hometocprevnextglossaryfeedbacksearchhelp
PDF

Table of Contents

Cisco IOS Release 11.3(5)WA4(8d) Release Notes for LightStream 1010 ATM Switch Software

Cisco IOS Release 11.3(5)WA4(8d) Release Notes for LightStream 1010 ATM Switch Software

March 11, 1999

LightStream 1010 ATM Cisco IOS Release 11.3(5)WA4(8d)

This document describes the features and caveats for Cisco IOS Release 11.3(5)WA4(8d) for the LightStream 1010 ATM switch software. This software is based on Cisco IOS Release 11.3(5).

Contents

This document includes the following sections:

Introduction

The LightStream 1010 ATM switch and Catalyst 8510 MSR provide switched ATM connections to individual workstations, servers, LAN segments, or other ATM switches and routers using fiber-optic, unshielded twisted-pair (UTP), and coaxial cable.

Note The ATM switch processor and port adapters can be installed in the Catalyst 5500 switch chassis.

Version and Part Number Matrix

Table 1 lists the release names, versions, and part numbers used with the LightStream 1010 ATM switch.


Table 1:
Release Name Release Version Part Number

WAS1-1

11.1.(4)

SF-WAS1-11.1.4

WAS1-2

11.1.(401)

SF-WAS1-11.1.401

WAS1-3

11.1.(6)

SF-WAS1-11.1.6

WAS1-4

11.1.(8)

SF-WAS1-11.1.8

WAS1-5

11.1.(9)

SF-WAS1-11.1.9

WAS1-6

11.1.(11)

SF-WAS1-11.1.11

WAS2-1

11.1.(401)

SF-WAS2-11.1.401

WAS2-2

11.1.(6)

SF-WAS2-11.1.6

WAS2-3

11.1.(8)

SF-WAS2-11.1.8

WAS2-4

11.1.(9)

SF-WAS2-11.1.9

WAS2-5

11.1.(11)

SF-WAS2-11.1.11

WAS3-1

11.2(2)WA3(1a)

SF-WAS3-11.2.2

WAS3-2

11.2(5)WA3(2b)

SF-WAS3-11.2.5

WAS3-3

11.2(8)WA3(3)

SF-WAS3-11.2.8

WAS3-4

11.2(10)WA3(4)

SF-WAS3-11.2.10

WAS3-5

11.2(12)WA3(5)

SF-WAS3-5

WAS3-6

11.2(15)WA3(6)

SF-WAS3-6

WAS3-7

11.2(15)WA3(7)

SF-WAS3-7

WAS4-1

11.2.0(8.0.1)FWA4(1)

SF-WAS4-11.2.8.1S

WAS4-2

11.3(0.8)TWA4(2)

SF-WAS4-2

WAS4-4

11.3(2a)WA4(4)

SF-WAS4-4

WAS4-6

11.3(3a)WA4(6)

SF-WAS4-6

WAS4-8

11.3(5)WA4(8b)

SF-WAS4-8

Release Name to Version and Part Number Matrix

System Requirements

This section describes the systems requirements for Release 11.3(5) and includes the following sections:

Memory Requirements

The standard default memory DRAM memory configuration is 64-MB DRAM (MEM-ASP64M) with release WA4.

Hardware Supported

Table 2 lists the interfaces supported on the LightStream 1010 ATM switch and Catalyst 8510 MSR and the minimum software release required to use them.


Table 2: Supported Interfaces and their Minimum Software Requirements
Part Number Description Minimum Software Requirement

WAI-OC3-4MM

4-port STS-3c/STM-1 multimode fiber port adapters

WAS1-1

WAI-OC3-4SS

4-port STS-3c/STM-1 single-mode fiber port adapters

WAS1-1

WAI-OC3-4U5

4-port STS-3c/STM-1 UTP-5 port adapters

WAS1-1

WAI-OC3-4SSLR

4-port STS-3c/STM-1 SMF long reach port adapters

WAS3-2

WAI-OC3-1S3M

OC-3 mix port adapter module, 1 IR+ port and 3 MM ports

WAS3-1

WAI-OC12-1MM

1-port STS-12c/STM-4c multimode fiber port adapters

WAS1-2

WAI-OC12-1SS

1-port STS-12c/STM-4c SMF port adapters

WAS1-1

WAI-OC12-1SSLR

1-port STS-12c/STM-4c SMF long reach port adapters

WAS3-2

WAI-ATM25-12P

12-port ATM 25 port adapters with 96-pin telco cable

WAS3-2

WAI-T3-2BNC

2-port DS-3 port adapters

WAS1-1

WAI-E3-2BNC

2-port E3 port adapters

WAS1-1

WAI-T3-4BNC

4-port DS-3 port adapters

WAS3-1

WAI-E3-4BNC

4-port E3 port adapters

WAS3-3

WAI-T1-4RJ48

4-port T1 (ATM) with RJ-48 interface port adapters

WAS3-1

WAI-E1-4RJ48

4-port E1 (ATM) with RJ-48 interface port adapters

WAS3-1

WAI-E1-4BNC

4-port E1 (ATM) with BNC interface port adapters

WAS3-1

WAI-T1C-4RJ48

4-port T1 (circuit emulation) with RJ-48 interface port adapters

WAS3-1

WAI-E1C-4RJ48

4-port E1 (circuit emulation) with RJ-48 interface port adapters

WAS3-1

WAI-E1C-4BNC

4-port E1 (circuit emulation) with BNC interface port adapters

WAS3-1

L1010-ASP-B-FC1

ASP1 with FC-per-class Queuing (FC-PCQ2)

WAS1-3

L1010-ASP-B-FCPFQ

ASP1 with FC-per-flow Queuing (FC-PFQ3)

WAS4-1

L1010-PWR-DC

Power Supply DC

WAS3-1

1ASP = ATM switch processor.
2FC-PCQ = feature card per-class queueing.
3FC-PFQ = feature card per-flow queueing.

Determining Your Software Release

To determine the version of Cisco IOS software currently running on the LightStream 1010 ATM switch and Catalyst 8510 MSR, log into the switch and use the show version EXEC command. The following is sample output from the show version command. The version number is indicated on the second line as shown below:

Cisco Internetwork Operating System Software
IOS (tm) LS1010 WA4-8 Software (LS1010-WP-M), Version 11.3 (5)WA4(8d)

Additional command output lines include more information, such as processor revision numbers, memory amounts, hardware IDs, and partition information.

Feature Set Tables

The Cisco IOS software is packaged in feature sets (also called software images) depending on the platform. Each feature set contains a specific set of Cisco IOS features. The following section lists the features set matrix and the features supported by each feature set.

Table 3 lists the Cisco IOS software feature sets available for the LightStream 1010 ATM switch and Catalyst 8510 MSR in Cisco IOS Release 11.3(5)WA4(8d) along with those features supported in previous releases.


Table 3: Feature Sets Supported by the Lightstream 1010 ATM Switch and Catalyst 8510 MSR
Feature Set 11.3(5)WA4(8d) 11.3(5)WA4(8c) 11.3(5)WA4(8b) 11.3(3a)WA4(6)

Left-justified E.164 AFI support

x

x

x

x

SNMP1

x

x

x

x

Asynchronous support

x

x

x

x

PPP2 (SLIP3/PPP)

x

x

x

x

IP4

x

x

x

x

NTP5

x

x

x

x

TACACS+6

x

x

x

x

Telnet

x

x

x

x

Point-to-point and point-to-multipoint permanent VCCs7 and VPCs8

x

x

x

x

Point-to-point and point-to-multipoint switched VCCs and VPCs (UNI 3.0)

x

x

x

x

Point-to-point and point-to-multipoint switched VCCs and VPCs (UNI 3.1)

x

x

x

x

Point-to-point and point-to-multipoint switched VCCs and VPCs (UNI 4.0)

x

x

x

x

Multipoint-to-point UNI signaling

x

x

x

x

Soft VCCs and VPCs

x

x

x

x

VP tunneling

x

x

x

x

VPI/VCI range support in ILMI 4.0

x

x

x

x

PNNI hierarchy

x

x

x

x

ILMI version 4.0

x

x

x

x

IISP9

x

x

x

x

LANE10 client (LEC11) and LANE Services (LES12/BUS13/LECS14) on ASP15

x

x

x

x

Token Ring LANE services

x

x

x

x

ATM ARP16 server on ASP

x

x

x

x

ATM ARP client on ASP

x

x

x

x

ATM tag switch router (TSR)

x

x

x

x

Port snooping

x

x

x

x

OAM17 F4 and F5

x

x

x

x

E.164 address translation

x

x

x

x

E.164 autoconversion

x

x

x

x

Circuit emulation

x

x

x

x

ATM access lists

x

x

x

x

ATM accounting

x

x

x

x

ATM RMON18

x

x

x

x

Multiple, weighted, dynamic thresholds for selective packet marking and discard

x

x

x

x

Shaped VP tunnels for CBR traffic (FC-PFQ only)

x

x

x

x

Substitution of other service categories in shaped VP tunnels (FC-PFQ feature cards only)

x

x

x

x

Dual leaky bucket policing (FC-PFQ feature cards only)

x

x

x

x

Scheduler/Service Class/PVC configuration for FC-PFQ feature cards

x

x

x

x

Logical multicast support (up to 254 leaves per output port, per point-to-multipoint VC) (FC-PFQ feature cards only)

x

x

x

x

Network clocking enhancements for smooth switchover (FC-PFQ feature cards only)

x

x

x

x

Per-VC or per-VP nondisruptive snooping (FC-PFQ feature cards only)

x

x

x

x

Support for non-zero MCR19 on ABR connections (FC-PFQ feature cards only)

x

x

x

x

Access lists on ILMI registration

x

x

x

x

CUGs

x

x

x

x

ATM soft restart

x

x

x

x

ATM accounting enhancements

x

x

x

x

CISCO-SYSLOG-MIB support

x

x

x

x

CISCO-CONFIG-COPY-MIB support

x

x

x

x

Signaling diagnostics and MIB

x

x

x

x

Supplemental AToM MIB

x

x

x

x

1SNMP = Simple Network Management Protocol.
2PPP = Point-to-Point Protocol.
3SLIP = Serial Line Internet Protocol.
4IP = Internet Protocol.
5NTP = Network Time Protocol.
6TACACS+ = Terminal Access Controller Access Control System.
7VCCs = virtual channel connections.
8VPCs = virtual path connections.
9IISP = Interim-Interswitch Signaling Protocol
10LANE = LAN Emulation.
11LEC = LAN Emulation Client.
12LES = LAN Emulation Server.
13BUS = broadcast and unknown server.
14LECS = LAN Emulation Configuration Server.
15ASP = ATM switch processor.
16ARP = Address Resolution Protocol.
17OAM = Operation, Administration, and Maintenance.
18RMON = Remote Monitoring
19MCR = minimum cell rate

New and Changed Information

This section contains information about new features that appear in this and previous releases of Cisco IOS Release 11.3.

