Port Concentrator. If the FRM8-2 detects that the firmware on the Port Concentrator is incompatible, the FRM8-2 downloads the current firmware to the Port Concentrator. This download operation takes about two minutes. An event is logged when a firmware download has either started or failed.

Operating software on the Port Concentrator is stored in Flash memory. Download should be required only if the PCS is connected to an FRM-2 with newer firmware or a PCS module is replaced and a software version difference exists.

Automatic Diagnostics—FRM-2 and FRI-2-X.21 Cards

The FRM-2 card runs a self-test diagnostic when it is in the standby state. The system software uses a reserved channel on the FRM-2 card to perform background loopback tests that include both the FRM-2 and FRI-2-X.21. This test verifies that all components up to the FRI-2-X.21 physical port are functioning. These diagnostics do not test the PCS.

StrataView Plus Interface

Information about PCS logical ports and frame relay connections automatically goes to StrataView Plus. StrataView Plus can also manage connections and other FRM-2 port functions.

SNMP Manager

The SNMP agent supports Port Concentrator logical ports. This includes configuring PCS port parameters, adding, or deleting frame relay connections, and retrieving statistics.

The SNMP agent also supports provisioning for 44 frame relay ports for FRM-2; the existing MIB variables are extended to the expanded number of logical ports. SNMP management functions are not supported for the Port Concentrator concentrated links.

User Interface

Interaction between the FRM-2 and PCS automatically updates the database to display the number of connected logical FRM-2 ports at the PCS. As a result, both the IGX user interface and the StrataView Plus interface automatically display the additional capacity of 44 ports for the FRM-2.

Concentrated Link Failure

If, during normal operation, communication between FRM-2 and PCS over a concentrated link stops, a concentrated link failure alarm results.

In addition, during start-up, a concentrated link is failed for any of the following reasons:

• Port Concentrator is not present.
  
  
• Code download from FRM-2 to Port Concentrator fails.
  
  
• Port Concentrator self-test fails.
  
  
• Cable is unplugged or bad.
  
  

Frame Trunk Module (FTM)

The Frame Trunk Module (FTM) supports devices that provide access for various types of traffic to the IGX node. Two lines of access devices are supported by the FTM. One series is the FastPAD family of products. The FastPADs support frame relay, voice, and serial data. The other series is the Cisco line of access devices.

The Cisco Line of Access Devices

An example of the Cisco line of access devices is the Cisco 3800 family of products. The Cisco access devices run the Cisco IOS (operating system) and have a control terminal separate from the node's control terminal. The access device itself uses IOS command, but once a control session has been established, you control the interface between the node and the access device by using Cisco WAN switching system software commands on the command line interface. For descriptions of the Cisco WAN switching system software commands that apply to the FTM and connections to the Cisco access devices, refer to the Cisco WAN Switching Command Reference. Individual manuals also exit for the Cisco access devices. Currently, you can refer to the Cisco 3800 Series Installation Guide and the Cisco 3800 Series Software Configuration Guide and Command Reference.

For the Cisco access devices, you can add connections between the following endpoints.:

• FTM and FTM or FTC
  
  
• FTM and CVM or CDP
  
  
• FTM and FRM or FRP
  
  

Note that, when you add connections between a 3800 on an FRM to a CDP or CVM, you must add the connections at the CDP or CVM. The FTM and FRP or FRM endpoints, you can add connections at either end. For more information on setting up connections between the Cisco 3800 access devices and an FTM, FRP/FRM, or CVM, refer to the Cisco IGX 8400 Series Installation publication.

The FastPAD Line of Access Devices

The interface cards for FastPADs are the FTM front card and FPC back card. The back card provides either a T1, E1, V.35, or X.21 interface. Each FPC V.35 or FPC X.21 provides four ports. Each port can support one FastPAD either locally or remotely (via modem). The T1 card has a DB15 for RX/TX and an alternate pair of RX/TX BNC connectors. The E1 connections are the same except for additional RX/TX-monitoring BNC connectors. Y-cable redundancy is also supported. Figure 4-34 shows the FTM front card and the FPC-V.35.

