|
|
Table Of Contents
Release Notes for the Catalyst 8540 for Cisco IOS Release 12.1(26)E1, Part 2: Caveats
Resolved Caveats—Cisco IOS Release 12.1(26)E1
Open Caveats—Cisco IOS Release 12.1(26)E
Resolved Caveats—Cisco IOS Release 12.1(26)E
Resolved Caveats—Cisco IOS Release 12.1(23)E1
Resolved Caveats—Cisco IOS Release 12.1(23)E
Resolved Caveats—Cisco IOS Release 12.1(22)E
Resolved Caveats—Cisco IOS Release 12.1(20)EB
Resolved Caveats—Cisco IOS Release 12.1(20)E
Resolved Caveats—Cisco IOS Release 12.1(19)EB
Resolved Caveats—Cisco IOS Release 12.1(19)E2
Resolved Caveats—Cisco IOS Release 12.1(14)EB1
Resolved Caveats—Cisco IOS Release 12.1(19)E
Resolved Caveats—Cisco IOS Release 12.1(14)EB
Resolved Caveats—Cisco IOS Release 12.1(14)E1
Resolved Caveats—Cisco IOS Release 12.1(13)EB1
Resolved Caveats—Cisco IOS Release 12.1(13)E1
Resolved Caveats—Cisco IOS Release 12.1(12c)EY
Resolved Caveats—Cisco IOS Release 12.1(12c)E1
Resolved Caveats—Cisco IOS Release 12.1(12c)E
Resolved Caveats—Cisco IOS Release 12.1(11)E1
Resolved Caveats—Cisco IOS Release 12.1(11)E
Resolved Caveats—Cisco IOS Release 12.1(10)EY
Resolved Caveats—Cisco IOS Release 12.1(10)E
Resolved Caveats—Cisco IOS Release 12.1(7a)EY1
Resolved Caveats—Cisco IOS Release 12.1(7a)EY
Resolved Caveats—Cisco IOS Release 12.1(6)EY
Resolved Caveats—Cisco IOS Release 12.1(5a)EY
Obtaining Technical Assistance
Obtaining Additional Publications and Information
Release Notes for the Catalyst 8540 for Cisco IOS Release 12.1(26)E1, Part 2: Caveats
April 12, 2005
Catalyst 8540 for Cisco IOS Release 12.1(26)E1
Text Part Number: OL-7461-01, B1
This document lists severity 1 and 2 caveats and select severity 3 caveats for Cisco IOS Release 12.1(5a)EY, up to and including Release 12.1(26)E1. Caveats describe unexpected behavior or defects in Cisco IOS software releases for the Catalyst 8540 MSR and the Catalyst 8510 MSR. Severity 1 caveats are the most serious caveats; severity 2 caveats are less serious.
To improve this document, we would appreciate your comments. If you are reading Cisco product documentation on the World Wide Web, you can submit comments electronically at http://www.cisco.com/feedback/ or you can contact caveats-doc@cisco.com.
For more information, see the "Documentation Feedback" section.
Note
This Cisco IOS Release 12.1(26)E1 is the last release supporting the Catalyst 8540 CSR platform.
How to Use This Document
This document describes open and resolved severity 1 and 2 caveats and select severity 3 caveats:
•
•
Within the sections the caveats are sorted by technology in alphabetical order. For example, AppleTalk caveats are listed separately from, and before, IP caveats. The caveats are also sorted alphanumerically by caveat number.
If You Need More Information
Cisco documentation and additional literature are available on Cisco.com.
For more information on caveats and features in Cisco IOS Release 12.1 T, see the following sources:
•
•
•
•
Note
Resolved Caveats—Cisco IOS Release 12.1(26)E1
Cisco IOS Release 12.1(26)E1 is a rebuild release for Cisco IOS Release 12.1(23)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(26)E1 but may be open in previous Cisco IOS releases.
•
Symptom: Bridge-groups may put a OSPF neighbor in the INIT state. With OSPF neighborship established when you add the bridge-group (in this case vlan-bridge), the neighborship is put into the INIT state. This also applies with DEC and IBM.
Workaround: None
•
A document that describes how the Internet Control Message Protocol (ICMP) could be used to perform a number of Denial of Service (DoS) attacks against the Transmission Control Protocol (TCP) has been made publicly available. This document has been published through the Internet Engineering Task Force (IETF) Internet Draft process, and is entitled "ICMP Attacks Against TCP" (draft-gont-tcpm-icmp-attacks-03.txt).
These attacks, which only affect sessions terminating or originating on a device itself, can be of three types:
1. Attacks that use ICMP "hard" error messages
2. Attacks that use ICMP "fragmentation needed and Don't Fragment (DF) bit set" messages, also known as Path Maximum Transmission Unit Discovery (PMTUD) attacks
3. Attacks that use ICMP "source quench" messagesSuccessful attacks may cause connection resets or reduction of throughput in existing connections, depending on the attack type.
Multiple Cisco products are affected by the attacks described in this Internet draft.
Cisco has made free software available to address these vulnerabilities. In some cases there are workarounds available to mitigate the effects of the vulnerability.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20050412-icmp.shtml.
The disclosure of these vulnerabilities is being coordinated by the National Infrastructure Security Coordination Centre (NISCC), based in the United Kingdom. NISCC is working with multiple vendors whose products are potentially affected. Its posting can be found at: http://www.niscc.gov.uk/niscc/docs/re-20050412-00303.pdf?lang=en.
•
A document that describes how the Internet Control Message Protocol (ICMP) could be used to perform a number of Denial of Service (DoS) attacks against the Transmission Control Protocol (TCP) has been made publicly available. This document has been published through the Internet Engineering Task Force (IETF) Internet Draft process, and is entitled "ICMP Attacks Against TCP" (draft-gont-tcpm-icmp-attacks-03.txt).
These attacks, which only affect sessions terminating or originating on a device itself, can be of three types:
1. Attacks that use ICMP "hard" error messages
2. Attacks that use ICMP "fragmentation needed and Don't Fragment (DF) bit set" messages, also known as Path Maximum Transmission Unit Discovery (PMTUD) attacks
3. Attacks that use ICMP "source quench" messagesSuccessful attacks may cause connection resets or reduction of throughput in existing connections, depending on the attack type.
Multiple Cisco products are affected by the attacks described in this Internet draft.
Cisco has made free software available to address these vulnerabilities. In some cases there are workarounds available to mitigate the effects of the vulnerability.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20050412-icmp.shtml.
The disclosure of these vulnerabilities is being coordinated by the National Infrastructure Security Coordination Centre (NISCC), based in the United Kingdom. NISCC is working with multiple vendors whose products are potentially affected. Its posting can be found at: http://www.niscc.gov.uk/niscc/docs/re-20050412-00303.pdf?lang=en.
•
A document that describes how the Internet Control Message Protocol (ICMP) could be used to perform a number of Denial of Service (DoS) attacks against the Transmission Control Protocol (TCP) has been made publicly available. This document has been published through the Internet Engineering Task Force (IETF) Internet Draft process, and is entitled "ICMP Attacks Against TCP" (draft-gont-tcpm-icmp-attacks-03.txt).
These attacks, which only affect sessions terminating or originating on a device itself, can be of three types:
1. Attacks that use ICMP "hard" error messages
2. Attacks that use ICMP "fragmentation needed and Don't Fragment (DF) bit set" messages, also known as Path Maximum Transmission Unit Discovery (PMTUD) attacks
3. Attacks that use ICMP "source quench" messagesSuccessful attacks may cause connection resets or reduction of throughput in existing connections, depending on the attack type.
Multiple Cisco products are affected by the attacks described in this Internet draft.
Cisco has made free software available to address these vulnerabilities. In some cases there are workarounds available to mitigate the effects of the vulnerability.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20050412-icmp.shtml.
The disclosure of these vulnerabilities is being coordinated by the National Infrastructure Security Coordination Centre (NISCC), based in the United Kingdom. NISCC is working with multiple vendors whose products are potentially affected. Its posting can be found at: http://www.niscc.gov.uk/niscc/docs/re-20050412-00303.pdf?lang=en.
Open Caveats—Cisco IOS Release 12.1(26)E
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(26)E. All of the caveats listed in this section are open in Cisco IOS Release 12.1(26)E1. This section describes severity 1 and 2 caveats and select severity 3 caveats.
•
Symptom: Static programming of bridge MAC entries does not work. If you configure the bridging entry on the ATM router module (ARM) subinterface, the ifindex value is programmed instead of the outgoing VC value and the traffic to this destination MAC is dropped.
Workaround: None.
•
Symptom: CRC errors are seen on the peer interface. When the 2-port Gigabit Ethernet interface module (EPIF based) is connected to other Gigabit interfaces (for example, the Catalyst 6000 or Catalyst 8540 XPIF interface modules) and traffic is transmitted at line rate (about 85 percent of line rate).
Workaround: None.
•
Symptom: The switch might experience high CPU utilization or the mroute cache might not be learned with point-to-multipoint VCs and multicast configured on the ATM router module. This occurs if the point-to-multipoint VCs are configured from an OC-3 interface (root for the PVC) and are terminated from a leaf on the ATM router module interface, which has multicast configured. Then, at high rates of traffic (over 100 Mbps) the mroute table might not contain the S,G entries that cause multicast traffic not to be routed. Sometimes the S,G might be available in the table but the traffic is still routed by the CPU, causing high CPU utilization.
Workaround: Use point-to-point PVCs instead of point-to-multipoint PVCs.
•
Symptom: RMON packet size based counters are not working for 8-port Gigabit Ethernet interfaces. The problem is not found using image version cat8540m-wp-mz.120-22.W5.25.
Workaround: None.
•
Symptom: When a VBR-RT PVC is configured with the UPC drop option, packets below traffic rate are also dropped.
Workaround: None.
•
Symptom: When reloading the Catalyst 8540 MSR switch, the first channel group of the 4CE1 interface module comes up in failed state in the firmware.
Note
Workaround: Enter the shutdown and no shutdown commands on the interface.
•
Symptom: Input and output packet counters do not increment when jumbo frames pass through the interface.
Note
Workaround: None.
•
Symptom: On a Catalyst 8540 MSR switch connected to a third party switch over an IMA connection with only one active link a synchronization error might occur on the third party switch.
This problem does not appear on connections between Cisco equipment, if there are at least two links in the IMA group, or if the E1/T1 link is not an IMA connection. This is caused by delayed ICP cells on the Cisco side causing the synchronization error.
Workaround: Use two or more active links in the IMA group or use a single link without the IMA configuration.
•
Symptom: When the switch tries to flood a PTSP (PNNI Topology State Packet), whose size is greater than 4096 bytes, the switch reloads due to memory corruption. This problem usually occurs a when large number of ATM prefixes are being advertised by a PNNI node.
This problem occurs only when the PTSP being flooded is greater than 4096 bytes.
Workaround: None.
•
Symptom: On a Catalyst 8540 switch with an Enhanced ATM router module installed, multicast traffic streams do not pass through unless "ip mroute-cache" is turned off on the interfaces.
Workaround: Disable "ip mroute-cache" and configure PVCs instead of SVCs.
•
Symptom: ILMI registration might fail on Cisco switches and routers connected to Lucent products.
For example, entering the show atm ilmi-status could display the following:
Switch# show atm ilmi-statusInterface : ATMX/Y Interface Type : Private UNI (User-side)ILMI VCC : (0, 16) ILMI Keepalive : Enabled/Up (5 Sec 4 Retries)ILMI State: UpAndNormalPeer IP Addr: 0.0.0.0 Peer IF Name:Peer MaxVPIbits: 8 Peer MaxVCIbits: 14ILMI PVC is UP, but address registration failed
Note
Workaround: Reload the switch or router.
•
Symptom: On the Catalyst 8540 CSR switch with Cisco IOS Release 12.0(26)W5(28), traffic might be process switched after OIR. This problem might occur if the ingress Gigabit Ethernet interface module, installed in slot 11, transmits traffic through a Fast Ethernet interface module installed in slot 2.
Workaround: None.
•
Symptom: A Catalyst 8540 CSR running Cisco IOS Release 12.1(19)E might continue to reload if more than one Gigabit Ethernet interface module is installed.
Workaround: Disable the IP multicast-routing on the switch.
•
Symptom: The 2-port Gigabit Ethernet (GE) interface module (C85GE-2X-16K) might count good packets as cyclic redundancy check (CRC) errors and cause a port stuck to occur.
Workaround: Instead of terminating the WAN link on the C85GE-2C-16K interface module, add a Layer 2 switch between the Catalyst 8540 MSR and the WAN link.
•
Symptom: On the Catalyst 8540 MSR running Cisco IOS Release 12.1(20)E, the input packet counter wraps when reaching around 65000 packets. The following example shows the interface counters:
53816 packets input, 2852248 bytes, 0 no bufferA few seconds later, you see the following counters for the same interface display:
3067 packets input, 162551 bytes, 0 no bufferThis problem starts after a reload of the switch and is seen on OC-3 interfaces.