New Features in Release 11.3(5)WA4(8d)

There are no new features being added for release 11.3(5)WA4(8d).

New Features in Release 11.3(5)WA4(8c)

There are no new features being added for release 11.3(5)WA4(8c).

New Features in Release 11.3(5)WA4(8b)

The sonet tx-ais on-rx-defect command was added to allow a SONET interface to send an alarm indication signal (AIS) if it detects that the receive port has failed.

New Cisco IOS Release 11.3(3a)WA4(6) Features

The following features have been added to the LightStream 1010 ATM switch software:

See the section "Documentation Updates" for a list of documents that describe these features.

Cisco IOS Release 11.3(2a)WA4(4) Features

The following features have been added to the LightStream 1010 ATM switch software:

See the "Documentation Updates" section for a list of documents that describe these features.

Cisco IOS Release 11.3(0.8)TWA4(2) Features

The following features have been added to the LightStream 1010 ATM switch software Cisco IOS Release 11.2(0.8)TWA4(2):

See the "Documentation Updates" section for a list of documents that describe these features.

Cisco IOS Release 11.2(8.0.1)FWA4(1) Features

The following features have been added to the LightStream 1010 ATM switch software 11.2(8.0.1)FWA4(1):

See the "Documentation Updates" section for a list of documents that describe these features.

Caveats

This section contains listings of the caveats and corrected caveats for each release. Use Table 4 to determine if a particular caveat applies to your software release, or if a particular caveat was corrected in your software release. These caveats are described in detail in the following sections:


Table 4: Cisco IOS Release Caveats and Caveats Corrected Reference
DDTS Number 11.3(5)WA4(8d) 11.3(5)WA4(8c) 11.3(5)WA4(8b) 11.3(3a)WA4(6) 11.3(2a)WA4(4) 11.3(0.8)TWA4(2) 11.2(8.0.1)FWA4(1)
Caveat Correct Caveat Caveat Correct Caveat Caveat Correct Caveat Caveat Correct Caveat Caveat Correct Caveat Caveat Correct Caveat Caveat Correct Caveat

CSCdi55937

x

x

x

x

x

x

x

CSCdi72017

x

CSCdi72255

x

x

x

x

CSCdi74229

x

x

x

x

x

x

x

CSCdi74298

x

CSCdi75584

x

x

x

x

x

x

x

CSCdi82954

x

x

x

x

x

x

x

CSCdi83275

x

x

x

x

x

x

x

CSCdi85540

x

CSCdi92142

x

x

x

x

x

x

x

CSCdi93291

x

CSCdj01016

x

x

x

x

x

x

x

CSCdj01757

x

x

x

x

x

x

x

CSCdj02004

x

CSCdj03281

x

CSCdj03621

x

CSCdj04898

x

CSCdj10889

x

x

x

x

x

x

x

CSCdj11070

x

x

x

x

x

x

x

CSCdj12108

x

x

CSCdj13326

x

CSCdj13565

x

x

x

x

x

x

x

CSCdj18430

x

x

x

x

x

x

x

CSCdj18583

x

x

x

x

x

x

x

CSCdj18678

x

x

x

x

x

x

x

CSCdj22451

x

CSCdj25769

x

x

CSCdj25772

x

x

x

x

x

x

x

CSCdj28496

x

x

x

x

CSCdj41665

x

CSCdj41709

x

CSCdj42070

x

x

CSCdj42967

x

x

x

x

x

x

x

CSCdj48158

x

CSCdj48371

x

CSCdj49255

x

CSCdj50627

x

x

CSCdj50993

x

CSCdj51042

x

CSCdj53797

x

x

CSCdj54954

x

x

x

x

x

x

x

CSCdj56808

x

x

CSCdj57479

x

x

CSCdj59621

x

x

x

CSCdj63058

x

CSCdj68412

x

x

x

x

CSCdj70489

x

x

CSCdj70619

x

x

x

x

CSCdj71109

x

x

x

x

x

x

x

CSCdj71876

x

x

x

x

x

x

x

CSCdj72651

x

x

x

x

CSCdj74422

x

x

CSCdj74442

x

x

x

CSCdj78305

x

x

x

x

x

x

x

CSCdj80396

x

x

x

x

CSCdj81680

x

x

CSCdj82930

x

x

x

x

x

x

CSCdj84344

x

x

x

x

x

x

CSCdj84379

x

x

x

x

CSCdj84656

x

x

CSCdj85627

x

x

x

x

CSCdj85853

x

x

x

x

x

x

x

CSCdj87463

x

x

CSCdj88091

x

x

x

CSCdj89223

x

x

x

CSCdj90806

x

x

CSCdj91890

x

x

CSCdj94322

x

x

CSCdk01629

x

x

CSCdk02119

x

x

CSCdk04816

x

x

CSCdk06610

x

x

x

x

x

x

CSCdk07378

x

x

x

x

x

x

CSCdk08318

x

x

CSCdk10398

x

x

x

x

CSCdk17127

x

x

CSCdk17133

x

x

CSCdk21557

x

x

CSCdk22484

x

x

CSCdk22791

x

x

CSCdk27716

x

x

CSCdk27905

x

x

CSCdk29039

x

x

CSCdk32548

x

x

CSCdk33601

x

x

x

x

CSCdk46071

x

x

CSCdk47194

x

x

CSCdk54019

x

x

CSCdk54678

x

x

CSCdk62547

x

x

x

x

CSCdk64465

x

x

x

x

CSCdk65901

x

x

CSCdk69639

x

x

CSCdk71494

x

x

CSCdk73583

x

x

CSCdk76280

x

x

CSCdk77032

x

x

CSCdk78469

x

x

CSCdk82708

x

x

CSCdk84355

x

x

CSCdm07874

x

x

Cisco IOS Release 11.3(5)WA4(8d) Caveats Corrected

Symptom: When using a 25-Mbps port adapter, the number of leaves in a point-to-multipoint connection cannot exceed 32.

Cisco IOS Release 11.3(5)WA4(8d) Caveats

Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX LED lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Version 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), are configured on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it is reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the Interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In a LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: If there are multiple parallel paths to the same destination on a LightStream 1010 switch with a FC-PFQ feature card installed, the tag switching VCs (TVCs) should be load balanced on a per network prefix basis over these parallel paths (up to a maximum of 4 parallel paths) instead of being VC merged. Load balancing does not happen in some cases and the TVCs might be VC merge and go out of the switch as a single VC.
Workaround: None.
Symptom: TCP: %SYS-3-MGDTIMER:Uninitialized timer : mgd_timer_complain_uninit.
Workaround: None.
Symptom: The ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might get stuck in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: Shutting down an interface with more than 1000 tag switching VCs might cause a CPUHOG occurrence for the EXEC process.
Workaround: There is no workaround for this problem. The CPUHOG occurrence does not impact the functionality of the ATM switch.
Symptom: IP host-routing does not disable when specified.
Workaround: Enable and disable ip routing and then save the configuration to NVRAM.
Symptom: CPU hog when setting up numerous calls.
Symptom: Open shortest path first (OSPF) does not recognize more than four parallel interfaces. This might cause some tag VCs (TVCs) to not get switched to other interfaces if a tunnel carrying the TVCs is shut down.
Workaround: Enter a clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches to switch to other interfaces. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make a conversion of a requested rate to the actual rate.
Symptom: When you set the ROM monitor environment variable boot to a nonexistent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: A LightStream 1010 switch installed in the Catalyst 5500 chassis might fail to boot up when using mixed power supplies at high temperatures.
When using power supplies from different vendors on the Catalyst 5500 chassis, make sure that the supplies are powered-up in serial order. If you power up the power supplies simultaneously, the ASP on the LightStream 1010 ATM switch might not come out of reset. For example, the ASP might hang and not boot.
Symptom: Funnel VCs can jeopardize quality of service (QoS) for services.
The current multipoint-to-point funnel implementation can compromise the QOS guarantees of other connections (guaranteed services) when the application that created the funnel SVC malfunctions. For example, if the application were to transmit traffic on more than one leg of the funnel SVC simultaneously, the rate scheduler on the output interface will over-subscribe and, potentially, affect the peak cell rate (PCR), sustained cell rate (SCR), and maximum cell rate (MCR) guarantees for other VCs on the interface.
Symptom: ILMI does not come up upon reload of vp-mux (vuni).
ILMI on the tunnel interface (connected to the far-end peer device through a vp-mux switch and virtual UNI) has trouble staying in "WaitDevType" after reload of the vp-mux switch.
Workaround: This situation can be cleared by shutting down the tunnel interface; disable the ILMI followed by the no shut command on the tunnel interface. After the interface settles down, the ILMI on the tunnel can be reenabled. Shut the interface. Use the commands atm ilmi-enable and atm auto-configuration (if desired), then reenable the tunnel interface with the no shut command.
Symptom: ADD Party over Tunnel does not work.
On a source node, if you are adding a subsequent leaf to a VP tunnel that is out of bandwidth, PNNI might reject the route request (for the leaf) leading to an eventual rejection of the ADD party by signaling.
Workaround: None.
Symptom: The snmp get command does not work for the following MIB variables:
Symptom: Crash with soft release 11-3-3aWA4-6.
Data flowing over OC12 links between two LightStream 1010 ATM switches can cause potential memory corruption leading to system crashes. This crash may be caused due to badly formed AAL5 packets.
Workaround: Reboot the switch.