---

Note FTM front cards exist in two forms. One uses an ACM1 adaptor. The other is a single-card or "native" version. Functionally, they are identical and use the same firmware.

---

You manage the FTM/FPC, the FastPAD, and their ports and connections by entering commands via the Cisco WAN switching system software command line interface. Statistics that relate to cards, ports, and the FastPAD are collected for StrataView Plus. Card management of the FTM/FPC includes detection of card installation or removal, mismatched back cards, or Y-cable redundancy.

Figure 4-34: FastPAD Cards: FTM and FPC (V.35)

Port management includes EIA signalling, LMI alarms, upping and downing of ports and the collection of port statistics available to StrataView Plus.

FastPAD management permits the management of cards and ports on the FastPAD device from the IGX. This management includes card and card removal detection, card mismatch, uploads and downloads between the FastPAD and the IGX node.

Connection management involves mapping FastPAD connection to Frame Relay-type virtual circuits (VCs). Connections that originate at a FastPAD must terminate at another FastPAD. Each FTM/FPC card set supports up to 252 connections. The card set collects statistics on these connections and provides them to StrataView Plus.

For descriptions of the FastPAD commands and detailed information on the FastPAD, refer to the FastPAD User's Guide. Refer also to the StrataView FastPAD User's Guide.

Data Cards

A data circuit has a direct interface to the IGX through either a High-speed Data Module (HDM) or Low-speed Data Module (LDM) card set. The HDM set consists of an HDM front card and a Synchronous Data Interface (SDI) back card. The LDM set consists of an LDM front card and a Low-speed Data Interface (LDI) back card. The back cards match the circuit type to the front card. Synchronous data card sets are listed in Table 4-34. An IGX 8430 node can have up to 25 HDM/LDM sets in a non-redundant system, for support of up to 200 full-duplex data connections.

The synchronous data cards support the ability to configure and monitor EIA leads; the ability to configure each channel for clocking, data rate, and DTE or DCE interface type; and complete loopback testing capability. Data channels can support null modem emulation as well as constant-carrier and switched-carrier operation. Data interfaces are transparent with respect to protocol. Asynchronous, binary synchronous, and bit synchronous protocols are supported with no impact on host or terminal software.

Table 4-34: Synchronous Data Cards Available

HDM, High-speed Data Module	SDI, RS-449 synchronous data interface (for X.21 or RS-422).
	SDI, V.35 synchronous data interface.
	SDI, RS-232D synchronous data interface.
	SDI, RS-232C synchronous data interface (for V.24).
LDM, Low -speed Data Module	LDI, RS-232C (V.24) four-port and eight-port low-speed data interface, non-interleaved EIA. 4-port models limited to 56 Kbps., 8-port models up to 19.2 Kbps.
	LDI, RS-232D four-port and 8-port, low-speed data interface, up to 19.2 Kbps., non-interleaved EIA.

High-speed Data Module (HDM)

The HDM front card in an IGX is a programmable communications processor that can support one to four high speed, synchronous data channels. It operates at speeds from 1.2 Kbps up to 1344 Kbps on all four ports while performing link error monitoring.

The HDM front data card:

• Performs cell adaptation
  
  
• Supports normal, looped, and split clocking
  
  
• Provides isochronous clocking circuitry
  
  
• Packetizes and depacketizes EIA lead sampling information
  
  
• Provides loopback capabilities, testing, and diagnostics
  
  

An internal baud rate generator provides transmit and receive data clocks to the SDI card at the selected rate. The HDM can accept data from an external data device with a non network synchronized clock (isochronous clock) up to 112 Kbps. With isochronous clocking, the HDM sends a clock control signal to the other end of the circuit to synchronize the far end HDM receive clock to the isochronous clock received at the near end.

Unless specified, a packet of data for EIA control lead information is built only at a very low rate or when a change of state is detected on one or more of the control leads. The data rate is specified as either "fast" or "not fast" (the default) by the addcon command for data connections. A fast EIA lead transmission can be specified in the software to send EIA control lead information in every FastPacket (interleaved EIA mode). This tightly couples the EIA lead states with the transmitted data but reduces the bandwidth efficiency.