Workaround: None.
•
Symptom: A fiber-optic Fast Ethernet interface module connecting a Catalyst 8540 to a Catalyst 5000 interface module, WS-U5533 or WS-U5538, might flap up-and-down intermittently. This might occur once per day or for longer intervals. Both the Catalyst 8540 and Catalyst 5000 ends of the connection log the link flap.
The problem occurs with the following interface modules:
–
–
This problem does not happen with WS-X5236-FX-MT.
Workaround: None.
•
Symptom: On the Catalyst 8540 MSR a cross-connection PVC with a high VCI number cannot be created.
The following error message appears:
Switch(config)# interface atm 0/0/1Switch(config-if)# atm pvc 16 12311 interface ATM0/0/0 16 12311../c8540-atms5000/NP5500/NP55_token_mgmt.c 864: Out of ovc entries for msc 0../c8540-atms5000/NP5500/NP55_conn_resources.c 208: Couldn't allocate ovc - msc 0../c8540-atms5000/NP5500/NP55_connection.c 587: Couldn't allocate output side resources for msc 0%connection creation fails: connection installation failureOn interface ATM0/0/1 (16/12311) - interface ATM0/0/0 (16/12311)Switch# show debugGeneric ATM MMC:ATM MMC ERRORS debugging is onATM Connection Management event debugging is onATM Connection Management error debugging isswitch#Workaround: None.
•
Symptom: Suppose you have four ATM links between a Catalyst 8500 switch and a Cisco 7200 router running Cisco IOS Release 12.2(13) over microwave if you combine these four links into the same IMA group, the router connection can become very unstable.
This problem occurs because there are so many packet drops that an OSPF neighbor connection problem appears over the IMA link.
Workaround: Remove one link from the IMA group and have only three ATM links. This allows you to keep the router connected.
•
Symptom: An unexpected reload on two Catalyst 8540 MSR switches was seen. After booting the switch, the problem is NOT seen again.
The traces pointed to "less stack space" and "mmc subsys init".
Workaround: None.
•
Symptom: The Catalyst 8540 CSR might display the following messages when running Cisco IOS Release 12.0(20)W5(26) software:
Jan 9 160611 EST %ALIGN-3-CORRECT Alignment correction made at0x603954BC reading 0x6200E625Jan 9 160611 EST %ALIGN-3-TRACE -Traceback= 603954BC 6036BD246036BB58 60076ABC 6005C948 6005CA80 6005CDCC 6007C4E4There is no noticeable impact to the operation of the switch.
Workaround: Downgrade to Cisco IOS Release 12.0(20)W5(24a) software or earlier.
•
Symptom: The Catalyst 8540 MSR might suddenly crash and display a traceback.
Workaround: None.
•
Symptom: Internal buffers (vbufs) are exhausted, ASSERTION FAILED, and some high-Priority IPC (Inter-process Communications) drops occur.
Workaround: None.
•
Symptom: PNNI adjacency between directly connected Catalyst 8500 MSR switches fails to come up using the following configuration: both switches running IOS Release 12.1(12c)EY and a Catalyst 8510 MSR connected to a Catalyst 8540 MSR over an OC-3 connection. The following conditions caused the problem to appear:
1.
Note
2.
>>> *Nov 19 09:22:39: %ATM-3-ILMIOPTIPOLLFAIL:>>> Internal error (3) in ILMI OPTIPOLL module3.
*Nov 19 09:30:45: PNNI:72.1 Hello at ATM0/0/3: Tx, state attempt*Nov 19 09:30:45: NodeId: 72:160:39.036F100A000000003100C900.00053243D901.00*Nov 19 09:30:45: Address: 39.036F100A000000003100C900.00053243D901.01*Nov 19 09:30:45: PgId: 72:39.036F.100A.0000.0000.0000.0000*Nov 19 09:30:45: Remote: port: 0 (0), NodeId:0:0:00.000000000000000000000000.000000000000.00*Nov 19 09:30:45: Local port: ATM0/0/3 (80003000)...*Nov 19 09:30:51: PNNI:72.1 Hello at ATM0/0/3: Received packet with unsupportedversion 494.
Nov 21 03:25:14: PNNI:72.1 Hello at ATM9/1/1: Tx, state 1way_inNov 21 03:25:14: NodeId: 72:160:39.036F100A000000003100C100.003094398501.00Nov 21 03:25:14: Address: 39.036F100A000000003100C100.003094398501.01Nov 21 03:25:14: PgId: 72:39.036F.100A.0000.0000.0000.0000Nov 21 03:25:14: Remote: port: ATM0/0/3 (80003000), NodeId:72:160:39.036F100A000000003100C900.00053243D901.00Nov 21 03:25:14: Local port: ATM9/1/1 (84901000)Workaround: None.
•
Symptom: On a Catalyst 8540 MSR running 12.1(12c)EY, the switch might crash if a show tech command is entered while many VCs are being created and torn down.
Workaround: None.
•
Symptom: A Catalyst 8540 CSR configured for IRB/BVI does not perform MAC learning in the microcode based on the reception of ARP requests. This includes gratuitous ARPs. This might cause problems with redundancy and failover mechanisms that rely on advertising the new location of a virtual IP/MAC pair through gratuitous ARP. Gratuitous ARP updates the bridge and ARP tables of devices in the network when a failover occurs. The Cisco PIX firewall also uses this mechanism for failover.
Workaround: There is no workaround for this problem in an IRB configuration on this platform. In a bridging only configuration, this problem does not occur.
Clearing the bridge table removes the condition. After the aging period (5 minutes), the connection is also restored.
•
Symptom: On rare occasions, when a route processor switchover occurs the microcode download on the Enhanced Gigabit Ethernet card might fail.
Workaround: Manually reloading the interface module using epc portstuck-manual-reload interface gigabitEthernet interface-number might solve the problem.
•
Symptom: You should not configure a structured CES circuit with more than 19 timeslots across a T1 ATM interface on a Catalyst 8540 MSR. While the circuit can be configured with more than 19 timeslots and will be built over the T1 ATM interface, the Catalyst 8540 MSR will discard cells on the configured VC causing errors on the emulated circuit.
With 20 timeslots configured you see cells queued and dropped (CLP0 Q full drop) on the PVC. This problem occurs with both hard and soft PVCs are configured over the T1 atm interface.
To detect the problem, use the show atm vc int atm card/subcard/port vpi vci command and confirm the Clp0 q full drops field is increasing.
Switch# show atm vc interfaces atm 2/0/3 0 50Interface: ATM2/0/3, Type: oc3suniVPI = 0 VCI = 50Status: UPTime-since-last-status-change: 00:29:20Connection-type: PVCCast-type: point-to-pointPacket-discard-option: disabledUsage-Parameter-Control (UPC): passWrr weight: Not-applicableNumber of OAM-configured connections: 0OAM-configuration: disabledOAM-states: Not-applicableCross-connect-interface: ATM0/0/0, Type: t1suniCross-connect-VPI = 0Cross-connect-VCI = 50Cross-connect-UPC: passCross-connect OAM-configuration: disabledCross-connect OAM-state: Not-applicableThreshold Group: 1, Cells queued: 126Rx cells: 6007758, Tx cells: 5866714Tx Clp0:5866714, Tx Clp1: 0Rx Clp0:6007758, Rx Clp1: 0Rx Upc Violations:0, Rx cell drops:140920Rx Clp0 q full drops:140920, Rx Clp1 qthresh drops:0 <-------Cells are dropped HereRx connection-traffic-table-index: 1448Rx service-category: CBR (Constant Bit Rate)Rx pcr-clp01: 1448Rx scr-clp01: noneRx mcr-clp01: noneRx cdvt: 1024 (from default for interface)Rx mbs: noneTx connection-traffic-table-index: 1448Tx service-category: CBR (Constant Bit Rate)Tx pcr-clp01: 1448Tx scr-clp01: noneTx mcr-clp01: noneTx cdvt: noneTx mbs: noneWorkaround: Configure the structured CES circuit to use only 19 or fewer time slots.
•
Symptom: EIGRP does not work on a BVI interface. The BVI interface cannot receive or generate EIGRP hello packets.
Workaround: None.
•
Symptom: A Soft VC or Soft PVC connecting two ATM interfaces in the same Catalyst 8540 MSR might become stuck for no reason.
Workaround: Remove and reapply the Soft VC configuration on the source ATM interface.
•
Symptom: The Catalyst 8540 MSR switch might crash if OAM enabled PVCs are configured between the Catalyst 8540 MSR switch and a Nortel switch.
Note
Workaround: Disable OAM management on the Nortel side of the PVC.
•
Symptom: On an ATM router module (ARM) that is running Cisco IOS Release 12.0(13)W5(19c), the ARM port may become stuck when a backup server is moved to a different switch.
Workaround: Configure the epc port-reload global configuration command for autorecovery on the stuck ARM port.
•
Symptom: If configuring RFC1483 bridging on the ATM router module, when certain RFC 1483 bridging PVC cable connections are unplugged the Catalyst 8540MSR CPU utilization will become very high. This could last a few hours until you shutdown the corresponding bridge-group subinterface on the ATM router module. Only shutting down the physical ATM port will not help.
Workaround: Shutdown the corresponding bridging group subinterface on ATM router module.
•
Symptom: The show ip traffic command displays a constant incrementing counter for "Drop No Route," but, there is a default route learned via OSPF in our routing table. Also, the show cef drop command displays numerous drops for "no route," but the debug ip cef drop command does not display any output.
Workaround: None.
•
Symptom: A point-to-multi point failure might occur if a physical hardware connection does not exist but a virtual software connection is configured. Due to the connection delink failure the following error message appears on the console:
%ATMCORE-3-INTERNAL_ERROR: connDeLinkP2mpPeerLegs: wrong root FSMWorkaround: None.
•
Symptom: A Catalyst 8540 MSR with an IMA interface module connection with an IMA interface connection to a Catalyst 8510 MSR fails if the Catalyst 8510 MSR interface is reset. You see the following error on the Catalyst 8540 MSR:
%IMAPAM-6-LOG: pam 2/0 changed state to FAILED%LINK-5-CHANGED: Interface ATM2/0/ima1, changed state to going downFor example, the show controller atm command for interface 2/0/0 also displays the following:
Cat8540# show controller atm 2/0/0ATM2/0/0 is downPAM State is FAILEDFirmware Version: 3.2FPGA Version : 1.5Boot version : 3.0Workaround: OIR of the carrier module clears the failed state of the IMA interface module and restores service.
•
Symptom: There is a 1 percent loss in ping packets to adjacent node with bidirectional traffic.
Workaround: None.
•
Symptom: The enhanced ATM router module subinterface counters do not get updated.
Workaround: None.
•
Symptom: When reloading the switch the following error message appears after a varying number of SVCs are set up.
00:20:42: %ATMCORE-3-INTERNAL_ERROR: atmCore_connCheckSpot: exhausted vpi/vciWorkaround: None
•
Symptom: PNNI and SSCOP signaling does not come back up on the Catalyst 8540 MSR OC48 interfaces after a powerlessly.
Workaround: None.
•
Symptom: LDP session does not go down when you shut the enhanced ATM router module interface. When tag-switching is configured on a Fast Ethernet interface and the controlling ATM router module is either not configured or shutdown, TDP will come up but the tag packets received on the Fast Ethernet interface will be dropped because there is no controlling ATM router module to process the packets.
Workaround: Make sure the Fast Ethernet interface is linked to an active controlling ATM router module.
•
Symptom: A tag switching traceback might appear when an interface is shut.
Workaround: None.
•
Symptom: When a circuit emulation service (CES) circuit that is carrying traffic is removed, no warning message is shown.
Workaround: None.
•
Symptom: Untagged packets are bridged between 8-port Gigabit Ethernet subinterfaces in the same card. This happens even when they are part of different bridge-groups.
Workaround: Configure a different native VLAN for each 8-port Gigabit Ethernet interface with subinterfaces in the same card.
•
Symptom: The soft-VC permanent virtual connection (PVC) status is indicated as ACTIVE only when the Soft VC is established, and its connection state is UP. If the Soft VC is not connected, the source end of the Soft VC indicates the PVC status as INACTIVE instead of ACTIVE, and the destination end of the Soft VC does not exist, and therefore indicates the PVC status as DELETED.
Workaround: None.
•
Symptom: The Catalyst 8540 MSR needs MIN VPI configuration capability to connect to a PBX VSI. Otherwise, the Catalyst 8540 MSR cannot tear down on a Soft PVC due to VPI/VCIs.
Workaround: None.