Cisco IOS Release 11.3(5)WA4(8c) Caveats Corrected

Symptom: When a LEC/LES/BUS are all on the same switch, a point-to-multipoint connection exits with the CPU as root and one leaf on the CPU (in addition to other leafs). When the LES sends a packet to all leafs, one packet needs to loop back to the CPU for the LEC. The code that performs loopback initializes hwidb->atm_db->fast_vc_info for the LANE client and eventually calls atm_input to queue the packet. However, another packet could come in on the same CPU port (for example, PNNI). This situation would cause hwidb->atm_db->fast_vc_info to be overwritten because hwidb is shared. When interrupt handling is complete for the new packet, the old packet can be misinterpreted as a totally different kind of packet.
Symptom: Disabling tag switching on an interface that carries tagged VCs (TVCs) does not remove the TVCs. Reenabling tag switching on that interface will put it in the "not TDP ready" state.
Symptom: A LightStream 1010 ATM switch with 8 LAN emulation clients configured on the CPU and 300 point-to-point and 60 point-to-multipoint connections crashes unexpectedly at the find-buddy function, due to an infinite loop.
Symptom: Connections where end-to-end delay IE is included in the connect message might fail. Examples of these connections include UNI40, CBR, UBR-rt.
A LightStream 1010 ATM switch that is an intermediate switch passes a bad CONNECT message from one side to another. The CONNECT message shows up as a CONNECT packet that is longer than the actual contents, with junk bytes at the end.
Symptom: The PNNI SNMP agent is not able to discover and allow PNNI-related configurations on ATM subinterfaces.
An alignment error message is printed when an SNMP walk (or other SNMP get next) request is issued.
Symptom: PNNI does not clean up and release some of its data structures properly, following process termination. A PNNI process is terminated when the associated node is either disabled or removed. As a result of not freeing the data structures, the associated memory is lost until the next reload.
Symptom: When running tag switching, no tag is created for the default route.
Symptom: Some ports on an ATM25 12-port adapter occasionally go into an alarm state. When this condition occurs, the port adapter must be hot-swapped.

Cisco IOS Release 11.3(5)WA4(8c) Caveats

Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX LED lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Version 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), are configured on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it is reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the Interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In a LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: If there are multiple parallel paths to the same destination on a LightStream 1010 switch with a FC-PFQ feature card installed, the tag switching VCs (TVCs) should be load balanced on a per network prefix basis over these parallel paths (up to a maximum of 4 parallel paths) instead of being VC merged. Load balancing does not happen in some cases and the TVCs might be VC merge and go out of the switch as a single VC.
Workaround: None.
Symptom: TCP: %SYS-3-MGDTIMER:Uninitialized timer : mgd_timer_complain_uninit.
Workaround: None.
Symptom: The ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might get stuck in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: Shutting down an interface with more than 1000 tag switching VCs might cause a CPUHOG occurrence for the EXEC process.
Workaround: There is no workaround for this problem. The CPUHOG occurrence does not impact the functionality of the ATM switch.
Symptom: IP host-routing does not disable when specified.
Workaround: Enable and disable ip routing and then save the configuration to NVRAM.
Symptom: CPU hog when setting up numerous calls.
Symptom: Open shortest path first (OSPF) does not recognize more than four parallel interfaces. This might cause some tag VCs (TVCs) to not get switched to other interfaces if a tunnel carrying the TVCs is shut down.
Workaround: Enter a clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches to switch to other interfaces. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make a conversion of a requested rate to the actual rate.
Symptom: When you set the ROM monitor environment variable boot to a nonexistent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: A LightStream 1010 switch installed in the Catalyst 5500 chassis might fail to boot up when using mixed power supplies at high temperatures.
When using power supplies from different vendors on the Catalyst 5500 chassis, make sure that the supplies are powered-up in serial order. If you power up the power supplies simultaneously, the ASP on the LightStream 1010 ATM switch might not come out of reset. For example, the ASP might hang and not boot.
Symptom: Funnel VCs can jeopardize quality of service (QoS) for services.
The current multipoint-to-point funnel implementation can compromise the QOS guarantees of other connections (guaranteed services) when the application that created the funnel SVC malfunctions. For example, if the application were to transmit traffic on more than one leg of the funnel SVC simultaneously, the rate scheduler on the output interface will over-subscribe and, potentially, affect the peak cell rate (PCR), sustained cell rate (SCR), and maximum cell rate (MCR) guarantees for other VCs on the interface.
Symptom: ILMI does not come up upon reload of vp-mux (vuni).
ILMI on the tunnel interface (connected to the far-end peer device through a vp-mux switch and virtual UNI) has trouble staying in "WaitDevType" after reload of the vp-mux switch.
Workaround: This situation can be cleared by shutting down the tunnel interface; disable the ILMI followed by the no shut command on the tunnel interface. After the interface settles down, the ILMI on the tunnel can be reenabled. Shut the interface. Use the commands atm ilmi-enable and atm auto-configuration (if desired), then reenable the tunnel interface with the no shut command.
Symptom: ADD Party over Tunnel does not work.
On a source node, if you are adding a subsequent leaf to a VP tunnel that is out of bandwidth, PNNI might reject the route request (for the leaf) leading to an eventual rejection of the ADD party by signaling.
Workaround: None.
Symptom: The snmp get command does not work for the following MIB variables:
Symptom: Crash with soft release 11-3-3aWA4-6.
Data flowing over OC12 links between two LightStream 1010 ATM switches can cause potential memory corruption leading to system crashes. This crash may be caused due to badly formed AAL5 packets.
Workaround: Reboot the switch.
Symptom: When using a 25-Mbps port adapter, the number of leaves in a point-to-multipoint connection cannot exceed 32.
Workaround: None.

Cisco IOS Release 11.3(5)WA4(8b) Caveats Corrected

Symptom: ATM Accounting: Exception in validblock_diagnose because resources at destination switch were not available for soft-VC setup.
Symptom: When a large amount of data is sent out of the Ethernet interface, the interface can become overwhelmed and start generating the error message:
%SONICT-3-INTERNAL_ERROR: sonic_send: no free tbufs
The interface stops sending data and the transmitter remains stuck until a shut/no shut command sequence is issued on the interface.
Symptom: Per VC Snooping, unconfigurable or broken.
Symptom: Left-justified AESA feature can cause the system to crash when enabling and disabling a node with 3000 or more e164/f69/X121 addresses.
Symptom: PNNI: crash at compare_avl_nodes.
When using the pnni_election command with a freed pnni_pdb where there are two scripts running pnni commands from different threads, the CLI command creates the pnni node and passes the pnni_pdb in the csb to be processed by the next issued command. In the meantime SNMP deletes the node from under CLI.
Symptom: PNNI: MIBS, wrong link-type in pnniLinkTable.
The PNNI agent returned a value of lowestLevelOutsideLink(4) for links between nodes on the same peer group for the pnniLinkType object. The correct value is lowestLevelHorizontalLink(2).
Symptom: Crash in process_handle_watchdog
This problem is caused by a routing loop, which is caused by metric overflow in metric aggregation.
Symptom: Power-on diagnostic test indicates snake-test failures from the ASP-PFQ.
If the snake test fails or the switch is not reset, the red LED on the left side of the ASP remains lit. Changing out power supplies and ASPs does not correct this problem. Consequently, there is no evidence that this alarm indicates a failed ASP.
Symptom: CES interfaces do not come up upon reload.
The line state stays at XmtAIS LossOfSignal. Sometimes the Switch comes up, but the IF status of show ces int always shows as DOWN and the line state is RcvAIS XmtAIS.
Symptom: CES: per-vc queues fill with cells on 64Kbps circuits.
With standard CES 64Kbps circuits, the per-VC queues build up with cells at a rate of about one per second. There is a mismatch between the cell-rate requested and the rate granted. This is caused by signalling a pcr of 170 cells/second, rather than the desired 172 cells/second. Because the CBR FC-PFQ traffic shaping in current software has excellent granularity, roughly a cell every two seconds is queued. After the queue fills, dropping occurs at that rate.