The HDM card is installed in a front slot. An SDI back card plugs directly into the P2 connector of the front card. The SDI back card provides the proper data channel interface.

The faceplate of the HDM has message lights and buttons for loopback control and signal monitoring. The buttons relate to loopback testing or scrolling through the FastPacket data ports for a snapshot of selected data port conditions (indicated by PORT, PORT UNDER TEST, loopback, and communication line state lights). Figure 4-35 illustrates and Table 4-35 lists the controls and indicators. When correlating the figure to the table, read from the top down.

Figure 4-35: HDM Controls and Indicators

Table 4-35: HDM Controls and Indicators

PORT light (yellow)	Indicates which data port on the back card is currently monitored.
SCROLL push-button	When pressed toggles through to the desired port. Information that is displayed by the remaining lights on the HDM faceplate applies to the port whose corresponding light is on.
LOOPBACK push-button	When pressed, toggles through the three loopback states—no loopback, local loopback, and remote loopback—on the port whose corresponding light is on. This button can be disabled by the cnffunc command.
PORT UNDER TEST light (yellow)	Indicates one of the ports has gone into loopback mode. If this is not the current port, press the SCROLL button to toggle in the port being tested.
LL light (yellow)	Indicates a local loopback is present on a port.
RL light (yellow)	Indicates a remote loopback is present on a port.
DTR light (green)	Indicates the Data Terminal Ready signal is ON at the selected port terminal.
TXD light (green)	Indicates that the Transmit Data signal is ON at the selected port terminal.
DCD light (green)	Indicates the Data Carrier Detect signal is ON at the selected port terminal.
RXD light (green)	Indicates that the Receive Data signal is ON at the selected port.
FAIL light (red)	Indicates an error occurred. Resetting the card with the resetcd f command is suggested first. If the LED comes on again, call Cisco Customer Engineering.
ACTIVE light (green)	Indicates that the card is active and functioning normally.

Redundancy for HDM data cards can be provided with a second front and back card set and a Y-cable connection on each port to the customer equipment. See Figure 4-36.

Figure 4-36: HDM Data Port Redundancy

Synchronous Data Interface Card (SDI)

The SDI card is a synchronous data interface back card that directly connects to a front HDM card. Each SDI card has four connectors and provides the physical and electrical connection interface to four data ports. Each port is independently configurable for DTE or DCE mode, baud rate, and so on. One for one port redundancy is provided with a second card set and a standard Y-cable arrangement.

The SDI card:

• Provides four ports for interfacing to the data equipment.
  
  
• Provides EIA control circuitry and samples EIA lead information.
  
  
• Performs serial to parallel conversion of data.
  
  
• Provides a jumper strap for configuring the ports as DTE or DCE.
  
  

Four types of SDI back cards can provide an interface between an HDM front card and the customer data equipment. Table 4-36 distinguishes each type of SDI card.

Table 4-36: SDI Physical Interfaces

RS-232C/V.24	4-ports	DB25 Subminiature, female
RS-232D/V.24	4-ports	DB25 Subminiature, female
RS-449/X.21	4-ports	DB37 Subminiature, female
V.35	4-ports	34-pin MRAC type (Winchester), female

Three clocking modes are available on the SDI for clocking in transmit data and clocking out receive data. In addition, the SDI can operate as either a DCE or DTE, which makes possible six combinations of clocking. (See Figure 4-37 and Figure 4-38 .) With loop clocking, the user device must loop the RxC to the XTC for clocking out the transmit data.

When the SDI is configured as DTE, the user device is the source of clock timing and is generally not synchronous with the network (IGX) timing. This is isochronous clocking. Isochronous clocking allows the customer data sets at each end of a circuit to operate at slightly different rates (non-synchronously) with minimum delay and loss of data. This feature limits the amount of data allowed to accumulate in the HDM receive buffers and forces a re-synchronization before the delay reaches an unacceptable level.