•
Symptom: Single bit errors have been observed on CES circuits originating on the same interface module if two other circuits always carry non-spec-conform AAL1 SNP for SN=0. Only this very high number of SNP errors in 1/8 of all AAL1 cells causes the issue. Multiple SNP corrections that could be seen in production networks are handled correctly without data-loss or corruption.
This is an interoperability issue with a Lucent PSAX 2300.
Workaround: None.
•
Symptom: The Catalyst 8540 MSR switch may ignore certain commands from the NVRAM after a reboot. This appears most often on switches with dual route processors.
Workaround: Create a connection traffic index and use that number for the set up of the PVC.
•
Symptom: On the Catalyst 8540 MSR with OC-48 interface module, the PIF3 corrupts the egress cell on the receiving resource management cell.
When receiving a resource management cell (PTI field = 6) the OC-48c interface module sets the payload bytes 1-4 and 44-46 to "FF".
Workaround: Removing any resource management traffic allows normal operation.
•
Symptom: The switch might hang during a manual switch processor "switchover" that follows a route processor switchover.
Workaround: Allow two minutes between manual route processor switchovers and manual switch processor switchovers.
•
Symptom: Signaling packets are dropped between the Catalyst 8540 and its neighboring devices during a route processor switchover. This is caused by automatic status inquiries, from both sides, that occur after a switchover.
Workaround: Perform one of the following to workaround this problem:
–
–
•
Symptom: There is a noise problem with the CES PAM hardware version one.
Workaround: None.
•
Symptom: When the routing table exceeds 50 K routes and if 8-port gigabit ethernet interface modules are installed, a CAM full condition might occurs. This leads to a port-stuck situation.
Workaround: OIR the interface module. The 8-port gigabit ethernet interface module was not designed to handle more than 50 K routes.
•
Symptom: Interoperability with the switch can cause cell loss on the Catalyst 6015 or the Catalyst 6260 if an E1 IMA group consists of either 1 or 3 active links.
Workaround: None.
•
Symptom: If SVC is configured on the ATM router module or the enhanced ATM router module and an OIR is performed on the module, the adjacency available across the SVC remains 'Valid' on the Ethernet port interface or Gigabit Ethernet interface cam. This will cause packets destined to that adjacency to go into a disappear.
Workaround: Before performing an OIR on the ATM router module or the enhanced ATM router module, remove any subinterfaces that have SVC configurations on them.
•
Symptom: After loading the secondary route processor running a release earlier than Cisco IOS Release 12.1(7a)EY (with the primary route processor running Cisco IOS Release 12.1(7a)EY), bringing down the secondary route processor and then reloading it with Cisco IOS Release 12.1(7a)EY might cause the secondary route processor to crash.
The following error messages might appear at the time of the crash:
00:00:24: %SYS-3-BADBLOCK: Bad block pointer A8000158-Traceback= 600C6EFC 600C87E8 60082388 600D0F5C 6039A2EC 6028E9B4 6028DFA0 600D9468 600D9520 600304E8 60030670 600BADE4 600BADD000:00:24: %SYS-6-MTRACE: mallocfree: addr, pcA8000180,60082380 61DC1D7C,600739FC 61DC1AE0,60082398 A8000430,6001FDE8A80002D8,60082380 61DC2018,600738AE 61DC1D7C,60082398 A8000588,6001FD3C00:00:24: %SYS-6-MTRACE: mallocfree: addr, pcA8000430,60082380 61DC22B4,60073760 61DC2018,60082398 A80006E0,6001FC90A8000588,60082380 61DC2550,60073612 61DC22B4,60082398 A8000838,6001FBE400:00:24: %SYS-6-BLKINFO: Corrupted previous pointer blk A8000158, words 152, alloc 600820E8, InUse, dealloc 0, rfcnt 1-Traceback= 600C4934 600C6F14 600C87E8 60082388 600D0F5C 6039A2EC 6028E9B4 6028DFA0 600D9468 600D9520 600304E8 60030670 600BADE4 600BADD000:00:24: %SYS-6-MEMDUMP: 0xA8000158: 0xAB1234CD 0xFFFFFFFE 0x0 0x60D7010800:00:24: %SYS-6-MEMDUMP: 0xA8000168: 0x600820E8 0xA80002B0 0xA8000014 0x8000009800:00:24: %SYS-6-MEMDUMP: 0xA8000178: 0x1 0x0 0x0 0x0
Workaround: Reload the secondary route processor.•
Symptom: When entering the shut down command on the OC-48 interface, the following traceback might be displayed with spurious memory access at snmp_trap:
601BABD8 60AE7404 60AE7448 60AE76B0 60AE7F38 60AEEAAC 60AF0398 60AE0D30 60096D1C 608C4924 608C5344 608C8C00 608E3380 60D06C8C 608E29DCThis does not affect functionality.
Workaround: None.
•
Symptom: The switch router might not send ICMP unreachable messages when an inbound access list is configured and being accessed on an interface module that has an ACL daughter card.
Workaround: None.
•
Symptom: Although packets are being sent out from an ATM OC-12 interface module, the 5-minute output rate shown using the show interface atm command indicates 0 bits per second output. This is not consistent with the output rate shown when the show atm traffic command is entered.
Workaround: None.
•
Symptom: Under certain conditions, the BVI output queue might get stuck when queuing outgoing packets but not dispatching them to hardware. This causes the output queue on the BVI interface to grow constantly.
Workaround: Create another BVI interface and move all the traffic over to the new BVI interface.
•
Symptom: The Catalyst 8540 MSR hangs after OIR of OC-3 multimode interface module.
Workaround: Reload the switch.
•
Symptom: The ifSpeed is reported as 100MB even when the auto-negotiation results in 10MBit.
Workaround: None.
•
Symptom: When auto negotiation is disabled on an enhanced two-port Gigabit Ethernet port, it is up even if the GBIC is not present.
Workaround: None.
•
Symptom: When the module is removed from the switch, the interfaces continue to show as inserted and also the IDB's are not deleted. This is not seen when time is given between the act of insertion and removal of the modules.
Workaround: Wait two minutes after OIR before inserting a new interface module into the switch.
•
Symptom: The switch router might reload after a bus error in handle_new_collect.
Workaround: None.
•
Symptom: When you change the port speed from 10 to 100 on a Catalyst 8540 CSR running 12.0(5)W5(13), enter the shutdown and no shutdown commands for the new setting to take effect.
Workaround: None.
•
Symptom: Fast Ethernet interfaces that are in shutdown mode have packet output and bytes incrementing in the show interface command.
Workaround: Enter the clear counters command.
•
Received malformed TLV.Symptom: The above message is sometimes received. This message can be ignored.
Workaround: None.
•
Symptom: The above messages may appear on the console. They are not reproducible. The message appears on subinterface ATM11/0/0.14 on the ATM router module port.
Workaround: None.
•
Symptom: The command show atm resources displays the number of cells in the UBR queue in an MSC and it increases continuously. The increase happens very slowly.
This occurs when traffic is sent through a Catalyst 8540 MSR at line rate. Increasing the threshold group per VC queue size does not eliminate the problem.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(26)E
Cisco IOS Release 12.1(26)E is a rebuild release for Cisco IOS Release 12.1(23)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(26)E but may be open in previous Cisco IOS releases.
•
Symptom: Unicast reverse path forwarding (uRPF) does not drop packets that are malformed or that have forged (spoofed) addresses of hardware-switched packets on the Catalyst 8540 or Catalyst 8510 switches running Cisco IOS Release 12.1(23)E. For example, if the switch is configured to perform unicast reverse path forwarding using the ip verify unicast source reachable-via rx command or the ip verify unicast reverse-path command and the switch is running Cisco IOS Release 12.1(23)E, those hardware-switched packets might be forwarded.
Workaround: None.
•
A document that describes how the Internet Control Message Protocol (ICMP) could be used to perform a number of Denial of Service (DoS) attacks against the Transmission Control Protocol (TCP) has been made publicly available. This document has been published through the Internet Engineering Task Force (IETF) Internet Draft process, and is entitled "ICMP Attacks Against TCP" (draft-gont-tcpm-icmp-attacks-03.txt).
These attacks, which only affect sessions terminating or originating on a device itself, can be of three types:
1. Attacks that use ICMP "hard" error messages
2. Attacks that use ICMP "fragmentation needed and Don't Fragment (DF) bit set" messages, also known as Path Maximum Transmission Unit Discovery (PMTUD) attacks
3. Attacks that use ICMP "source quench" messagesSuccessful attacks may cause connection resets or reduction of throughput in existing connections, depending on the attack type.
Multiple Cisco products are affected by the attacks described in this Internet draft.
Cisco has made free software available to address these vulnerabilities. In some cases there are workarounds available to mitigate the effects of the vulnerability.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20050412-icmp.shtml.
The disclosure of these vulnerabilities is being coordinated by the National Infrastructure Security Coordination Centre (NISCC), based in the United Kingdom. NISCC is working with multiple vendors whose products are potentially affected. Its posting can be found at: http://www.niscc.gov.uk/niscc/docs/re-20050412-00303.pdf?lang=en.
•
A Cisco device running Cisco IOS and enabled for the Border Gateway Protocol (BGP) is vulnerable to a Denial of Service (DoS) attack from a malformed BGP packet. Only devices with the command `bgp log-neighbor-changes' configured are vulnerable. The BGP protocol is not enabled by default, and must be configured in order to accept traffic from an explicitly defined peer. Unless the malicious traffic appears to be sourced from a configured, trusted peer, it would be difficult to inject a malformed packet.
If a misformed packet is received and queued up on the interface, this bug may also be triggered by other means which are not considered remotely exploitable such as the use of the command `show ip bgp neighbors' or running the command `debug ip bgp <neighbor> updates' for a configured bgp neighbor.
Cisco has made free software available to address this problem.
For more details, please refer to this advisory, available at http://www.cisco.com/warp/public/707/cisco-sa-20050126-bgp.shtml
•
Symptom: The ATM interface on a router connected to a Catalyst 8540 MSR is shut down, but the ATM interface on the Catalyst 8540 MSR continues to stay up. The expected result would be the "down" state. This occurs if a C85MS-8T1-IMA interface module in the Catalyst 8540 MSR is running Cisco IOS Release 12.1(19)E and the ATM interfaces are configured as part of an IMA group.
Workaround: None.
•
A document that describes how the Internet Control Message Protocol (ICMP) could be used to perform a number of Denial of Service (DoS) attacks against the Transmission Control Protocol (TCP) has been made publicly available. This document has been published through the Internet Engineering Task Force (IETF) Internet Draft process, and is entitled "ICMP Attacks Against TCP" (draft-gont-tcpm-icmp-attacks-03.txt).
These attacks, which only affect sessions terminating or originating on a device itself, can be of three types:
1. Attacks that use ICMP "hard" error messages
2. Attacks that use ICMP "fragmentation needed and Don't Fragment (DF) bit set" messages, also known as Path Maximum Transmission Unit Discovery (PMTUD) attacks
3. Attacks that use ICMP "source quench" messagesSuccessful attacks may cause connection resets or reduction of throughput in existing connections, depending on the attack type.
Multiple Cisco products are affected by the attacks described in this Internet draft.
Cisco has made free software available to address these vulnerabilities. In some cases there are workarounds available to mitigate the effects of the vulnerability.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20050412-icmp.shtml.
The disclosure of these vulnerabilities is being coordinated by the National Infrastructure Security Coordination Centre (NISCC), based in the United Kingdom. NISCC is working with multiple vendors whose products are potentially affected. Its posting can be found at: http://www.niscc.gov.uk/niscc/docs/re-20050412-00303.pdf?lang=en.
•
Symptom: Using the show switch fabric command to request information about the different interfaces connected to the chassis does not reflect all of the different interface connections. Only the 155-Mbps interfaces appear in the display.
Workaround: None.
•
Symptom: On the Catalyst 8500 switch, a PNNI PVC link on an NNI interface might flap after the link stabilizes and the ATM PNNI VC component does not come back up.
Workaround: Bounce the PNNI process using the no atm router pnni and atm router pnni configuration commands to disable and enable the PNNI processes.
•
Symptom: Packets are switched out on native VLANs leading to routing by the CPU (with BVI). This occurs on EPIF-based interface modules when untagged packets coming in on the dot1q native VLAN are not processed by the microcode. Instead, they are given to the CPU, and the CPU performs the processing. This causes high CPU utilization if untagged packets are received at a high rate on the native VLAN subinterfaces.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(23)E1
Cisco IOS Release 12.1(23)E1 is a rebuild release for Cisco IOS Release 12.1(23)E. The caveats in this section are resolved in Cisco IOS Release 12.1(23)E1 but may be open in previous Cisco IOS releases.
•
Symptom: TCP connections may be vulnerable to spoofed ICMP packets. A spoofed ICMP packet may cause the TCP connection to use a very low segment size for 10 minutes at a time.