Cisco IOS Release 11.3(5)WA4(8b) Caveats

Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX LED lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Version 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), are configured on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it is reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the Interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In a LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: If there are multiple parallel paths to the same destination on a LightStream 1010 switch with a FC-PFQ feature card installed, the tag switching VCs (TVCs) should be load balanced on a per network prefix basis over these parallel paths (up to a maximum of 4 parallel paths) instead of being VC merged. Load balancing does not happen in some cases and the TVCs might be VC merge and go out of the switch as a single VC.
Workaround: None.
Symptom: TCP: %SYS-3-MGDTIMER:Uninitialized timer : mgd_timer_complain_uninit.
Workaround: None.
Symptom: The ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might get stuck in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: Shutting down an interface with more than 1000 tag switching VCs might cause a CPUHOG occurrence for the EXEC process.
Workaround: There is no workaround for this problem. The CPUHOG occurrence does not impact the functionality of the ATM switch.
Symptom: IP host-routing does not disable when specified.
Workaround: Enable and disable ip routing and then save the configuration to NVRAM.
Symptom: CPU hog when setting up numerous calls.
Symptom: Open shortest path first (OSPF) does not recognize more than four parallel interfaces. This might cause some tag VCs (TVCs) to not get switched to other interfaces if a tunnel carrying the TVCs is shut down.
Workaround: Enter a clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches to switch to other interfaces. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make a conversion of a requested rate to the actual rate.
Symptom: When you set the ROM monitor environment variable boot to a nonexistent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: A LightStream 1010 switch installed in the Catalyst 5500 chassis might fail to boot up when using mixed power supplies at high temperatures.
When using power supplies from different vendors on the Catalyst 5500 chassis, make sure that the supplies are powered-up in serial order. If you power up the power supplies simultaneously, the ASP on the LightStream 1010 ATM switch might not come out of reset. For example, the ASP might hang and not boot.
Symptom: Funnel VCs can jeopardize quality of service (QoS) for services.
The current multipoint-to-point funnel implementation can compromise the QOS guarantees of other connections (guaranteed services) when the application that created the funnel SVC malfunctions. For example, if the application were to transmit traffic on more than one leg of the funnel SVC simultaneously, the rate scheduler on the output interface will over-subscribe and, potentially, affect the peak cell rate (PCR), sustained cell rate (SCR), and maximum cell rate (MCR) guarantees for other VCs on the interface.
Symptom: ILMI does not come up upon reload of vp-mux (vuni).
ILMI on the tunnel interface (connected to the far-end peer device through a vp-mux switch and virtual UNI) has trouble staying in "WaitDevType" after reload of the vp-mux switch.
Workaround: This situation can be cleared by shutting down the tunnel interface; disable the ILMI followed by the no shut command on the tunnel interface. After the interface settles down, the ILMI on the tunnel can be reenabled. Shut the interface. Use the commands atm ilmi-enable and atm auto-configuration (if desired), then reenable the tunnel interface with the no shut command.
Symptom: When a LEC/LES/BUS are all on the same switch, a point-to-multipoint connection exits with the CPU as root and one leaf on the CPU (in addition to other leafs). When the LES sends a packet to all leafs, one packet needs to loop back to the CPU for the LEC. The code that performs loopback initializes hwidb->atm_db->fast_vc_info for the LANE client and eventually calls atm_input to queue the packet. However, another packet could come in on the same CPU port (for example, PNNI). This situation would cause hwidb->atm_db->fast_vc_info to be overwritten because hwidb is shared. When interrupt handling is complete for the new packet, the old packet can be misinterpreted as a totally different kind of packet.
Symptom: ADD Party over Tunnel does not work.
On a source node, if you are adding a subsequent leaf to a VP tunnel that is out of bandwidth, PNNI might reject the route request (for the leaf) leading to an eventual rejection of the ADD party by signaling.
Workaround: None.
Symptom: The snmp get command does not work for the following MIB variables:
Symptom: Crash with soft release 11-3-3aWA4-6.
Data flowing over OC12 links between two LightStream 1010 ATM switches can cause potential memory corruption leading to system crashes. This crash may be caused due to badly formed AAL5 packets.
Workaround: Reboot the switch.
Symptom: Disabling tag switching on an interface that carries tagged VCs (TVCs) does not remove the TVCs. Reenabling tag switching on that interface will put it in the "not TDP ready" state.
Symptom: A LightStream 1010 ATM switch with 8 LAN emulation clients configured on the CPU and 300 point-to-point and 60 point-to-multipoint connections crashes unexpectedly at the find-buddy function, due to an infinite loop.
Symptom: Connections where end-to-end delay IE is included in the connect message might fail. Examples of these connections include UNI40, CBR, UBR-rt.
A LightStream 1010 ATM switch that is an intermediate switch passes a bad CONNECT message from one side to another. The CONNECT message shows up as a CONNECT packet that is longer than the actual contents, with junk bytes at the end.
Symptom: The PNNI SNMP agent is not able to discover and allow PNNI-related configurations on ATM subinterfaces.
An alignment error message is printed when an SNMP walk (or other SNMP get next) request is issued.
Symptom: PNNI does not clean up and release some of its data structures properly, following process termination. A PNNI process is terminated when the associated node is either disabled or removed. As a result of not freeing the data structures, the associated memory is lost until the next reload.
Symptom: When running tag switching, no tag is created for the default route.
Symptom: Some ports on an ATM25 12-port adapter occasionally go into an alarm state. When this condition occurs, the port adapter must be hot-swapped.

Cisco IOS Release 11.3(3a)WA4(6) Caveats Corrected

This section describes caveats from Cisco IOS Release 11.3(2) TWA4(4) that have been corrected by Cisco IOS Release 11.3(3a) WA4(6) of the LightStream 1010 ATM switch software.

Symptom: E3 port adapters are limited to 65535 cells per second per PVC regardless of the framing type. There is no benefit from G832 framing.
Symptom: The fail over time for the circuit emulation switch (CES) soft VC on a redundant link might be inconsistent, from less than 1 second up to 15 seconds.
Symptom: A CES PVC cannot be cross connected to a VP tunnel interface.
Symptom: The CBR interface on the CES port adapter is always in the up/up state, even when no cable is attached, unless it is administratively shut down. If the CBR line goes down, the interface stays up/up, and the only error reported is a Linestate: LossOfSignal.
Symptom: PNNI support of left-justified E.164 AESA addresses is not available in Cisco IOS Release 11.3 (0.8)TWA4(2) release.
Symptom: Some LANE clients cannot join an ELAN.
If the LANE clients try to join the ELAN, they are rejected.
Symptom: Using more than 13 time slots with cas on-hook, the voice CBR connections were unstable. Without on-hook-detect configuration, all CBR VCs are stable.
Symptom: A ring-back tone cannot be heard from the other side of a PBX when on-hook-detect is set.
Symptom: You cannot use SNMP to create a soft PVC on the ATM switch.
Symptom: The ATM switch might crash if you remove a snooped subinterface by entering the no interface command and then entering the show atm snoop-vc command on the snoop interface.

Cisco IOS Release 11.3(3a)WA4(6) Caveats

This section describes unsupported features and limitations, caveats, and some potentially unexpected behavior by the LightStream 1010 ATM switch for Cisco IOS Release 11.3(3a)WA4(6).

The following caveats have been identified for this release of the LightStream 1010 ATM switch software:

Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX light emitting diode (LED) lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Version 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), are configured on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it is reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the Interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In an LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: The system does not support CUGs on soft PVC connections. For example, if you define a set of access policies and a CUG on an interface with the PVC leg of a soft PVC call, the CUG parameters do not apply to those calls.
Workaround: None.
Symptom: If there are multiple parallel paths to the same destination on a LightStream 1010 ATM switch with a FC-PFQ feature card installed, the tag switching VCs (TVCs) should be load balanced on a per network prefix basis over these parallel paths (up to a maximum of 4 parallel paths) instead of being VC merged. Load balancing does not happen in some cases and the TVCs might be VC merge and go out of the switch as a single VC.
Workaround: None.
Symptom: TCP: %SYS-3-MGDTIMER:Uninitialized timer : mgd_timer_complain_uninit.
Workaround: None.
Symptom: The ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might get stuck in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: Shutting down an interface with more than 1000 tag switching VCs may cause a CPUHOG occurrence for the EXEC process.
Workaround: There is no Workaround for this problem. The CPUHOG occurrence does not impact the functionality of the ATM switch.
Symptom: IP host-routing does not disable when specified.
Workaround: Enable and disable ip routing and then save the configuration to NVRAM.
Symptom: CPU hog while setting up a lot of calls.
Symptom: Open shortest path first (OSPF) does not recognize more than four parallel interfaces. This might cause some tag VCs (TVCs) to not get switched to other interfaces if a tunnel carrying the TVCs is shut down.
Workaround: Enter a clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches to bring everything back up most of the time. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut to bring everything back up.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make a conversion of a requested rate to the actual rate.
Symptom: When you set the ROM monitor environment variable boot to a nonexistent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: A LightStream 1010 ATM switch installed in the Catalyst 5500 chassis might fail to boot up when using mixed power supplies at high temperatures.
When using power supplies from different vendors on the Catalyst 5500 chassis, make sure that the supplies are powered-up in serial order. If you power up the power supplies simultaneously, the ASP on the LightStream 1010 ATM switch might not come out of reset. For example, the ASP might hang and not boot.
Symptom: Funnel VCs can jeopardize quality of service (QoS) for services.
The current multipoint-to-point funnel implementation can compromise the QOS guarantees of other connections (guaranteed services) when the application that created the funnel SVC malfunctions. For example, if the application were to transmit traffic on more than one leg of the funnel SVC simultaneously, the rate scheduler on the output interface will over-subscribe and, potentially, affect the peak cell rate (PCR), sustained cell rate (SCR), and maximum cell rate (MCR) guarantees for other VCs on the interface.
Symptom: ILMI does not come up upon reload of vp-mux (vuni).
ILMI on the tunnel interface (connected to the far-end peer device through a vp-mux switch and virtual UNI) has trouble staying in "WaitDevType" after reload of the vp-mux switch.
Workaround: This situation can be cleared by shutting down the tunnel interface; disabling the ILMI followed by the no shut command on the tunnel interface. After the interface settles down, the ILMI on the tunnel can be re-enabled. Shut the interface. Use the commands atm ilmi-enable and atm auto-configuration (if desired), then re-enable the tunnel interface with the no shut command.
Symptom: ATM Accounting : Exception in validblock_diagnose because resources at destination switch were not available.
Symptom: When a large amount of data is sent out of the ethernet interface, the interface can become overwhelmed and start generating the error message:
%SONICT-3-INTERNAL_ERROR: sonic_send: no free tbufs
The interface stops sending data and the transmitter remains stuck.
Workaround: Issue a shut/no shut command sequence on the interface.
Symptom: Per VC Snooping unconfigurable or broken.
Workaround: None.
Symptom: Crash at pnni_route-bg
Left-justified AESA feature can cause the system to crash when enabling and disabling a node with 3000 e164/f69/X121 addresses.
Workaround: None.
Symptom: ADD Party over Tunnel does not work.
On a source node, if you are adding a subsequent leaf to a VP tunnel that is out of bandwidth, PNNI might reject the route request (for the leaf) leading to an eventual rejection of the ADD party by signaling.
Workaround: None.
Symptom: PNNI: crash at compare_avl_nodes.
When using the pnni_election command with a freed pnni_pdb where there are two scripts running pnni commands from different threads, the CLI command creates the pnni node and passes the pnni_pdb in the csb to be processed by the next issued command. In the mean time SNMP deletes the node from under CLI.
Workaround: When CLI resumes the next command, use a copy of the node pdb that has been freed by SNMP. Get the pnni_pdb from the PNNI_node_tbl entry if it exists. Otherwise, the switch aborts the execution of the command. Modify all of the CLI node commands to use the new scheme.
Symptom: PNNI: MIBS, wrong link-type in pnniLinkTable.
The PNNI agent returned a value of lowestLevelOutsideLink(4) for links between nodes on the same peer group for the pnniLinkType object. The correct value is lowestLevelHorizontalLink(2).
Workaround: None.
Symptom: Crash in process_handle_watchdog
This problem is caused by a routing loop, which is caused by metric overflow in metric aggregation.
Workaround: None.
Symptom: Power-on diagnostic test indicates snake-test failures from the ASP-PFQ.
If the snake-test fails or the switch is not reset, the red led on the left side of the ASP remains lit. Changing out power supplies and ASPs does not correct this problem. Consequently, there is no evidence that this alarm indicates a failed ASP.
Workaround: None.
Symptom: The snmp get command does not work for the following MIB variables:
Workaround: None.
Symptom: Crash with soft release 11-3-3aWA4-6.
Data flowing over OC12 links between two LightStream 1010 ATM switches can cause potential memory corruption leading to system crashes. This crash may be caused due to badly formed AAL5 packets.
Workaround: Reboot the switch.
Symptom: CES interfaces do not come up upon reload.
The Line state stays at XmtAIS LossOfSignal. Sometimes the Switch comes up, but the IF status of show ces int always shows as DOWN and the line state is RcvAIS XmtAIS.
Workaround: None.
Symptom: CES: per-vc queues fill with cells on 64Kbps circuits.
With standard CES 64Kbps circuits, the per-VC queues build up with cells at a rate of about one per second. There is a mismatch between the cell-rate requested and the rate granted. This is caused by signalling a pcr of 170 cells/second, rather than the desired 172 cells/second. Because the CBR FC-PFQ traffic shaping in current software has excellent granularity, roughly a cell every two seconds is queued. After the queue fills, dropping occurs at that rate.

Cisco IOS Release 11.3(2a)WA4(4) Caveats Corrected

This section describes caveats from Cisco IOS Release 11.3(0.8) TWA4(2) that have been corrected by Cisco IOS Release 11.3(2) TWA4(4) of the LightStream 1010 ATM switch software.

Symptom: When a write bus error occurs, the system memory could be corrupted.
The LightStream 1010 ATM switch supports configuration of multiple switch addresses using the global atm address configuration command. Previously, only the active switch address was used to generate interface related soft PVC addresses. In this release the soft PVC addresses are generated based on all switch addresses, instead of just the active switch address. If you change the switch address, the soft PVC configuration based on old address continues to be valid, allowing you to modify soft PVC configuration later to migrate to the new address.
Symptom: Some vendor switches running LANE 3.0 reject UNI 3.1 connections because of the absence of the ATM signaling AAL5 IE mode field.
In the LANE 3.0 specification, the mode field is mandatory in the AAL5 IE. In the UNI 3.1 specification, the mode field was removed. When the setup originates from the UNI3.1 side, it does not have the mode field. For interpretability, the new CLI atm signalling ie aal5 mode command has been added to allow you to configure a mode field statically to be carried in AAL5 IE while running UNI 3.0. By default, this command is on. On a UNI 3.0 connection side a message mode will be passed if a setup is received from a UNI 3.1 side connection. You can change the ATM signalling AAL5 IE mode from message to stream, or disable the command.
Symptom: An OC12 port adapter on a LightStream 1010 ATM switch, which was configured as a network clock source, results in bad clock quality.
Symptom: When the soft VC connected over a VP tunnel route is broken, it takes approximately 45 seconds to reestablish the soft VC over an alternative path, because the destination end of the soft VC is not released soon enough. Shortening the time it takes ILMI to detect the keepalive failure can speed the process to release the destination end of the soft VC and reestablish an alternative route.
In Cisco IOS Release 11.3 (0.8)TWA4(2), ILMI would retry the keepalive polling five times (a fixed number) before declaring the link down. You can configure the keep-alive timer between 1 to 429,4967 seconds (default is 5 seconds).
To solve this problem, Cisco IOS Release 11.3(2a)WA4(4) has an additional keyword, retry, in the atm ilmi-keepalive command, and the timer accuracy has been improved.
The new command changes are as follows:
atm ilmi-keepalive [seconds [retry retry]]
Symptom: If you deleted and recreated the PNNI lowest-level nodes simultaneously on multiple nodes in the same network using the no node node_number lowest command in the config-atm-router mode, you might cause one of these nodes to crash.
Symptom: The input queue on interface atm 2/0/0 of the LightStream 1010 ATM switch, configured as a Token Ring LANE client, might fill up with 34-byte packets. The packets originate from a Cisco 4700 router LEC, and are destined to the LEC of the LightStream 1010 ATM switch.
Symptom: When a LightStream 1010 ATM switch running Cisco IOS Release 11.3 is connected to another running Cisco IOS Release 11.2, and it tries to allocate VCI to equal 32 (VCI=32), the switch running 11.3 does not accept the value because the VCI values 32, 33, and 34 are reserved in Cisco IOS Release 11.3 and later.
In this case the call will not succeed for the first three attempts, and will only succeed once the vpi/vci assignments increment to 35. There are no operational problems with this caveat.
Symptom: An ATM-signaling SETUP message does not automatically negotiate-down the PCR to the minimum line rate through which the SETUP passes.
This occurs if the SETUP enters the switch by way of a network to network interface (NNI). If the SETUP enters the switch by way of a UNI 4.0 interface, the PCR to the minimum line rate negotiation is performed.
Symptom: PNNI port numbers for internal prefixes are updated incorrectly in the routing table when the received prefixes advertise non-zero port numbers. This problem occurs only with internal prefixes and not with exterior routes. This is only important when the routes have to be flushed and are not removed from the routing table; you can see these routes using the show atm route command. However, these entries do get removed from the PNNI database, and the routes are marked as "DN" in the routing table; you can see the stale entries using the show atm route command. This problem does not occur in an all Cisco environments.

Cisco IOS Release 11.3(2a)WA4(4) Caveats

This section describes unsupported features and limitations, caveats, and some potentially unexpected behavior by the LightStream 1010 ATM switch for Cisco IOS Release 11.3(2a)WA4(4).

The following caveats have been identified for this release of the LightStream 1010 ATM switch software:

Symptom: Shutting down an interface with more than 1000 tag switching VCs may cause a CPUHOG occurrence for the EXEC process.
Workaround: There is no Workaround for this problem. The CPUHOG occurrence does not impact the functionality of the ATM switch.
Symptom: If there are multiple parallel paths to the same destination on a LightStream 1010 ATM switch with a FC-PFQ feature card installed, the tag switching VCs (TVCs) should be load balanced on a per network prefix basis over these parallel paths (up to a maximum of 4 parallel paths) instead of being VC merged. Load balancing does not happen in some cases and the TVCs might be VC merge and go out of the switch as a single VC.
Workaround: None.
Symptom: A LightStream 1010 ATM switch installed in the Catalyst 5500 chassis might fail to boot up when using mixed power supplies at high temperatures.
When using power supplies from different vendors on the Catalyst 5500 chassis, make sure that the supplies are powered-up in serial order. If you power-up the power supplies simultaneously, the ASP on the LightStream 1010 ATM switch might not come out of reset. For example, the ASP might hang and not boot.
Symptom: Funnel VCs can jeopardize quality of service (QoS) for services.
The current multipoint-to-point funnel implementation can compromise the QOS guarantees of other connections (guaranteed services) when the application that created the funnel SVC malfunctions. For example, if the application were to transmit traffic on more than one leg of the funnel SVC simultaneously, the rate scheduler on the output interface will over-subscribe and, potentially, affect the peak cell rate (PCR), sustained cell rate (SCR), and maximum cell rate (MCR) guarantees for other VCs on the interface.
Symptom: Occasionally, some LANE clients cannot join an ELAN.
Workaround: None.
Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: E3 port adapters are limited to 65535 cells per second per PVC regardless of the framing type. There is no benefit from G832 framing.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX light emitting diode (LED) lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Version 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), are configured on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it is reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the Interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: The fail over time for the circuit emulation switch (CES) soft VC on a redundant link might be inconsistent, from less than 1 second up to 15 seconds.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In an LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: The system does not support CUGs on soft PVC connections. For example, if you define a set of access policies and a CUG on an interface with the PVC leg of a soft PVC call, the CUG parameters do not apply to those calls.
Workaround: None.
Symptom: The ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might get stuck in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: When IP host routing is enabled and saved to NVRAM, the only way you can remove this command from the configuration is to enable and disable IP routing in Cisco IOS Release 11.3 (0.8) TWA4(2).
Workaround: None.
Symptom: Open shortest path first (OSPF) does not recognize more than four parallel interfaces. This might cause some tag VCs (TVCs) to not get switched to other interfaces if a tunnel carrying the TVCs is shut down.
Workaround: Enter a clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches to bring everything back up most of the time. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut to bring everything back up.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make a conversion of a requested rate to the actual rate.
Symptom: A CES PVC cannot be cross connected to a VP tunnel interface.
Workaround: Use CES soft-VCs instead; a soft CES VC will connect to a tunnel interface.
Symptom: When you set the ROM monitor environment variable boot to a nonexistent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: The CBR interface on the CES port adapter is always in the up/up state, even when no cable is attached, unless it is administratively shut down. If the CBR line goes down, the interface stays up/up, and the only error reported is a Linestate: LossOfSignal.
Workaround: Administratively shut down the interface.
Symptom: An ATM-signaling SETUP message does not automatically negotiate-down the PCR to the minimum line rate through which the SETUP passes.
This occurs if the SETUP enters the switch by way of a NNI interface. If the SETUP enters the switch by way of a UNI 4.0 interface, the PCR to the minimum line rate negotiation is performed.
Workaround: None.
Symptom: PNNI support of left-justified E.164 AESA addresses is not available in Cisco IOS Release 11.3 (0.8)TWA4(2) release. This feature will be released at the next maintenance release.
Workaround: None.
Symptom: Occasionally, some lane clients cannot join an ELAN.
Workaround: None.
Symptom: Using more than 13 time slots with cas on-hook, the voice CBR connections were unstable. Without on-hook-detect configuration, all CBR VCs are stable.
Symptom: A ring-back tone cannot be heard from the other side of a PBX when on-hook-detect is set.
Symptom: You cannot use SNMP to create a soft PVC on the ATM switch.
Workaround: None.
Symptom: The ATM switch might crash if you remove a snooped subinterface by entering the no interface command and then entering the show atm snoop-vc command on the snoop interface.
Workaround: None.

Cisco IOS Release 11.3(0.8)TWA4(2) Caveats Corrected

This section describes caveats from Cisco IOS Release 11.2(8.0.1)FWA4(1) that have been corrected by Cisco IOS Release 11.3(0.8)TWA4(2) of the LightStream 1010 ATM switch software.

Symptom: The bug corrupts the stack and the stack eventually crashes while calling find_buddy repeatedly. A watchdog timeout then causes a reload.
Symptom: Spurious memory access messages appear when you enter the debug atm packet command on one of the interfaces of the LightStream 1010 ATM switch. This caveat was first found in Cisco IOS Release 11.2 (8.0.1) FWA4(1).
Symptom: With multiple LES/BUS pairs configured on a LightStream 1010 ATM switch and about 20 LECs configured to join the ELANs, repeated shutdown or no shutdown of the interface for an extended time could produce the following error message:
ATMS2000_add_party_to_output_only_errors
Symptom: The system sends out excessive ARPs on the Ethernet interface when ip host-routing is enabled.
Symptom: When you configure LES/BUS LANE service on the CPU port, and there is heavy signaling traffic, usually when the switch resets and tries to come back up with a large number of clients for each LES/BUS, the following warning message appears:
CPUHOG - atmsig_output
This has no impact on the system functionality or performance. This caveat was first found in Cisco IOS Release 11.2 (8.0.1) FWA4(1).
Symptom: When you set multiple VP tunnels between the same neighbor, PNNI does not switch to a better tunnel for flooding when an alternate is possible. Using a VP tunnel that has less tolerance for cell delay variation (CDV) causes cell drops and a failure to complete database synchronization. This caveat was first noticed in Cisco IOS Release 11.2 (8.0.1) FWA4(1).
Symptom: OAM cells with invalid type are passed through the switch rather than being discarded.
Symptom: You can not configure the LightStream 1010 switch for a core-dump using FTP because the FTP subsystem is not included in the LightStream 1010 image. This caveat was first found in Cisco IOS Release 11.2(8.0.1)FWA4(1).
Symptom: Globally disabling ATM accounting did not turn off file_full traps. This caveat was first discovered in Cisco IOS Release 11.2(8.0.1)FWA4(1).

Cisco IOS Release 11.3(0.8)TWA4(2) Caveats

This section describes unsupported features and limitations, caveats, and some potentially unexpected behavior by the LightStream 1010 ATM switch for Cisco IOS Release 11.3TWA4.

The following caveats have been identified for this release of the LightStream 1010 ATM switch software:

The LightStream 1010 ATM switch supports configuration of multiple switch addresses using the global atm address configuration command. Only the active switch addresses are used to generate interface related soft PVC addresses.
This problem has been resolved in Cisco IOS Release 11.3(2a)WA4(4).
Workaround: Update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: Some vendor switches running LANE 3.0 reject UNI 3.1 connections because of the absence of the ATM signaling AAL5 IE mode field.
In the LANE 3.0 specification, the mode field is mandatory in the AAL5 IE. In the UNI 3.1 specification, the mode field was removed. When the setup originates from the UNI 3.1 side, it does not have the mode field. For interoperability, the new CLI atm signalling ie aal5 mode command has been added to Cisco IOS Release 11.3(2a)WA4(4) of the software to allow you to configure a mode field statically to be carried in AAL5 IE while running UNI 3.0. By default, this command is on. On a UNI 3.0 connection side, a message mode will be passed if a setup is received from a UNI 3.1 side connection you can change the ATM signaling AAL5 IE mode from message to stream, or disable the command.
Workaround: Update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: An OC12 port adapter on a LightStream 1010 ATM switch, which was configured as a network clock source, results in bad clock quality.
This problem has been resolved in Cisco IOS Release 11.3(2a)WA4(4).
Workaround: Update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: When the soft VC over a VP tunnel route is broken, it takes approximately 45 seconds to reestablish the soft VC over an alternative path because the destination end of the soft VC is not released soon enough. Shortening the time it takes ILMI to detect the keepalive failure can speed the process to release the destination end of the soft VC and re-establish an alternative route.
In release11.3(0.8) TWA4(2), ILMI would retry the keepalive polling five times (a fixed number) before declaring the link down. You can configure the keepalive timer between 1 to 429,4967 seconds (default is 5 seconds).
To solve this problem, Cisco IOS Release 11.3(2a)WA4(4) has an additional keyword, retry, in the atm ilmi-keepalive command, and the timer accuracy has been improved.
The new command changes are as follows:
atm ilmi-keepalive [seconds retry]]
This problem has been resolved in Cisco IOS Releases 11.3(2a)WA4(4) and 11.3(02.03)CI07.
Workaround: Modify the ATM ILMI keepalive timer to less than 5 seconds, or update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: If you delete and recreate the PNNI lowest-level nodes simultaneously on multiple nodes in the same network using the no node node_number lowest command in config-atm-router mode, you might cause one of these nodes to crash.
This problem has been resolved in Cisco IOS Release 11.3(2a)WA4(4).
Workaround: Update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: The input queue on interface atm 2/0/0 of the LightStream 1010 ATM switch, configured as a Token Ring LANE client, might fill up with 34-byte packets. The packets originate from a Cisco 4700 router LEC, and are destined to the LEC of the LightStream 1010 ATM switch.
This problem has been resolved in Cisco IOS Release 11.3(2a)WA4(4).
Workaround: Update the software to Cisco IOS Release 11.3(2a)WA4(4).
Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: E3 port adapters are limited to 65535 cells per second per PVC regardless of the framing type. There is no benefit from G832 framing.
Workaround: None.
Symptom: When more than 1,000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions, the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX light emitting diode (LED) lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Release 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If you configure different Emulated Local Area Networks (ELANs), using Cisco IOS Release 11.1(X), on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN:
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might be occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it reenabled after a system reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: Supporting ABR VP tunnels require interaction between VP flow-control and VC marking that neither FC-PCQ nor FC-PFQ hardware can perform. The configuration of an ABR VP tunnel subinterface is prevented.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: The failover time for the circuit emulation switch (CES) soft VC on a redundant link might be inconsistent, from less than 1 second up to 15 seconds.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up, leaving it in a releasing or inactive state.
Workaround: None.
Symptom: In an LightStream 1010 ATM switch equipped with an FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: The system does not support CUGs on soft PVC connections. For example, if you define a set of access policies and a CUG on an interface with the PVC leg of a soft PVC call, the CUG parameters do not apply to those calls.
Workaround: None.
Symptom: The LightStream 1010 ATM switch does not currently support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might stick in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: When a LightStream 1010 ATM switch running Cisco IOS Release 11.3 is connected to another running Cisco IOS Release 11.2, and it tries to allocate VCI to equal 32(VCI=32), the LightStream1010 ATM switch running Cisco IOS release 11.3 does not accept the value because the VCI values 32, 33, and 34 are reserved in Cisco IOS Release 11.3 and later.
In this case, the call will not succeed for the first three attempts, and will only succeed once the vpi/vci assignments increment to 35. There are no operational problems with this caveat.
Workaround: Configure atmfAtmLayerMinSvccVci to 32 on the switch running Cisco IOS Release 11.3.
Symptom: When IP host routing is enabled and saved to NVRAM, the only way to remove this command from the configuration is to enable and disable IP routing in Cisco IOS Release 11.3 (0.8) TWA4(2).
Workaround: None.
Symptom: OSPF does not recognize more than four parallel interfaces causing some tag VCs (TVCs) to not getting switched to other interfaces if a tunnel carrying the TVCs is shutdown.
Workaround: Enter the clear ip route command on the switch on the interface that was shut, or a clear ip route command on all the switches seems to bring everything back up most of the time. If the clear ip route command does not work, enter a shut/no shut command on the UNI interfaces on the switch on which the physical interface was shut to bring everything back up.
Symptom: There is a small divergence in the measured output of the shaped VP tunnels at rates of 90 Mbps and above.
Workaround: None.
Symptom: The granularity provided by the FC-PFQ feature card scheduling hardware does not allow an exact match of all requested cell rates. To satisfy the traffic contract guarantee, the next higher available scheduling value is used. A shaped VP tunnel is used frequently to pass data to a WAN VP trunk, and limits the traffic transmitted to the scheduled rate. This can cause cells to be dropped in the WAN. Any dropping must be done prior to multiplexing onto the VP, so that a packet discard can be performed.
Workaround: Refer to the information in the DDTs for information on how to make the conversion of a requested rate to the actual rate.
Symptom: A CES PVC cannot be cross connected to a VP tunnel interface.
Workaround: Use CES soft-VCs instead; a soft CES VC will connect to a tunnel interface.
Symptom: When you set the ROM monitor environment variable boot to a non existent file (using the Cisco IOS command boot system flash) and the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch; a break character is sent to the switch to force it to the ROM monitor prompt. You can then manually reboot the switch.
Symptom: The CBR interface on the CES port adapter is always in the up/up state, even when no cable is attached, unless it is administratively shut down. If the CBR line goes down, the interface stays up/up, and the only error reported is a Linestate: LossOfSignal.
Workaround: Administratively shut down the interface.
Symptom: An ATM signaling SETUP message progressing through the switch does not automatically negotiate-down the PCR to the minimum line rate through which the SETUP passes.
This occurs if the SETUP enters the switch by way of an NNI interface. If the SETUP enters the switch by way of a UNI 4.0 interface, the PCR to the minimum line rate negotiation is performed.
Workaround: None.
Symptom: PNNI support of left justified E.164 AESA addresses is not available in Cisco IOS Cisco IOS Release 11.3 (0.8)TWA4(4) release. This feature will be released at the next maintenance release.
Workaround: None.