Figure 4-37: Clocking Modes for SDI in DCE Mode

Figure 4-38: Clocking Modes for SDI in DTE Mode

Isochronous clocking involves a mechanism that lets a node at the far end compensate for an unstable clock in the user device at the near end. Data transmission in an isochronous network is reliable up to 112 Kbps. You can use isochronous clocking on only one input at a time per port. The SDI does not support two isochronous clock inputs in the same direction as required by some modems that generate the TxC and RxC clocks independent of each other.

---

Note The IGX automatically uses isochronous mode when the DTE/DCE endpoints justify it.

---

Split clocking uses the user-device timing for timing data transmission in one direction and the IGX timing for the other direction.

Low Speed Data Module (LDM)

The LDM front card supports up to 8 synchronous or asynchronous data ports. Each port is independently configurable for DTE or DCE mode, baud rate, and so on. The LDM card is a low speed data module for use on RS-232C ports with data rates up to 19.2 Kbps, where the higher speed capabilities of an HDM are unnecessary.

The LDM can process either synchronous or non-synchronous input data. With non-synchronous inputs, the data is over-sampled at a rate determined by how much jitter your equipment can tolerate. Using an external device is also possible for synchronizing the asynchronous data before the data enters the IGX.

The LDM front data card:

• Performs cell adaptation of customer data and EIA control leads.
  
  
• Supports normal and looped clocking.
  
  
• Provides loopback capabilities, testing and diagnostics.
  
  

Additional features, such as embedded (fast) EIA, sixth EIA lead support, and pleisochronous clocking, are also supported. The fast EIA control lead lets the user include the RTS/CTS EIA control leads in the same FastPacket as customer data. The EIA control lead status is encoded as the eighth data bit in each data byte. This provides a quick EIA response without significantly affecting bandwidth requirements. It is limited to data rates of 19.2 Kbps and below.

The LDM can reside in any empty front slot and requires an LDI back card. The LDI card plugs directly into the P2 connector of the LDM card.

The faceplate of the LDM has message lights and buttons for loopback control and signal monitoring. Figure 4-39 shows and Table 4-37 lists these indicators and buttons. When correlating the figure to the table, read from the top down. The buttons are for loopback testing and scrolling through the FastPacket data ports to obtain a snapshot of selected port conditions (indicated by PORT, PORT UNDER TEST, loopback, and communication line status lights).

Figure 4-39: LDM Connections and Indicators

Table 4-37: LDM Connections and Indicators

PORT Readout Display	Indicates which FastPacket data port (1 through 8) on the Back card is currently under scrutiny.
SCROLL push-button	When pressed, toggles through the ports. Information displayed by other LEDs on the faceplate applies to the port shown by the 7-segment display.
LOOPBACK push-button	When pressed, toggles through the three loopback states on the port whose light is on. These states are: no loopback, local loopback, remote loopback.
PORT UNDER TEST light (yellow)	Indicates that a port has gone into the loopback mode. If this is not the current port, push the SCROLL button to toggle in the port being tested.
LL light (yellow)	Indicates that a local loopback is occurring on one of the ports.
RL light (yellow)	Indicates a remote loopback is occurring on one a port.
DTR light (green)	Indicates the Data Terminal Ready signal is ON at the selected port terminal.
TXD light (green)	Indicates the Transmit Data signal is ON at the selected port terminal.
DCD light (green)	Indicates the Data Carrier Detect signal is ON at the selected port terminal.
RXD light (green)	Indicates that the Receive Data signal is ON at the selected port.
FAIL light (red)	Indicates an error occurred. Resetting the card with the resetcd f command is suggested first. If the LED comes on again, call Cisco Customer Engineering.
ACTIVE light (green)	Indicates that the card is active and functioning normally.

Redundancy for LDM data card types is available through a second front and back card set and a Y-cable connection on each port to the customer data equipment. Figure 4-40 illustrates redundancy.