Conditions: This symptom is observed when TCP connections are configured for PMTU discovery. Note that PMTU discovery is disabled by default on a router.
Workaround: Disable PMTU discovery.
Resolved Caveats—Cisco IOS Release 12.1(23)E
Cisco IOS Release 12.1(23)E is a rebuild release for Cisco IOS Release 12.1(22)E. The caveats in this section are resolved in Cisco IOS Release 12.1(23)E but may be open in previous Cisco IOS releases.
•
Symptom: If the switch at one end of a VP tunnel uses Cisco IOS Release 12.0(13)W5(19c) and the switch at the other end of the tunnel uses Cisco IOS Release 12.0(16)W05(21), there could be interoperability issues causing call establishment failures.
Workaround: None.
•
Symptom: A Catalyst 8540 MSR switch DS-3 interface configured with "clock source loop-timed" changes to "admin shut" state if the switch is reloaded due to the following OIR test failure.
"ONLINEDIAG-5-OIR_TEST_FAIL: PHY Loop back test failed"Workaround: Enter the no shutdown command on the DS-3 interface.
•
Symptom: The getmany utility program displays the wrong instances of the "ifStackTable".
Workaround: None.
•
Symptom: The SNMP "get" request on the "ifSpeed" object of the IF-MIB returns the wrong value for the IMA group interface.
There is no other functionality impact.
Workaround: None.
•
Symptom: After a switch processor switchover the input and output cells and the rate counters for IMA, CES, and OC-48 interfaces display incorrect values.
Workaround: None.
•
Symptom: After creating a CES hard PVC on the Catalyst 8540 MSR, if the circuit is deleted the corresponding VPI/VCI on the ATM port and the ATM psuedo port are sometimes not deleted.
If a CES hard PVC with the same ATM VPI/VCI is configured again, configuration fails with the following message.
"%connection creation fails: vpi/vci has been used by other connections"Workaround: None.
•
Symptom: Soft PVC call setup might fail going through third-party ATM switches. These third-party ATM switches might use Extended QoS Information Element (0xEC) with "Acceptable Backward Peak-to-Peak Cell Delay Variation Identifier" without the corresponding cumulative forward and/or backward peak-to-peak CDV.
Workaround: None.
•
Symptom: The 8-port Gigabit Ethernet interface module (C85GE-8X-64K) does not forward broadcasts when spanning is disabled. All connected ports have problems transmitting broadcasts out if the 8-port Gigabit Ethernet interface module:
–
–
–
Workaround: Make the connection on the 8-port Gigabit Ethernet interface module starting with the highest port number. For example, starting with port 8.
•
Symptom: On an interface module with a soft VC configured, if an interface is shut down or if the interface module is removed from the chassis the following cause code appears in the "debug atm sig-all" display of subsequent soft-VC requests:
cause = normal call clearing, location = UserThis cause code is not correct. The "Resource Unavailable class definitions" cause code should appear.
Workaround: None.
•
Symptom: Input errors increment along with the giant counter if jumbo frames are switched through the interface. This behavior is cosmetic and no packets are being dropped.
Workaround: None.
•
Symptom: The Catalyst 8540 MSR might experience a memory leak on Pool Manager and Frame Relay LMI
Workaround: None.
•
Symptom: If the ATM interface on a router connected to a Catalyst 8540 MSR is shut down, the ATM interface on the Catalyst 8540 MSR continues to stay up. The expected result would be the "down" state. This occurs if a C85MS-8T1-IMA interface module in the Catalyst 8540 MSR is running Cisco IOS Release 12.1(19)E and the ATM interfaces are configured as part of an IMA group.
Workaround: None.
•
Symptom: Connectivity problems might occur across STM-4 interfaces after upgrading the switch from Cisco IOS Release 12.0(1a)W5(5b) to 12.0(24)W5(26b). This upgrade might cause all PVCs having a leg on an STM-4 interface to begin experiencing connectivity problems.
Workaround: None.
•
Symptom: Using the show switch fabric command to request information about the different interfaces connected to the chassis does not reflect all of the different interface connections. Only the 155-Mbps interfaces appear in the display.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(22)E
Cisco IOS Release 12.1(22)E is a rebuild release for Cisco IOS Release 12.1(20)E. The caveats in this section are resolved in Cisco IOS Release 12.1(22)E but may be open in previous Cisco IOS releases.
•
Symptom: The wrong payload pattern in idle cells might be transmitted from DS-3 and E3 interfaces.
As per recommendation I.361 (of the ITU-T), the payload pattern of the idle cell is "01101010" or "0x6A". But in DS-3 and E3 interfaces of the Catalyst 8540 MSR, Catalyst 8510 MSR, and LightStream1010 ATM switches, the payload pattern of the idle cell is set to "00000000" or "0x00". This is incorrect as per the ITU-T specification.
Some interoperability issues with other devices might appear if those devices expect only the "0x6A" pattern.
Workaround: None.
•
Symptom: Using the show vc command displays garbage data for TVC(I) type interfaces with multiple leaves. Data for TVC(O) type interfaces with multiple leaves are displayed correctly.
Note
Workaround: Use the show atm vc command instead of the show vc command.
•
Symptom: The input and output counters for the OC-12 interface display a huge value. This problem is seen with traffic running just after bootup or after performing a "clear counters" when traffic is flowing across the link. For example:
4294944186 packets input, 4293742466 bytes, 0 no bufferWorkaround: None.
•
Symptom: The show ncdp sources command always displays "health: unknown" after configuration.
Workaround: None.
•
Symptom: During an upgrade of the Catalyst 8540 MSR from or to the software versions listed in Table 1, an ATM interface with tag switching configured might change to going-down state if an interface flap occurs after the upgrade.
Workaround: Perform route processor switchover.
•
Symptom: Either QSAAL VC or all other well known VCs go down if you revert back to default well known VCs from the manually configured well known VCs.
Workaround: When the you find the well known VCs are missing, use the following two commands again, atm manual-well-known-vc delete and no atm manual-well-known-vc. Then the default well known VCs are reestablished.
•
Symptom: The aal5 buffers command when used with the Catalyst 8540 CSR and Catalyst 8510 CSR switches has no affect and should be a hidden command.
Workaround: None.
•
Symptom: If you configure more than sixteen static MAC entries on the Catalyst 8540 MSR or Catalyst 8540 CSR, the device might generate error messages indicating that it cannot chunk_malloc.
The errors might look similar to the following:
*Dec 9 13:59:28: Can't chunk_malloc*Dec 9 13:59:28: Can't chunk_malloc*Dec 9 13:59:28: Can't chunk_mallocWorkaround: None.
•
Symptom: IP routes through route processor interface (ATM0) are not programmed in EPIF based Ethernet interface modules.
Note
Workaround: None.
•
Symptom: The ATM router module gets stuck after approximately 60 minutes of operation when a particular PVC (IP address, A.B.18.52 atm-vcvc-num broadcast) is configured.
Symptoms include pings to IP address A.B.18.52 fail.
Note
Workaround: Use an IP address that does not have "18.52" as the last two octets for the source IP address for any traffic going to the route processor through the ATM router module.
•
Symptom: A Catalyst 8540MSR with a Gigabit Ethernet interface module, ATM router module, and redundant MSRP might fail its DNS and TFTP functionality after an MSRP switchover occurs. This problem is not seen if the network connection is through the Ethernet ports of an MSRP.
Workaround: Power-cycle the switch if DNS and TFTP are needed.
•
Symptom: Debugs remain enabled for removed interfaces.
1. If you use the debug atm sig-all command, the debugs are enabled for an interface. If that interface module is removed from the chassis, the show debug command output continues to show that "sig-all" debugs are enabled for the removed interface.
2. If you use the deb atm sig-all command, the debugs are enabled for all interfaces and if an interface module is removed from the chassis, the show debug command output continues to show that "sig-all" debugs are enabled for the removed interface module.
This is only a display issue. There is no impact on functionality.
Workaround: None.
•
Symptom: A Catalyst 8540 MSR running Cisco IOS Release 12.1(19)E2 may report a "Spurious Memory Access" when a network management station attempts to do an SNMP get or set of any object in ciscoLS1010ModuleTable. The following error is seen:
Sep 8 02:12:12.615 EST: %ALIGN-3-SPURIOUS: Spurious memory access made at 0x6077D30C reading 0xCSep 8 02:12:12.619 EST: %ALIGN-3-TRACE: -Traceback= 6077D30C 6077BAD0 60779904601D8D54 601DC6D0 601CDDD4 601F167C 600D43F4This problem is only reported when one of the following 4-port OC-12c interface modules is installed in the chassis in the slot0 position.
–
–
This spurious memory access does not affect operation of the Catalyst 8540 MSR.
Workaround: None.
•
Symptom: Building integrated timing supply (BITS) clocking problems.
If the clock source is changed to BITS and the network clock module fails to lock on to the BITS source, the network clock on the Catalyst 8540 MSR switches back to "System" (the internal Stratum 3 NetClk module). Immediately after that change, the primary route processor of the Catalyst 8540 MSR reloads because of a memory alignment error.
Workaround: None.
•
Symptom: Many port interrupts cause the console and route processor to hang.
Under certain circumstances, a Catalyst 8540 MSR console and route processor might hang or "freeze". During this time, the route processor is receiving a high amount of "Port Interrupts" as shown by the show stack command output in the following example:
Switch# show stack<snip/>Interrupt level stacks:Level Called Unused/Size Name1 4 8220/9000 Ehsa Interrupts2 238527938 31272/32768 Port Interrupts^^^^^^^^^3 0 32416/32768 OIR interrupt4 0 9000/9000 PCMCIA Interrupt5 2924484 8456/9000 Console Uart6 0 9000/9000 Error Interrupt7 326062987 8604/9000 NMI Interrupt HandlerUse the hidden command show interrupt, as in the following example, to display where these interrupts are coming from:
Switch# show interruptLevel Intr count----------------------------------------1(ehsa) 42(port) 2369520603(stall) 04(pcmcia) 05(console) 29184066(error) 07(timer) 310363716Number of cougar port interrupts: 236952060Number of Switch Card Interrupts: 230292624Number of Ethernet Interrupts : 3579138Number of Cubi Interrupts : 3137906Number of Nclk Interrupts : 0Number of pam Interrupts : 51Sum of level2 interrupts not equal to isr_countNumber of ints from msc 0 on swc: 0 = 266735Number of ints from msc 1 on swc: 0 = 230080191Number of ints from msc 2 on swc: 0 = 14MSC 0 Non-Zero interrupt counts:NP55 marker int count: 1137NP55 port int count: 267154NP55 cpu port int count: 35NP55 imr disable count: 267189MSC 1 Non-Zero interrupt counts:MSC 2 Non-Zero interrupt counts:NP55 marker int count: 2416NP55 port int count: 230041043NP55 imr disable count: 230041043MSC 3 Non-Zero interrupt counts:MSC 4 Non-Zero interrupt counts:NP55 marker int count: 14MSC 5 Non-Zero interrupt counts:MSC 6 Non-Zero interrupt counts:MSC 7 Non-Zero interrupt counts:In this example, the interrupts are coming from MSC number 2.
Workaround: Use the hidden command show interrupt to determine from which MSC the interrupts are originating. Then force a switchover of the switch processors to use the redundant switch processor as in the following example:
Switch# redundancy preferred-switch-card-slots 6 7Wait for one or two minutes then revert to the default switch processors (slot 5 and 7). This problem is related to one of the switch processors being stuck in a "strange" state as in the following example:
Switch# redundancy preferred-switch-card-slots 5 7•
Symptom: When connecting the IMA interface module from a Catalyst 8540 MSR to a Nortel Passport 15000, the IMA interface does not come up.
Workaround: None.
•
Symptom: Same issue as DDTS CSCdx58649.
Workaround: None.
•
Symptom: On the LightStream 1010 switch, a PNNI PVC link on an NNI interface might flap after the link stabilizes and the ATM PNNI VC component does not come back up.
Workaround: "Bounce" the PNNI process using the no atm router pnni and atm router pnni configuration commands to disable and enable the PNNI processes.
•
Symptom: The switch might drop OAM F5 END loopback cells under its default configuration of OAM intercept (which is enabled by default).
Note
Workaround: Disable OAM intercept to avoid dropping F5 OAM cells.
•
Symptom: When an E1 IMA (inverse multiplexing over ATM) adapter is inserted into a chassis in which several other IMA adapters have already been added, the clock cannot be derived from the network.
Note
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(20)EB
Cisco IOS Release 12.1(20)EB is a rebuild release for Cisco IOS Release 12.1(19)EB. The caveats in this section are resolved in Cisco IOS Release 12.1(20)EB but may be open in previous Cisco IOS releases.
•
Symptom: After route processor switchover, in the following two cases the leaves of a point-to-multipoint soft-VC might change to INACTIVE state:
–
The connections are still up in the hardware and the traffic keeps flowing.