Cisco IOS Release 11.2(8.0.1)FWA4(1) Caveats Corrected

This section describes caveats from Cisco IOS Release 11.2(8)WA3(3) that have been corrected by Release 11.2(8.0.1)FWA4(1) of the LightStream 1010 ATM switch software.

Symptom: Under some circumstances, the Ethernet port on the LightStream 1010 ATM switch hangs. When you enter the show interface ethernet 2/0/0 command, the interface is shown up, but the output queue is full.
Symptom: The ATM accounting periodic collection writes more records than the number of VCs it is monitoring.
Symptom: The clear counters command can cause a T1 frame slip on the CES ports.
Symptom: For PVPs, virtual path identifier (VPI)=0 is not allowed. This restriction has been relaxed for the hard transit PVPs.
Symptom: On 25-Mbps port adapters, entering the show atm vc command can potentially crash the switch.
Symptom: There is a memory leak in Cisco IOS Release 11.2.8.
Symptom: The Service Specific Connection Oriented Protocol (SSCOP) layer sequence number wraparound condition leads to memory leaks and memory fragmentation problems. The problem occurs when the SSCOP's send sequence number reaches a maximum value of 16777215. You must reset the switch to restore normal functionality.
Symptom: The switch does not support restart based on the virtual path identifier/virtual channel identifier (VPI/VCI).
Symptom: When a reload is entered, the following message might appear. 
%ALIGN-3-SPURIOUS: Spurious memory access made at 0x600D2EBC reading 0x0
This caveat does not affect the operation of the switch.
Symptom: Logical multicasting is not supported.
Symptom: A DS3 port adapter alarm indication signal (AIS) is not sent from a shutdown port when reloading the switch software. This might cause one side of a link to appear up and the other side of a link to appear down. It only occurs if one side of the link is shut down during a power-up cycle. If the interface is brought up, both sides of the link resume normal operation.
Symptom: If a T1/E1 port is configured as the clock source for the system, and the port receives transitions in link status that affect the clock, other interfaces that derive their clock from that clock source might go down for a moment. This occurs only when transitions are less than one second.
Symptom: The receive cell counter on the virtual circuit VPI=0, VCI=32 displays erroneous cell counts.
Symptom: The SNMP agent on the switch always returns a "Slot empty" value, even if a Flash memory card is present in the Personal Computer Memory Card International Association (PCMCIA) slot.
Symptom: ATM accounting background notification is not supported.
Symptom: The LightStream 1010 ATM switch might log the following message:
%SYS-3-MGDTIMER:Uninitialized timer, timer stop, timer = [hex].
This message does not affect the switch operation.
Symptom: If you hot swap a T1/E1 port adapter with an OC3 port adapter, the switch might hang.
Symptom: The TTB length bit is not set for G832 framing. The TTB messages, which check for the continued presence of the intended transmitter, must be enabled for G832Adm framing.

Cisco IOS Release 11.2(8.0.1)FWA4(1) Caveats

This section describes unsupported features and limitations, caveats, and some potentially unexpected behavior by the LightStream 1010 ATM switch for Cisco IOS Release 11.2(8.0.1)FWA4(1).

The following caveats have been identified for this release of the LightStream 1010 ATM switch software:

Symptom: Remote defect identification (RDI) cells sent by an end point in response to alarm indication signal (AIS) cells generated at an intermediate switch with a fault condition on an interface are not propagated beyond the intermediate switch. The intermediate switch removes the connection leg entries for both interfaces participating in the connection when a fault is discovered on one of the interfaces, even though the other interface might still be up. As a result, the RDI cells are dropped at the intermediate switch.
Workaround: None.
Symptom: E3 port adapters are limited to 65535 cells per second per PVC regardless of the framing type. There is no benefit from G832 framing.
Workaround: None.
Symptom: When more than 1000 SVCs are active on an interface and the shutdown command is entered, all SVCs on that interface are released and the following message appears:
%SYS-3-CPUHOG
This message indicates that the release process runs for a long time before returning control to the kernel, which can then schedule other tasks. This process does not affect normal operation of the switch.
Workaround: None.
Symptom: Under very heavy traffic conditions the switch might experience temporary queue cell failures. This should clear after the traffic congestion clears.
Workaround: None.
Symptom: The receiver circuitry on DS3 port adapters can interpret noise as a valid signal. This signal is framed incorrectly and does not contain real data. The DS3 controller interprets the signal as a bad signal instead of no signal, and the red RX light emitting diode (LED) lights up.
Workaround: None.
Symptom: When a large number of LANE clients come up and down constantly over extended periods of time, the system can run out of AAL5 buffers and the following message appears:
%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
In most cases, the system continues to function normally, but occasionally the system denies additional calls from end systems.
Workaround: Toggle the interface to reinitialize all connections and restore normal operation.
Symptom: A LECS, using Cisco IOS Release 11.2(X), expects all LESs to establish an individual control VC to the LECS in order to validate clients.
If different Emulated Local Area Networks (ELANs), using Cisco IOS Version 11.1(X), on multiple subinterfaces of the same physical interface, then all LES(s) multiplex the control messages (which validate the clients) into a single VC.
For example, see the following LES router configuration: 
atm1/0.1    sysa_70k_31_a1.1_LAN sysa_70k_31_a1.1_LAN
atm1/0.2    sysa_70k_31_a1.2_LAN sysa_70k_31_a1.2_LAN
atm1/0.3    sysa_70k_31_a1.3_LAN sysa_70k_31_a1.3_LAN
atm1/0.4    sysa_70k_31_a1.4_LAN sysa_70k_31_a1.4_LAN
This configuration of an LECS sends the following warning messages to the console stating that an LES of one ELAN is attempting to obtain information about another ELAN: 
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.1_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.2_LAN'
%LANE-4-LECS_WARNING: interface ATM2/0/0: elan 'sysa_70k_31_a1.4_LAN' LES asking for elan 
'sysa_70k_31_a1.3_LAN'
The clients are still allowed to join the ELAN. Disregard the warning message.
Workaround: None.
Symptom: You can create variable bit rate (VBR), available bit rate (ABR), and unspecified bit rate (UBR) VCs across the switch with peak cell rate (PCR) values greater than the interface line rate. However, the actual allocated bandwidth continues to be:
Workaround: None.
Symptom: When one of the installed power supplies is powered OFF, and you are copying an image to bootflash, a power supply failure message appears.
Workaround: None.
Symptom: The call attempt counters for PortSelectGroups might not count the outgoing calls on its NNI interfaces because of switch crankback attempts. This might result in a discrepancy between the call attempt counters shown on Portselgroup representing the interface on which the call came in and the counters shown on the PortSelgroup representing the interface over which an attempt was made to forward the call. This problem might occur when a call fails.
Workaround: None.
Symptom: If you disable Cisco Discovery Protocol (CDP) on the 2/0/0.x interface and it is configured as a LEC, the information is not saved to NVRAM when you enter the copy running-config startup-config command. CDP is disabled in the running configuration, but it reenabled after a system reload.
Workaround: None.
Symptom: This caveat is caused by a compiler bug. The bug corrupts the stack and eventually crashes while calling find_buddy repeatedly. A watchdog timeout then causes a reload.
Workaround: None.
Symptom: ATM RMON data collection is not supported on subinterfaces or tunnels. The atm rmon collect command is accepted on hardware interfaces only; it is ignored on subinterfaces. SNMP and NVGEN support (by way of portSelTable) is not possible until the interfaces MIB (RFC 1573) entries are added for tunnel subinterfaces. ATM RMON counters for a hardware interface do not include any of the traffic through tunnels configured on the interface.
Workaround: None.
Symptom: The cell count reported on a snooped interface is twice the actual number of cells transmitted.
Workaround: None.
Symptom: SNMP support of the ciscoAtmIfPhysEntryData table and LED information is not available on the 25-MB port adapter.
Workaround: None.
Symptom: Cell loss might occur while hot swapping a power supply.
Workaround: None.
Symptom: The fail over time for the circuit emulation switch (CES) soft VC on a redundant link might be inconsistent, from less than 1 second up to 15 seconds.
Workaround: None.
Symptom: Spurious memory access messages appear when you enter the debug atm packet command on one of the interfaces of the LightStream 1010 ATM switch.
Workaround: None.
Symptom: When memory is almost or completely exhausted and a soft PVC goes down, it might not come back up.
Workaround: None.
Symptom: With multiple LES/BUS pairs configured on a LightStream 1010 ATM switch and about 20 LECs configured to join the ELANs, repeated shutdown or no shutdown of the interface for an extended time could produce the following error message:
ATMS2000_add_party_to_output_only_errors
Workaround: None.
Symptom: The system sends out excessive ARPs on the Ethernet interface when ip host-routing is enabled.
Workaround: None.
Symptom: In an LightStream 1010 ATM switch equipped with a FC-PFQ, the maximum number of cells available for use is 64511. The number of cells in the switch fabric is 65535.
Workaround: None.
Symptom: When you set multiple VP tunnels between the same neighbor, PNNI does not switch to a better tunnel for flooding when an alternate is possible. Using a VP tunnel that has less tolerance for CDV causes cell drops and a failure to complete database synchronization.
Workaround: None.
Symptom: When you configure LES/BUS LANE service on the CPU port, and there is heavy signalling traffic, usually when the switch resets and tries to come back up with a large number of clients for each LES/BUS, the following warning message appears:
CPUHOG - atmsig_output
This has no impact on the system functionality or performance.
Workaround: None.
Symptom: You can configure the LightStream 1010 ATM switch for a core-dump using FTP. This is caused because the FTP subsystem is not included in the LightStream 1010 image. This caveat was first found in Cisco IOS Release 11.2 (8.0.1).
Workaround: None.
Symptom: The LightStream 1010 ATM switch does not support maxvc-number negotiation through ILMI.
Workaround: None.
Symptom: Under some rare conditions (not yet identified), some ports might stick in the WaitDevType state.
Workaround: Reboot the switch.
Symptom: When a LightStream 1010 ATM switch running Cisco IOS Release 11.2 is connected to another running Cisco IOS Release 11.3, and the switch running Cisco IOS Release 11.2 tries to allocate vci to equal 32 (vci=32), the switch running 11.3 does not accept the value, because the vci values 32, 33, and 34 are reserved in Cisco IOS Release 11.3 and later. In this case the call will not succeed for the first three attempts, and will only succeed once the vpi/vci assignments has incremented to 35. There are no operational problems with this caveat.
Workaround: Configure atmfAtmLayerMinSvccVci to 32 on the switch running Cisco IOS Release 11.3.
Symptom: When IP host routing is enabled and saved to NVRAM, the only way to remove this command from the configuration is to delete and rebuild the NVRAM in Cisco IOS Release 11.2 (8.0.1) FWA4(1).
Workaround: None.
Symptom: OAM cells with invalid type are passed through the --switch rather than being discarded.
Workaround: None.
Symptom: Globally disabling ATM accounting did not turn off file_full traps.
Workaround: None.
Symptom: A CES PVC cannot be cross-connected to a tunnel interface.
Workaround: Use CES soft-VCs instead; a soft CES VC will connect to a tunnel interface.
Symptom: When the ROM monitor environment variable boot is set to a non-existent file (using the boot system flash command) and if the configuration register is set to 0x2102 (autoboot), the switch hangs during the subsequent reload command issued by the software.
Workaround: Power-cycle the switch, a break character is sent to the switch to force it to the rom monitor prompt. The switch can then be manually re-booted.
Symptom: An ATM signalling SETUP message progressing through the switch does not automatically negotiate-down the peak cell rate (PCR) to the minimum line rate through which the SETUP passes.
This occurs if the SETUP enters the switch by way of an NNI interface. If the SETUP enters the switch by way of a UNI 4.0 interface, such negotiation is performed.
Workaround: None.

Documentation Updates

An Update to the LightStream 1010 ATM Switch and Catalyst 8510 MSR Command Reference document was created to reflect information added after printing the manual.

Related Documentation

The following sections describe the documentation available for the LightStream 1010 ATM switch and Catalyst 8510 MSR. These documents consist of hardware installation guides, software installation guides, Cisco IOS configuration and command references, system error messages, and feature modules, which are updates to the Cisco IOS documentation. Documentation is available as printed manuals or electronic documents, except for feature modules, which are available online only.

The most up-to-date documentation can be found on the Web by way of Cisco Connection Online (CCO) and the Documentation CD-ROM. These electronic documents might contain updates and modifications made after the hard-copy documents were printed.

These release notes should be used in conjunction with the documents listed in this section.

Platform Documents

Following is a list of the platform specific documentation available for the LightStream 1010 ATM switch and Catalyst 8510 MSR:

Software Documents

Following is a list of the software documentation available for the LightStream 1010 ATM switch and Catalyst 8510 MSR:

Service and Support

For service and support for a product purchased from a reseller, contact the reseller. Resellers offer a wide variety of Cisco service and support programs, which are described in the section "Service and Support" in the information packet that shipped with your product.

Note If you purchased your product from a reseller, you can access CCO as a guest. CCO is Cisco Systems' primary real-time support channel. Your reseller offers programs that include direct access to CCO services.

For service and support for a product purchased directly from Cisco, use CCO.

Software Configuration Tips on the Cisco TAC Home Page

For helpful tips on configuring Cisco products, follow this path on CCO:

Software & Support: Technical Tips (button on left margin)

"Hot Tips" are popular tips and hints gathered from Cisco's Technical Assistance Center (TAC). Most of these documents are also available from the TAC's Fax-on-Demand service. To access Fax-on-Demand and receive documents at your fax machine, call 888-50-CISCO (888-502-4726). From international areas, call 650-596-4408.

The following sections are provided from the Technical Tips page:

Cisco Connection Online

Cisco Connection Online (CCO) is Cisco Systems' primary, real-time support channel. Maintenance customers and partners can self-register on CCO to obtain additional information and services.

Available 24 hours a day, 7 days a week, CCO provides a wealth of standard and value-added services to Cisco's customers and business partners. CCO services include product information, product documentation, software updates, release notes, technical tips, the Bug Navigator, configuration notes, brochures, descriptions of service offerings, and download access to public and authorized files.

CCO serves a wide variety of users through two interfaces that are updated and enhanced simultaneously: a character-based version and a multimedia version that resides on the World Wide Web (WWW). The character-based CCO supports Zmodem, Kermit, Xmodem, FTP, and Internet e-mail, and it is excellent for quick access to information over lower bandwidths. The WWW version of CCO provides richly formatted documents with photographs, figures, graphics, and video, as well as hyperlinks to related information.

You can access CCO in the following ways:

For a copy of CCO's Frequently Asked Questions (FAQ), contact cco-help@cisco.com. For additional information, contact cco-team@cisco.com.

Note If you are a network administrator and need personal technical assistance with a Cisco product that is under warranty or covered by a maintenance contract, contact Cisco's Technical Assistance Center (TAC) at 800 553-2447, 408 526-7209, or tac@cisco.com. To obtain general information about Cisco Systems, Cisco products, or upgrades, contact 800 553-6387, 408 526-7208, or cs-rep@cisco.com.

Documentation CD-ROM

Cisco documentation and additional literature are available in a CD-ROM package, which ships with your product. The Documentation CD-ROM, a member of the Cisco Connection Family, is updated monthly. Therefore, it might be more current than printed documentation. To order additional copies of the Documentation CD-ROM, contact your local sales representative or call customer service. The CD-ROM package is available as a single package or as an annual subscription. You can also access Cisco documentation on the World Wide Web at http://www.cisco.com, http://www-china.cisco.com, or http://www-europe.cisco.com.

If you are reading Cisco product documentation on the World Wide Web, you can submit comments electronically. Click Feedback in the toolbar and select Documentation. After you complete the form, click Submit to send it to Cisco. We appreciate your comments.





hometocprevnextglossaryfeedbacksearchhelp
Posted: Thu Mar 11 12:14:27 PST 1999
Copyright 1989-1999©Cisco Systems Inc.