Figure 4-40: LDM Data Port Redundancy

The 4-port and 8-port LDM supports only a subset of the full RS-232C/D control leads. The LDM supports only non-isochronous DCE normal and DCE or DTE looped clocking modes, transmission of 3 EIA lead states (non-interleaved), and baud rates of up to 19.2 Kbps on the 8-port version and 38.4 Kbps on the 4-port version. Split clock mode is not supported.

Low Speed Data Interface Card (LDI)

The Low-Speed Data Interface (LDI) card is a low-speed data interface back card that operates in conjunction with an LDM front card. The LDI provides the physical and electrical connection interface between the user low-speed data circuit and the LDM data PAD. Three models of the LDI exist. Two are four-port cards, and one is an eight-port card, as Table 4-38 indicates.

Some of the functions and features of the LDI are:

• Four or eight ports for interfacing to the data equipment
  
  
• Sampling of EIA lead status for the LDM to monitor
  
  
• Serial-to-parallel conversion of user data
  
  
• Support for DTE or DCE operation
  
  

Table 4-38: LDI Physical Interfaces

LDI 4	RS-232C/D (V.24)	4-ports	DB15 Subminiature, female
LDI 8	RS-232C/D (V.24)	8-ports	DB15 Subminiature, female

The LDI can operate either as a DCE or DTE. Selection is made by using a Cisco DTE or DCE adapter cable between the port connector and the cable from the user device. This cable is terminated with a standard DB25 on the customer end. Each port is configured separately.

Three EIA control leads are brought out to the rear connectors, three when used as a DCE and three for DTE. These leads are:

Table 4-39: EIA Control Leads

RTS	CTS
DSR	DTR
DCD	RL

The RL lead allows the use of a previously unused DTE EIA output lead. You can use remote loopback (RL) to enable a far-end modem loopback. Local loopback (LL) is not provided as an output on the LDI. If local loopback is required, use the DTR lead by wiring it to the connector pin for local loopback.

The LDI supports two clocking modes: normal and looped (Figure 4-41). The normal mode is used when the LDI port is configured as a DCE. Looped clock is only used when the LDI port is configured as a DTE. The user device must take the external transmit clock and loop it back to the RxC for clocking in the receive-data. In both cases, the LDI is the source of clock timing. Table 4-40 shows the accuracy and worst case jitter that can be expected from an end-to-end circuit using LDIs at each end.

Table 4-40: LDI Clocking Accuracy and Jitter Performance

1200	.06	.03
2400	.12	.06
4800	.24	.12
9600	.48	.23
14,400	.72	.35
19,200	.96	.47

Figure 4-41: LDI Clocking Modes

Optional Peripherals

At least one node in a network has a StrataView Plus terminal, a control terminal, or a dial-in modem connected to it. Any control terminal connected in the network can configure, manage, monitor, and diagnose the entire network. In addition, at least one node in a network may have a connected printer for error and event reports.

The control terminal and printer connect to two RS-232 serial ports. These ports are the Control Terminal and Auxiliary Port on the SCM faceplate. These serial ports support all standard asynchronous data rates from 1200 to 19,200 baud. The default rate is 9600 baud. Data rates and the type of equipment connected to the ports are software-configurable.

Cisco  recommends that at least one IGX in the network be connected to a direct-dial modem so that TAC personnel can perform remote diagnostic tests. (Contact the TAC through Cisco Customer Engineering.) A direct-dial modem connects to the backplane at the Control Terminal port. In addition, an auto-dial modem can be connected to the AUX port connector on the SCM at any node in the network so that remote error messages and alarms can be sent to the TAC.

An external clock source can be connected to the SCM card using the external clock adapter cable. The external clock device can be either 1.544 MHz or 2.048 MHz RS-422 square wave signals depending on the primary application of the IGX (T1 or E1). Selection is made through software. The clock stability should be at least as good as a Stratum 3 clock source. The Cisco IGX 8400 Series Installation publication lists the terminals, printers, modems and clock sources that have been tested and approved for use with the IGX.

Posted: Wed Sep 11 00:33:15 PDT 2002
All contents are Copyright © 1992--2002 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.

---