If the destination VPI/VCI are different then no problems are seen.
–
Corresponding to each INACTIVE leaf, a "NOT CONNECTED" soft-VC appears on the outgoing interface for that leaf.
During the INACTIVE state, the ADD_PARTY retries keep occurring.
Workaround: Depending on the case, perform the following workaround:
Case 1—Delete and reconfigure the point-to-multipoint soft PVC or use the shutdown and the no shutdown commands on the source interface.
Case 2—Use the clear atm atm-vc out_going_interface VPI_of_hanging_leg VCI_of_hanging_leg command to clear the hanging leg and the leaf becomes ACTIVE.
•
Symptom: If tag switching is configured and VPI 1 is configured on the ATM interface (for a PVC, for example), the tag switching configuration is lost when the switch reboots.
Workaround: Use the copy startup-config running-config command to copy the startup-configuration to the running-configuration.
•
Symptom: IP routes through route processor interface (ATM0) are not programmed in EPIF based Ethernet interface modules.
Note
Workaround: None.
•
Symptom: The ATM router module gets stuck after approximately 60 minutes of operation when a particular PVC (IP address, A.B.18.52 atm-vc vc-num broadcast) is configured.
Symptoms include pings to IP address A.B.18.52 fail.
Note
Workaround: Use an IP address that does not have "18.52" as the last two octets for the source IP address for any traffic going to the route processor through the ATM router module.
•
Symptom: Many port interrupts cause the console and route processor to hang.
Under certain circumstances, a Catalyst 8540 MSR console and route processor might hang or "freeze". During this time, the route processor is receiving a high amount of "Port Interrupts" as shown by the show stack command output in the following example:
Switch# show stack<snip/>Interrupt level stacks:Level Called Unused/Size Name1 4 8220/9000 Ehsa Interrupts2 238527938 31272/32768 Port Interrupts^^^^^^^^^3 0 32416/32768 OIR interrupt4 0 9000/9000 PCMCIA Interrupt5 2924484 8456/9000 Console Uart6 0 9000/9000 Error Interrupt7 326062987 8604/9000 NMI Interrupt HandlerUse the hidden command show interrupt, as in the following example, to display where these interrupts are coming from:
Switch# show interruptLevel Intr count----------------------------------------1(ehsa) 42(port) 2369520603(stall) 04(pcmcia) 05(console) 29184066(error) 07(timer) 310363716Number of cougar port interrupts: 236952060Number of Switch Card Interrupts: 230292624Number of Ethernet Interrupts : 3579138Number of Cubi Interrupts : 3137906Number of Nclk Interrupts : 0Number of pam Interrupts : 51Sum of level2 interrupts not equal to isr_countNumber of ints from msc 0 on swc: 0 = 266735Number of ints from msc 1 on swc: 0 = 230080191Number of ints from msc 2 on swc: 0 = 14MSC 0 Non-Zero interrupt counts:NP55 marker int count: 1137NP55 port int count: 267154NP55 cpu port int count: 35NP55 imr disable count: 267189MSC 1 Non-Zero interrupt counts:MSC 2 Non-Zero interrupt counts:NP55 marker int count: 2416NP55 port int count: 230041043NP55 imr disable count: 230041043MSC 3 Non-Zero interrupt counts:MSC 4 Non-Zero interrupt counts:NP55 marker int count: 14MSC 5 Non-Zero interrupt counts:MSC 6 Non-Zero interrupt counts:MSC 7 Non-Zero interrupt counts:In this example, the interrupts are coming from MSC number 2.
Note
Workaround: Use the hidden command show interrupt to determine from which MSC the interrupts are originating. Then force a switchover of the switch processors to use the redundant switch processor as in the following example:
Switch# redundancy preferred-switch-card-slots 6 7Wait for one or two minutes then revert to the default switch processors (slot 5 and 7). This problem is related to one of the switch processors being stuck in a "strange" state as in the following example:
Switch# redundancy preferred-switch-card-slots 5 7•
Symptom: If you upgrade the Catalyst 8540 from Cisco IOS Release 12.1(11b)E5 to Cisco IOS Release 12.1(19)E, the "UNI version" configuration is lost from the running configuration.
Workaround: Use the copy startup-config running-config command to replace the running configuration.
•
Symptom: On a Catalyst 8500 MSR or LightStream 1010 switch with a DS3 interface, the interface goes down after only 7ms of AIS and then comes back up. For example, the following message appears:
%LINK-5-CHANGED: Interface ATM1/0/0, changed state to going downThen approximately three seconds later, the next message appears:
%LINK-3-UPDOWN: Interface ATM1/0/0, changed state to up
Note
Workaround: None.
•
Symptom: The switch might drop OAM F5 END loopback cells under its default configuration of OAM intercept (which is enabled by default).
Note
Workaround: Disable OAM intercept to avoid dropping F5 OAM cells.
•
Symptom: When an E1 IMA (inverse multiplexing over ATM) adapter is inserted into a chassis in which several other IMA adapters have already been added, the clock cannot be derived from the network.
Note
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(20)E
Cisco IOS Release 12.1(20)E is a rebuild release for Cisco IOS Release 12.1(19)EB. The caveats in this section are resolved in Cisco IOS Release 12.1(20)EB but may be open in previous Cisco IOS releases.
•
Symptom: Interface up/down events occur even when the carrier delay is configured.
Carrier delay support for DS3 and E3 interfaces on the LightStream 1010 and the Catalyst 8500 MSR is required.
Note
Workaround: None.
•
Symptom: In certain situations the 8-port Gigabit Ethernet interface module input queue shows negative values. This has been seen with a TTL value of 1 only in the lab.
This issue is cosmetic and has no effect on the performance of the Catalyst 8540 or on the 8-port Gigabit Ethernet interface module.
Workaround: None.
•
Symptom: After upgrading to Cisco IOS Release 12.1(19)E, a few of the ATM interfaces might stay in "going down" state as shown using the show interface atm card/subcard/port command.
Workaround: None.
•
Symptom: When upgrading the Catalyst 8540 from Cisco IOS Release 12.1(11b)E5 to Release 12.1(19)E, the "UNI version" configuration is lost from the running configuration.
Workaround: Copy startup-config to running-config.
•
Symptom: The clear counters command used with an OC-48 ATM interface module installed does not reset the FEBE counters in the show controllers command display. See the following example:
Switch# show controllers atm11/0/0...LINEAIS = 0 RDI = 0 FEBE = 2393 BIP(B2) = 0PATHAIS = 0 RDI = 0 FEBE = 228 BIP(B3) = 0LOP = 0Switch# clear countersClear "show interface" counters on all interfaces [confirm] YSwitch# show controllers atm11/0/0...LINEAIS = 0 RDI = 0 FEBE = 2393 BIP(B2) = 0PATHAIS = 0 RDI = 0 FEBE = 228 BIP(B3) = 0LOP = 0Workaround: Reload the Cisco IOS image on the switch and do not OIR the OC-48 interface module.
•
Symptom: When establishing soft-VP between switches, if a VP number is used by the previous switch and if that switch sends an error stating that "VPI" is already used, then the switch receiving that error should use a different VPI number during its next retry attempt. For example:
Switch-1 (1/0/0)-------(1/0/0) Switch-2 (1/1/1)---------(1/1/1) Switch-3 (1/1/0)1. Assume a soft-VP call is getting initiated by Switch-1.
2. Assume that the node ID of Switch-3 is higher than Switch-2.
3. Assume that the Switch-2 ATM1/1/1 interface has a PVP configured with VPI 4 (ATM PVP 4).
In this scenario, when a call is initiated by Switch-1, it is forwarded by Switch-2 to Switch-3. Because the node ID of Switch-3 is higher than Switch-2, Switch-3 requests a VP number. If the request is for VPI 4 on Switch-3's ATM1/1/1, it will reach Switch-2's ATM1/1/1. Because there was a PVP configured on Switch-2's ATM1/1/1 interface with PVP 4, it is rejected stating that VPI is "unavailable". This error is sent to both Switch-1 and Switch-3. So, during the next attempt when Switch-3 requests a VPI number, it should request a different VPI number from 4.
Workaround: None.
•
Symptom: Problems occur in CISCO-ATM-IF-PHYS-MIB implementation:
1. ciscoAtmIfPhysLineFebeErrors returns Path FEBE errors instead of Line FEBE errors.
2. ciscoAtmIfPhysPathFebeErrors returns Line FEBE errors instead of Path FEBE errors.
This only happens with OC-3/STM-1 or OC-12/STM-4 interfaces.
Workaround: Poll ciscoAtmIfPhysLineFebeErrors when you need Path FEBE errors and poll ciscoAtmIfPhysPathFebeErrors when you need Line FEBE errors.
•
Symptom: On a Catalyst 8500 MSR or LightStream 1010 switch with a DS3 interface, the interface goes down after only 7ms of AIS and then comes back up. For example, the following message appears:
%LINK-5-CHANGED: Interface ATM1/0/0, changed state to going downThen approximately three seconds later, the next message appears:
%LINK-3-UPDOWN: Interface ATM1/0/0, changed state to up
Note
Workaround: None.
•
Symptom: The 64-bit ifHCInOctets and ifHCOutOctets counters, queried using UCD snmpget routine returns invalid values. The difference between the two values, retrieved in succession (10 seconds, 20 seconds, 30 seconds periods) is invalid for the speed of interface and known traffic on that interface. Often the difference is very close to the 32-bit maximum value, indicating a memory or network-byte ordering problem. These same symptoms have been noted on the show interface command output as well.
Workaround: None.
•
Symptom: Load balancing between provider edge (PE) to provider (P) switches and provider edge to provider edge does not work. If MPLS is enabled on the switch router running Release 12.1 software, MPLS might not re-resolve all dependant routes for a path if you enter the shut/noshut or the clear ip route commands on the outgoing interface of the next hop router.
Workaround: None.
•
Symptom: Configuring different encapsulations on 2-port Gigabit Ethernet subinterfaces clears the existing active configuration when multiple subinterfaces are present on a main interface.
Workaround: All subinterfaces have to be configured with the same encapsulation (either 802.1 Q or Inter-Switch Link [ISL]).
•
Symptom: The system image might crash while configuring VC bundling. This might be related to a memory cleanup process and configuring or unconfiguring VC bundles quickly might increase its likelihood.
Workaround: None.
•
Symptom: A switch router with an 8-port Gigabit Ethernet interface module might forward duplicate broadcasts when bridging is configured on a trunk.
Workaround: None.
•
Symptom: The Internetwork Packet Exchange (IPX) statistics sometimes fluctuate when no traffic is flowing through the interface. The only affected functionality is IPX statistics.
Workaround: None.
•
Symptom: If SVC is configured on the ATM router module or the Enhanced ATM router module and an OIR is performed on the module, the adjacency available across the SVC will remain 'Valid' on the Ethernet port interface or Gigabit Processor interface cam. This will cause packets destined to that adjacency to go into a disappear.
Workaround: Before performing an OIR on the ATM router module or the Enhanced ATM router module, remove any subinterfaces that have SVC configurations on them.
•
Symptom: After entering the shutdown command and the no shutdown command on an ATM router module interface with more than 32 CLIP clients the following error message might display on the console:
00:58:16: %SYS-3-MGDTIMER: Uninitialized timer, timer stop, timer = 644B0550.-Process= "ATM ARP INPUT", ipl= 0, pid= 30-Traceback= 600CE1BC 600CF2A4 60792B0C 60796850 60797744 600BBA64 600BBA50shutEnter hex value: 600CE1BC 600CF2A4 60792B0C 60796850 60797744 600BBA64 600BBA500x600CE1BC:mgd_timer_complain_uninit(0x600ce190)+0x2c0x600CF2A4:mgd_timer_stop(0x600cf26c)+0x380x60792B0C:atm_arp_stampvc(0x60792ad4)+0x380x60796850:atm_inarp_rcvdreq(0x6079620c)+0x6440x60797744:atm_arp_process(0x60796f30)+0x8140x600BBA64:r4k_process_dispatch(0x600bba50)+0x140x600BBA50:r4k_process_dispatch(0x600bba50)+0x0
This does not impact functionality.Workaround: After waiting a few minutes, enter the shutdown command and the no shutdown command.
•
Symptom: Spurious memory access at pim_igmp_new_dr_querier might be seen when running PIM-SM with multicast traffic on ATM router module interfaces. This happens when DR changes are seen. This does not impact functionality.
Workaround: None.
•
Symptom: When two Catalyst 8540 CSRs with 2-port Gigabit Ethernet interface modules are connected back-to-back, packets are lost. On one side the counters might show input and output increasing, but the other side shows only output packets increasing. The Address Resolution Protocol (ARP) from the other side is lost, and the ability to ping the other side is lost.
Workaround: Reset the Gigabit Ethernet interface module.
•
Symptom: The prompt on the secondary route processor does not reflect the hostname of the switch router after a sync.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(19)EB
Cisco IOS Release 12.1(19)EB is a rebuild release for Cisco IOS Release 12.1(14)EB1. The caveats in this section are resolved in Cisco IOS Release 12.1(19)EB but may be open in previous Cisco IOS releases.
•
Symptom: The 64-bit ifHCInOctets and ifHCOutOctets counters, queried using UCD snmpget routine returns invalid values. The difference between the two values, retrieved in succession (10 seconds, 20 seconds, 30 seconds periods) is invalid for the speed of interface and known traffic on that interface. Often the difference is very close to the 32-bit maximum value, indicating a memory or network-byte ordering problem. These same symptoms have been noted on the show interface command output as well.
Workaround: None.
•
Symptom: Load balancing between provider edge (PE) to provider (P) switches and provider edge to provider edge does not work. If MPLS is enabled on the switch router running Release 12.1 software, MPLS might not re-resolve all dependant routes for a path if you enter the shut/noshut or the clear ip route commands on the outgoing interface of the next hop router.
Workaround: None.
•
Symptom: The debug sscop packet command output includes an I or an O to indicate if the packet was input or output on the interface. The output of the debug sscop packet command incorrectly indicates the direction of an inbound POLL PDU as outbound. This is a cosmetic issue affecting ATM routers and ATM switch routers in all versions of Cisco IOS.
Switch# debug sscop packetSSCOP Packets debugging is onSwitch#*Jan 18 17:45:38.306: SSCOP(O - ATM1/0/0): 00 01 3E DC 0A 00 00 06*Jan 18 17:45:38.306: SSCOP(O - ATM1/0/0): 00 01 3E DC 00 00 00 24 0B 00 00 06*Jan 18 17:45:39.178: SSCOP(O - ATM1/0/0): 00 01 3E DF 0A 00 00 06*Jan 18 17:45:39.178: SSCOP(I - ATM1/0/0): 00 01 3E DF 00 00 00 24 0B 00 00 06*Jan 18 17:45:47.786: SSCOP(O - ATM1/0/0): 00 01 3E DD 0A 00 00 06*Jan 18 17:45:47.786: SSCOP(O - ATM1/0/0): 00 01 3E DD 00 00 00 24 0B 00 00 06*Jan 18 17:45:48.898: SSCOP(O - ATM1/0/0): 00 01 3E E0 0A 00 00 06*Jan 18 17:45:48.898: SSCOP(I - ATM1/0/0): 00 01 3E E0 00 00 00 24 0B 00 00 06*Jan 18 17:45:57.186: SSCOP(O - ATM1/0/0): 00 01 3E DE 0A 00 00 06*Jan 18 17:45:57.186: SSCOP(O - ATM1/0/0): 00 01 3E DE 00 00 00 24 0B 00 00 06*Jan 18 17:45:58.162: SSCOP(O - ATM1/0/0): 00 01 3E E1 0A 00 00 06*Jan 18 17:45:58.162: SSCOP(I - ATM1/0/0): 00 01 3E E1 00 00 00 24 0B 00 00 06 und allAll possible debugging has been turned off.
Workaround: None.
•
Symptom: Spurious memory access might occur at lss_arm_atm_adjacency_mac(). This happens after the following warning message:
%TBRIDGE-4-VCPAKDROP: Subscriber SA 0000.0c01.50b0 moved or loopbetween circuits 600 and 200
Workaround: None.•
Symptom: After loading the secondary route processor running a release earlier than Cisco IOS Release 12.1(7a)EY (with the primary route processor running Cisco IOS Release 12.1(7a)EY), bringing down the secondary route processor and then reloading it with Cisco IOS Release 12.1(7a)EY might cause the secondary route processor to crash.
The following error messages might appear at the time of the crash:
00:00:24: %SYS-3-BADBLOCK: Bad block pointer A8000158-Traceback= 600C6EFC 600C87E8 60082388 600D0F5C 6039A2EC 6028E9B4 6028DFA0 600D9468 600D9520 600304E8 60030670 600BADE4 600BADD000:00:24: %SYS-6-MTRACE: mallocfree: addr, pcA8000180,60082380 61DC1D7C,600739FC 61DC1AE0,60082398 A8000430,6001FDE8A80002D8,60082380 61DC2018,600738AE 61DC1D7C,60082398 A8000588,6001FD3C00:00:24: %SYS-6-MTRACE: mallocfree: addr, pcA8000430,60082380 61DC22B4,60073760 61DC2018,60082398 A80006E0,6001FC90A8000588,60082380 61DC2550,60073612 61DC22B4,60082398 A8000838,6001FBE400:00:24: %SYS-6-BLKINFO: Corrupted previous pointer blk A8000158, words 152, alloc 600820E8, InUse, dealloc 0, rfcnt 1-Traceback= 600C4934 600C6F14 600C87E8 60082388 600D0F5C 6039A2EC 6028E9B4 6028DFA0 600D9468 600D9520 600304E8 60030670 600BADE4 600BADD000:00:24: %SYS-6-MEMDUMP: 0xA8000158: 0xAB1234CD 0xFFFFFFFE 0x0 0x60D7010800:00:24: %SYS-6-MEMDUMP: 0xA8000168: 0x600820E8 0xA80002B0 0xA8000014 0x8000009800:00:24: %SYS-6-MEMDUMP: 0xA8000178: 0x1 0x0 0x0 0x0
Workaround: Reload the secondary route processor.•
Symptom: While sending traffic with super CAM (for example, the OC48 combo super CAM or OC12 super CAM) and enabling the online diags, then performing an OIR of the super CAM. If you use shutdown and no shutdown commands on one of the interfaces you might see the following error message:
ALIGN-3-READEXCEPTION messages.Workaround: None.
•
Symptom: The atm arp-server nsap command is removed from the running configuration upon reloading the switch.
Workaround: The copy startup-config running-config command will restore the original configuration.
Resolved Caveats—Cisco IOS Release 12.1(19)E2
Cisco IOS Release 12.1(19)E2 is a rebuild release for Cisco IOS Release 12.1(19)E. The caveats in this section are resolved in Cisco IOS Release 12.1(19)E2 but may be open in previous Cisco IOS releases.
•
Symptom: If your switch is running Cisco IOS Release 12.1(7a)EY and you perform an OIR of a CES interface module, this might cause the route processor to crash.
Workaround: None.
•
Symptom: A corrupted HW-CEF table on the Gigabit interface module causes punts (redirects) and CPU usage.
A Catalyst 8540 MSR switches running 12.1(14)E1 might experience high CPU usage due to IP input.
This is caused by corrupted CEF entries in the interface module HW forwarding table that cause the packets to be punted (redirected) to the main route processor CPU.
Workaround: None. For a temporary workaround, use the clear adjacency command to clear the HW -CEF table but, eventually the CPU returns to normal.
•
Symptom: Whenever you attempt to update the Access Control List (ACL) attached to the management interface (Ethernet2/0/0) on the LightStream 1010 ATM switch the following error message appears:
ACL is not supportedACL functionality on the interface is not affected.
Workaround: None.
•
Symptom: The "ifindex persistent" feature support was addedbeginning with Cisco IOS Release 12.1(13)E.
The SNMP ifIndex persistence feature provides an interface index (ifIndex) value that is retained and used when the switch reboots. The ifIndex value is a unique identifying number associated with a physical or logical interface.
For additional information reference, Configuring SNMP IfIndex Persistence.
Workaround: None.
•
Symptom: After a switch processor or route processor switchover, incorrect values appear in the show interface atm display counters for the OC-3 interface module.
Workaround: Reload the switch.
•
Symptom: The ATM router module interface might indicate outgoing packets when the interface is administratively down. This does not impact the functionality.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(14)EB1
Cisco IOS Release 12.1(14)EB1 is a rebuild release for Cisco IOS Release 12.1(14)EB. The caveats in this section are resolved in Cisco IOS Release 12.1(14)EB1 but may be open in previous Cisco IOS releases.
•
Symptom: When you boot the Catalyst 8540 MSR switch the following error messages pertaining to ONLINE diag failure appears.
%ONLINEDIAG-5-OIR_TEST_FAIL:This happens because an E1 RJ-45 interface module is installed in the chassis. This might cause the E1 RJ-45 interfaces to change state to Admin down.
Workaround: Use the no shutdown command on the interface that changed state to Admin down.
•
Symptom: There is a connectivity problem over bridged PVCs on an enhanced ATM router module. This is a configuration issue.
TCAM Layer 2 switching region space on the enhanced ATM router module interface (set to default 1024 entries) becomes exhausted. This causes local learning to not occur on the enhanced ATM router module interface which, in turn, creates the connectivity problem.
Workaround: Increase the size of the Layer 2 switching region on the TCAM using the global configuration command sdm size l2-switching #_of_entries. Save the new configuration to NVRAM and reboot the switch (the configuration takes effect after reboot).
•
Symptom: On the Catalyst 8540 CSR the Underruns counter on the egress interface increments. This occurs because the Catalyst 8540 CSR is a cut through switch and it starts forwarding once a predefined number of bytes are received by the ingress interface. However, if any input errors occur (for example, bad CRC, alignment errors, or giant packet errors) after the switch starts forwarding, the switch would have already started forwarding the frame and will register an underrun error on the egress port.
To determine the cause of underruns on a particular interface, the counters on all other Fast Ethernet or Gigabit Ethernet interfaces need to be checked for input errors and eliminated.
Note
Workaround: None.
•
Symptom: After upgrading to Cisco IOS Release 12.1(13)E1, the Catalyst 8540 MSR might displaying the following error message continuously:
Feb 5 12:53:43: Modifying Unallocated EntryFeb 5 12:53:44: Modifying Unallocated EntryFeb 5 12:53:44: Modifying Unallocated EntryThis might cause connectivity problems.
Workaround: None.
•
Symptom: The Catalyst 8540 MSR might display following error messages:
xpif_read_media_type: Null ds structure for Port-channel1xpif_read_media_type: Null ds structure for Port-channel1Workaround: None.
•
Symptom: The Catalyst 8540MSR and LightStream 1010 console hangs when using the hw-module {slot number | subslot subslot/subcard} reset command on the Frame Relay E1 interface nodule.
Workaround: Shutdown all controllers in the module before using the hw-module {slot number | subslot subslot/subcard} reset command.
•
Symptom: When you reload a Catalyst 8540 MSR with a redundant route processor, the secondary route processor will become the primary route processor after the reload.
The uptime shown when you enter the show version command is the uptime of the boot image. This should be the uptime of the running IOS software.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(19)E
Cisco IOS Release 12.1(19)E is a rebuild release for Cisco IOS Release 12.1(14)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(19)E but may be open in previous Cisco IOS releases.
•
Symptom: The Frame Relay retry-interval is shown twice in the running configuration. This means the configured Frame Relay Soft VCs are not preserved after an image reload or a route processor switchover.
Workaround: None.
•
Symptom: PNNI routing fails from one access device to a complex-peer in some branches. For example, in the following network example, calls from the Ls1010 switch to an interface on the Catalyst 8540 switch are not set up.
But, calls from the Catalyst 8540 switch to the Ls1010 switch are set up correctly. If the Catalyst 8540.2 switch is configured as a simple single-level hierarchy the calls originated from either switch to the other switch are set up without any problems.
Workaround: Configure the complex node (Catalyst 8540.2 switch) as a simple hierarchy.
•
Symptom: The switch crashes when you run the no tag-switching mtu command.
Workaround: None.
•
Symptom: The following issues could be seen on the simultaneous execution of certain PNNI CLIs.
The switch reloads when:
–
–
–
–
The following issues are also seen:
–
–
–
–
Workaround: None.
•
Symptom: On a Catalyst 8540 MSR, if an OC-3 interface is overbooked and configured with PVCs that consume more than 599 Mbps of bandwidth, during bootup these PVCs are lost from the running configuration.
The following error message appears during bootup for every PVC that is rejected.
%connection creation fails: specified traffic/resource parameters are invalidWorkaround: Configure the PVCs that were lost once the switch boots up.
•
Symptom: After a Route Processor switchover, no traffic flows through structured CES circuits (Hard PVCs, Soft PVCs and SVCs). This does not occur on unstructured CES circuits.
Workaround: Use the shutdown and no shutdown commands on the CES interface and traffic starts flowing again.
•
Symptom: The marking feature does not work on the Enhanced Gigabit Ethernet module if the outgoing interface is TC-ATM. It works if the outgoing VC is a normal 1483 PVC.
Workaround: None.
•
Symptom: Subinterfaces of route processor and ATM router module ports are not registered with SNMP ifTable.
Workaround: None.
•
Symptom: The ATM router module might have a connectivity problem for bridged traffic. The connectivity is broken for five minutes when the soft VC between the ATM router modules are rerouted using another path.
Workaround: Run the clear bridge command on the Catalyst 8540 switch that is not switching traffic.
•
Symptom: Soft PVC does not come up.
If a Soft PVC originates from a Catalyst 8540 MSR switch directly connected to an Alcatel switch, the Soft PVC never comes up. If the call originated from the Alcatel side, the Soft PVC comes up with no problem.
Note
For example:
At UNI ATM interface:
atm soft-vc 1 455 dest-add xx.348f.8001.bbfc.0000.0ba0.6f10.4080.210a.2a01.00 1 455 rx-cttr 1 tx-cttr 1The Alcatel switch is directly connected, ATM NNI and PNNI are up, and adjacency is up:
show atm pnni neighbor---Neighbor Name: unknown, Node number: 40Neighbor Node Id: 88:160:xx.348F8001BBFC00000BA06F10.408021810A00.00Neighboring Peer State: FullLink Selection For CBR : minimize blocking of future callsLink Selection For VBR-RT : minimize blocking of future callsLink Selection For VBR-NRT: minimize blocking of future callsLink Selection For ABR : balance loadLink Selection For UBR : balance loadPort Remote Port Id Hello stateATM12/0/3 1281000 2way_in (Flood Port)---Switch# show atm routeP I 40 0 UP 0 39.348f.8001.bbfc.0000.0ba0.6f10/104---But, if the show atm pnni dtl node 40 ubr command is used:
*Apr 10 23:11:21.218: PNNI: UBR route request from ATM_OWNER_UNKNOWN*Apr 10 23:11:21.218: PNNI: To node 88:160:xx.348F8001BBFC00000BA06F10.408021810A00.00*Apr 10 23:11:21.218: PNNI: selected target node 40*Apr 10 23:11:21.218: priority: 2 (40 0) pnni-remote-internal*Apr 10 23:11:21.218: PNNI: Compute On-Demand Route Based On Admin Weight*Apr 10 23:11:21.218: PNNI: Failed To Find An On-Demand Route,Code: PNNI_RESOURCE_UNAVAILABLE*Apr 10 23:11:21.218: PNNI: No Route To88:160:xx.348F8001BBFC00000BA06F10.408021810A00.00*Apr 10 23:11:21.218: PNNI: Send Source Route ReplyTo Requestor: Code PNNI_RESOURCE_UNAVAILABLE---When the call is initiated from the LightSteam 1010 switch, nothing comes to the Alcatel switch. At the same time, when the call is initiated from the Alcatel switch, the Soft PVC comes up:
Switch# show atm vc interface atm 9/1/1Interface VPI VCI Type X-Interface X-VPI X-VCI Encap StatusATM9/1/1 0 5 PVC ATM13/0/0 0 72 QSAAL UPATM9/1/1 0 16 PVC ATM13/0/0 0 40 ILMI UP>>Alcatel Init <<<<ATM9/1/1 0 400 SoftVC ATM12/0/3 0 227 UP>> LS1010 Init <<<<ATM9/1/1 1 455 SoftVC NOT CONNECTEDWorkaround: None.
•
Symptom: The loopback diagnostic does not work with a 4-port OC -12 interface module in the Catalyst 8540 MSR.
This occurs after running the shutdown and no shutdown commands on the interface.
Workaround: Remove and reapply loopback diagnostic.
•
Symptom: Under certain circumstances, when an OC-12 interface flaps up-and-down on a Catalyst 8540 MSR running Cisco IOS Release 12.1(12)E or EY software, the switch fabric might start dropping cells on VCs exiting this interface.
Workaround: Try reloading the switch image.
•
Symptom: On the Catalyst 8540 MSR, after powering off and turning the power back on, the PVC configuration disappeared from running configuration but still exits in startup config.
Workaround: Either reconfigure the PVC manually or copy startup configuration to running configuration.
•
Symptom: Ethernet interface modules fail the Online Access Test after upgrading from Cisco IOS Release 12.0.5.W5.13 to Cisco IOS Release 12.0.22.W5.25.
Workaround: This access test failure is caused by the incompatibility of hardware version (4.1) and firmware version (2.1) of the Ethernet interface modules installed in the switch. Upgrade to the latest firmware version (3.3) to resolve this problem.
•
Symptom:
The Catalyst 8540 switch running Cisco IOS Release 12.0(18)W5(22) might experience a memory leak in pool manager.
The system logs may indicate tracebacks similar to the following:
%SYS-2-MALLOCFAIL: Memory allocation of 496 bytes failed from 0x6006CA58, poolProcessor, alignment 0 -Process= "Pool Manager", ipl= 6, pid= 4-Traceback= 6009E434 6009FEE0 6006CA60 600AA200 6009968C 60099678For further information on memory leaks, refer to the Troubleshooting Memory Problems publication at http://www.cisco.com/warp/public/63/mallocfail.shtml#subfour
Workaround: None.
•
Symptom: An ATM router module interface with a PVC configured might experience a port stuck condition.
Workaround: None.
•
Symptom: On the CES module interface, the status reads UP/UP and the status of the PVC to that interface is UP in even though there is no physical connection at that port.
Workaround: Perform shutdown and no shutdown on the interface to clear the erroneous status.
•
Symptom: Cell loss on PVC connections might occur if you remove the active route processor module without using the redundancy prepare-for-cpu-removal command prior to removing the module. Using the redundancy prepare-for-cpu-removal command prevents any cell loss outage from affecting the established PVCs during removal.
Workaround: None.
•
Symptom: In the Catalyst 8540 switch, the output of a show ethernet command has lost the carrier count following the collision counter. This is a counter error only and does not actually indicate a lost carrier and does not impact the operation of the Ethernet interface.
Workaround: None.
•
Symptom: On the Catalyst 8540 MSR switch the following object MIBs do not work as expected:
CISCO-ATM-CONN-MIB, CISCO-RHINO-MIB, OLD-CISCO-SYS-MIB, CISCO-ATM-RM-MIB, CISCO-ATM-IF-PHYS-MIB, CISCO-OAM-MIBWorkaround: None.
•
Symptom: When you try to connect Racal-Datacomm CSUs together using ports on the same CES interface module, timing problems might prevent the CSUs from starting up.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(14)EB
Cisco IOS Release 12.1(14)EB is a rebuild release for Cisco IOS Release 12.1(14)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(14)EB but may be open in previous Cisco IOS releases.
•
Symptom: If a Frame Relay-Frame Relay PVC is deleted through SNMP Resource Management the information is incorrect in the show frame-relay interface resource serial card/subcard/port:chn_grp_# command display.
This allows for creation of more VCs than the physical bandwidth available.
Workaround: Delete and recreate the channel group.
Resolved Caveats—Cisco IOS Release 12.1(14)E1
Cisco IOS Release 12.1(14)E1 is a rebuild release for Cisco IOS Release 12.1(13)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(14)E1 but may be open in previous Cisco IOS releases.
•
Symptom: Input errors on interface ATM0 increment one per minute. Only Input errors are incrementing and no other counters (For example, CRC and other errors).
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(13)EB1
Cisco IOS Release 12.1(13)EB1 is a rebuild release for Cisco IOS Release 12.1(13)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(13)EB1 but may be open in previous Cisco IOS releases.
•
Symptom: Primary Route Processor crashes on copying a diag-image from TFTP to PCMCIA flashcard.
Workaround: Configured "no file verify auto" to avoid the automatic IOS image verification. This will prevent the crash.
•
Symptom: When you configure HSRP on a GEC subinterface to track a parallel link which already has a standby group, a memory leak occurs and HSRP starts changing status up and down continuously.
Workaround: None.
•
Symptom: The show diag power-on command displays incorrect slot number information for slots 9 through 12.
Workaround: None.
•
Symptom: If the BITS port in T1 mode is the primary clock and it changes to "unlockable", the clock switches to the next priority clock but does not revert back to the BITS port even when it is reconfigured correctly.
Workaround: Reconfigure the BITS clock source as in the following example:
switch(config)# no network-clock-select 1 BITS 0switch(config)# network-clock-select 1 BITS 0•
Symptom: The CBR interface module sends a "remote alarm" to the peer interface if the line flaps continuously. This causes the peer interface to change to "yellow alarm" state.
Workaround: Enter the shutdown and no shutdown commands on the CBR interface.
•
Symptom: Packets received with a destination multicast MAC address and destination unicast IP address are forwarded to the Route Processor and then forwarded back out on the VLAN where the packets were received.
Workaround: None.
•
Symptom: When you downgrade a Catalyst 8540 image from an image that supports 128K ITT entries per MSC module to an image that supports only 32K ITT entries per MSC module, and more than 32K of ITT space was used before the downgrade, the results are undefined. The values displayed for free and used ITT space are incorrect. This is caused by a mismatch in capabilities of the primary and secondary images. Such a downgrade is not recommended.
Workaround: None.
•
Symptom: When issuing privilege interface level 10 atm .... commands and a further subset of these commands, the prompt does not return for an extended period of time and virtual exec CPU utilization goes very high.
Workaround: None.
•
Symptom: A Route Processor failover on Catalyst 8540 MSR might cause a DS3 interface with scrambling enabled to disrupt traffic passing through it. This might result in CRC errors being seen on the edge device.
Workaround: Enter the shutdown and no shutdown commands on the DS3 interface of the Catalyst 8540 MSR.
•
Symptom: When the Catalyst 8540 receives a gratitude ARP that updates the ARP table with a new MAC address the ingress port TCAM still points to an old adjacency. Once the IP address is pinged from the Route Processor, the ingress port TCAM adjacency is updated.
Workaround: Use the clear lane le-arp card/subcard/port command to update the ingress port TCAM adjacency.
•
Symptom: A Cisco 8540 running IOS Release 12.1(12c)E1 may have packets for the default route routed on the Route Processor instead of the interface modules if default-network is configured. This causes an increased CPU utilization in the IP Input process.
This is caused by a switching table in the interface module not being cleared when the ip default-network command is not configured. This leads to traffic being switched or dropped incorrectly.
Workaround: The ip route 0.0.0.0 0.0.0.0 next-hop can be used instead of the ip default-network command. The Catalyst 8540 must be reloaded with no ip default-network configuration to clear the redundant entries in the interface module switching table.
•
Symptom: On a Catalyst 8540 MSR switch, in the show switch fabric command display, the Rx Cells and Tx Cells counters remain at 0 under MSC3 even though traffic is flowing through the ports linked to that MSC.
Note
Workaround: None.
•
Symptom: After a LAN emulation client (LEC) goes down, a router that has an enhanced ATM port adapter (PA-A3) may not be able to register a network service access point (NSAP) address with an Integrated Local Management Interface (ILMI). Several messages that are similar to the following may be logged in the log file:
%LANE-3-NOREGILMI: ATM1/1/0.7 LEC cannot register39.000000000000000000000022.000163728428.07 with ILMI%LANE-3-NOREGILMI: ATM1/1/0.6 LEC cannot register39.000000000000000000000022.000163728428.06 with ILMIThis symptom is observed on a Cisco router that has a PA-A3 port adapter and that is running Cisco IOS Release 12.2(10a). This symptom is observed after the software is upgraded from Cisco IOS Release 12.2(8)T4.
Workaround: Reload the router.
•
Symptom: While pinging (sweep range) across a Soft PVC between two ARM (I) interface modules, cell drops occur at a rate of less than one percent.
Workaround: None.
•
Symptom: Cell drops are observed for OC12 interface modules with a Hierarchical VP tunnel.
The problem occurs if a Hierarchical VP tunnel is defined on interface in slot 0 and the primary Route Processor is in slot 8 and the secondary Route Processor has come up in slot 4.
The same problem can be seen if there is a Hierarchical VP tunnel defined on interface in slot 9 and the primary is Route Processor is in slot 4 and the secondary Route Processor has come up in slot 8
This occurs when the secondary Route Processor boots up and there is a Hierarchical VP tunnel on any of the interfaces that share the same MSC as the secondary RP. There are no issues if the Redundant Route Processor is never booted.
Workaround: For OC12 interface modules with Hierarchical VP tunnel use slot 2 or slot 11.
•
Symptom: The K1 CPU on the 8-port Gigabit Ethernet interface module might reset itself, when the following message appears on the console:
*Aug 17 14:05:56.214: %SYS-2-GETBUF: Bad getbuffer, bytes= 18288-Process= "SK-IPC Input", ipl= 0, pid= 2-Traceback= 6006D214 6006F078 6044A3CC 6044A648 6009968C 60099678CDP and other well-known multicast-MAC traffic received at rates greater than 700 Mbps by the 8-port Gigabit Ethernet interface module cause the K1 CPU to run out of memory. The K1 CPU might also malfunction by concatenating multiple packets causing getbuffer to fail in IOS.
Workaround: None.
•
Symptom: After upgrading the switch image, the LightStream 1010 switch might lose PNNI neighbors. When the "Auto-Config" function becomes stuck, it causes PNNI neighbor loss and prevents soft-PVCs from connecting.
Note
Workaround: Use the shutdown and no shutdown commands on the interfaces or reboot the switch to recover.
•
Symptom: The IMA interface module does not generate a syslog message when the link goes down.
Workaround: None.
•
Symptom: When using version UNI 4.0, during SVC call set ups, a Catalyst 8500 MSR switch in the "Call Received" state does not send a connect-ack upon reception of a connect. This causes the SVC to be released.
Workaround: Use version UNI 3.1
•
Symptom: A Catalyst 8500 MSR running Cisco IOS version 12.0(20)W5(24b) might crash with a bus error in the SNMP code.
Workaround: Disabling polling chassis MIBs might prevent the crash.
•
Symptom: If you disconnect the transmit cable from an E1 CBR interface on a Catalyst 8540, Catalyst 8510, or LightStream 1010 switch, the interface still appears up/up. No indication of error messages appear in the log files.
When the transmit cable is reconnected, it starts to transmit traffic again. However, if the transmit cable is disconnected multiple times, the link does not pass traffic even when the transmit cable is reconnected.
Workaround: Use the shutdown and no shutdown commands on the interface and it returns to normal, or disconnect the receive cable and reconnect it.
•
Symptom: On a Cisco 6400 Node Switch Processor (NSP) running Cisco IOS Release 12.2(2)B and its derivatives, using the atm ping command to test a valid and active VP might fail with the following message:
Jun 20 15:56:17.742: %ATMCORE-3-INTERNAL_ERROR: atmCore_oamRcvPing: No match.The failure occurs when running the ping towards specific destination ATM switches, and is dependent on the format of the loopback cell received from the destination ATM switches. The occurrence of this problem is difficult to predict and depends on the particular ATM switch implementation of the destination switch.
Workaround: None.
•
Symptom: A Cisco 6400 NSP, running Cisco IOS Release 12.1(5)DB or 12.2(2)B, displays the following error message when you use the atm ping interface atm slot/subslot/port VPI VCI end-loopback command and enter the wrong VP/VC combination:
connection doesn't existAny future atm ping commands to any connections (both valid and invalid) always result in the same error message and the ping commands fails. It fails even if the connection is valid, and even if a VP ping is used (for example, no VC number is specified).
You can still use the IP ping command to diagnose the connectivity between the NRP2 (Node Route Processor 2) and NSP.
Workaround: None.
•
Symptom: On Frame Relay soft VCs, all frames sent with DE (discard eligibility) bit set are dropped at the serial interface even when the UPC is configured as "tag-drop" for the softVC.
Workaround: None.
•
Symptom: When an IMA interface module is installed in the switch chassis, and if "loopback PIF" is configured on any OC-3 or OC-12 interface, loopback is NOT reflected when queried through SNMP. Loopback is NOT actually being set.
Workaround: None.
•
Symptom: Inverse ARP does not succeed for a Frame Relay PVC terminated on ATM router module.
Workaround: None.
•
Symptom: The counter "Rx Frames with DE Tagged Locally" (counting DE bits tagged locally) should increment for all frames above the CIR (committed information rate). This conforms to the policy feature available on the Stratacom ATM switches.
Workaround: None.
•
Symptom: All MPLS VPN connectivity is broken on Cisco 7500, Cisco 7200, and all common platforms if there is a TCATM link between the two provider edges (PEs) that falls in the IGP path and LSP. The routers are disabled from initiating headend VC request over TCATM link.
Workaround: None.
•
Symptom: A subinterface might be stuck in the waiting state after a CPU reload. The CPU reload is displayed with the show atm statistics command. This will cause the P2MP VC count to be less than what it was before CPU reload.
Workaround: To bring the subinterface up, enter the shutdown command and then the no shutdown command on the subinterface.
•
Symptom: If an 8-port Gigabit Ethernet interface module is hot swapped after configuring PAgP port channel between two Catalyst 8540 CSRs, members on the hot swapped side of the port channel move out of it while the members on the side that was not hot swapped remain in the port channel. This happens only with the 8-port Gigabit Ethernet interface module.
Workaround: After some time, PAgP might add the hot swapped interface back to the port channel. If this does not happen, configure the ports manually.
Resolved Caveats—Cisco IOS Release 12.1(13)E1
Cisco IOS Release 12.1(13)E1 is a rebuild release for Cisco IOS Release 12.1(12c)EY. The caveats in this section are resolved in Cisco IOS Release 12.1(13)E1 but may be open in previous Cisco IOS releases.
•
Symptom: When bursty traffic is sent to the ARM via a carrier module installed 1-port OC-12 interface module in a peer Catalyst 8540 MSR, the ARM interface displays underruns and drops packets.
Note
Workaround: None.
•
Symptom: An ATM interface might stick in the state "going down" if the PVC is created from the ATM router module with VCI "34" on the ATM side. For example, "atm pvc 2 35 pd on interface ATM0/0/0 7 34" causes the line to flap "up" and "down".
Note
Workaround: Recreate the PVC using any VCI other than "34" on the ATM side.
•
Symptom: A Route Processor crash occurs when address registration fails at the switch. Also, an error message is flashed to the console. This problem occurs when traffic is flowing over a Flexwan module and either a hw-module-reset or OIR has occurred.
Workaround: None.
•
Symptom: When you try to upgrade the FPGA from 1.8 to A.1.9 on the carrier module C85MS-SCAM-2P with HW version 3.1, the following error message appears:
"Hardware revision 3.1, of controller in slot 9 does not match list in file headerIncompatible hardware for given image bootflash:fi-c8540-scam.A.1-9"Workaround: Use carrier modules with HW version 3.0 or 4.0.
•
Symptom: Hosts connected to the Catalyst 8540 switch through an Eight Port Gigabit Ethernet module, that do not transmit data frequently are intermittently unreachable.
The connectivity to the hosts eventually recovers.
Workaround: None.
•
Symptom: When a point-to-multipoint SVC connection is set up through a switch (root and leaf are active and cross-connected), and a request is received to add a leaf through another interface, and the allocation of this half-leg fails, the connection structures associated with this half-leg are not cleaned up. Therefore, when the connection is cleared, that failed connection cannot be released, which ties up bandwidth and VPI and VCI resources for the root half-leg of the connection.
Note
Workaround: None.
•
Symptom: Avaya PBX and Definity PBX interoperability testing.
The following caveats exist for signalled ATM point-to-multipoint (P2MP) connections and the debugging of issues related to them on the Catalyst 8500 ATM switches:
–
–
–
–
Workaround: None.
•
Symptom: When the Cisco ATM switch is running UNI 4.0 with a Fore or Marconi switch, it might releasing a call because of unsupported traffic parameters.
This occurs when the Cisco ATM switch is connected to the other vendor equipment and configured to operate with UNI 4.0 version across the link and as a non-negotiated link. If any traffic parameter values are modified the Cisco ATM switch treats this as fatal error and releases the call.
Workaround: Hardcode the UNI version to 3.1
•
Symptom: When using ATM Frame Relay IWF on a Catalyst 8500 MSR running Release 12.0(18)W5(22) or later, the output queue of the pseudo-ATM interface might go to 40/40 and remain there even if no traffic is going through the switch on the ATM-FR circuits.
Note
Workaround: None.
•
Symptom: ATM PVCs configured on the ATM0.X subinterfaces might appear under the ATM0 main interface. This leaves the subinterfaces with out ATM PVCs and no connectivity. This only occurs after a switch reload, power cycle or using the force-failover main-cpu command on Catalyst 8540 MSRs running Cisco IOS Release 12.1(7a)EY, 12.1(7a)EY1, or 12.1(10)E.
Workaround: Use the no sync dynamic-info command to disable sync dynamic-info.
•
Symptom: Frame relay PVC goes down and remains down after enabling the OAM feature and performing a shut / no shut on the serial interface. However, traffic flow is not affected through the PVC even though it appears down.
Workaround: Disable OAM and shut/no shut the serial interface to force the interface to appear
•
Symptom: The ATM router module has only 147 Mbps of resources available on the ATM interfaces. The following display shows the available bit rate for an ATM router module ATM interface:
Switch# show atm interface resource atm 0/1/1[Information Deleted]Resource Management state:Available bit rates (in Kbps):147743 cbr RX, 147743 cbr TX, 147743 vbr RX, 147743 vbr TX,147743 abr RX, 147743 abr TX, 147743 ubr RX, 147743 ubr TXAllocated bit rates:0 cbr RX, 0 cbr TX, 0 vbr RX, 0 vbr TX,0 abr RX, 0 abr TX, 0 ubr RX, 0 ubr TXBest effort connections: 0 pvcs, 0 svcsWorkaround: None.
•
Symptom: Catalyst 8540 MSRs connected in a ring configuration (via OC-48 modules) experience problems recovering from a power outage. Symptoms include OC-48 interfaces on one side of the connection being DOWN/DOWN while the directly connected interface is UP/UP.
Workaround: This problem occurs only as a result of a bad clock configuration on the entire network. Power cycle the router, then restart it from the command line; after this has been done, enter the shutdown/no shutdown command sequence.
•
Symptom: The ATM router module interface shuts down after entering the epc portstuck-manual-reload command with IP traffic.
Workaround: None.
•
Symptom: The display might show that oversized packets are on an enhanced ATM router module interface when the interface is a member of a BVI. This happens even when there is no traffic except BPDU and OSPF routing updates. These "large packets" are not the real traffic transiting through the ATM router module. They are CPU-bound traffic (i.e. routing updates). The following is the message that might occur on the console:
00:41:36: %TBRIDGE-4-GIANT: Giant received on ATM3/0/1, 1514 exceeds 1500DA 0100.5e00.0005 SA 0001.4274.ee09 [0x080045C0]
The system recovers from this.Workaround: None.
•
Symptom: In a multi-vendor environment, if the non-Cisco IMA interface side does not support Test pattern, as specified in Inverse Multiplexing for ATM (IMA) Specification Version 1.1 by the ATM Forum, the Cisco IMA interface status might change from up to down continuously.
Workaround: None.
•
Symptom: When the Catalyst 8540 is powered-up or power cycled, the shutdown/no shutdown command has to be entered on the CBR ports to activate the CBR ports.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(12c)EY
Cisco IOS Release 12.1(12c)EY is a rebuild release for Cisco IOS Release 12.1(12c)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(12c)EY but may be open in previous Cisco IOS releases.
•
Symptom: A memory leak might occur during an "atmSoft_timerProcess". With every "soft-vc" retry the holding memory on "atmsig CC" and "atm soft CV timer" increases and the "free" memory decreases. The free memory might fall to as low as 700K at times.
Workaround: None.
•
Symptom: After you configure a Point-to-Multi Point connection, if you try to remove the connection from the leaf side, the following message is displayed:
% PVC de-configuration not allowed with snooping on!!% De-configure snooping first.Workaround: Delete the PVC or PVP from the root end.
•
Symptom: The Catalyst 8540 MSR might experiences an unexpected reload when reset. This creates a few new entries in the snmpTargetAddrTable. This operation is caused by SNMPV3.
Workaround: None.
•
Symptom: If you have hard CES PVCs whose ATM legs belong to a shaped VP tunnel, and you perform a route processor switchover, one of the following two problems might occur:
1. The new primary route processor crashes after displaying the following message:
"Initializing feature card"2. All CBR interfaces go down and they come up only after you perform no shutdown on the interfaces.
Workaround: None for problem 1. For problem 2, perform a no shutdown on all CBR interfaces.
•
Symptom: The switch crashes when the IPX encapsulation is changed on Fast Ethernet or Portchannel interfaces.
This might occur after you change encapsulation and then perform a shutdown/no shutdown.
Workaround: None.
•
Symptom: Many PHY parity errors occur on the traffic going through the OC-12 interface module. This causes PHY Parity Error interrupts due to high CPU utilization (increasing to approximately 90 percent) during route processor switchover.
This occurs because the PHY Parity Check bit in the SMR register is set to "1" (enabled) during route processor switchover. Then SMR register is set from 0x0 to 0xF during switchover, enabling the PHY Parity Check bit. The default value of this register should be 0x0 (PHY Parity Check disabled).
Workaround: None.
•
Symptom: On a Frame Relay to ATM Soft VC, all the frames sent with the DE bit set are dropped at the serial interface even when the UPC intent is configured as "tag-drop" for the Soft VC.
Workaround: None.
•
Symptom: Prefix entry pointing to invalid adjacency in EPIF Patricia tree.
Workaround: None.
•
Symptom: When a port on a Catalyst 8540 CSR is configured as a Snoop (monitor) interface, the traffic on the Snoop interface appears corrupted. Sniffing the same traffic on the attached Catalyst switch shows that the traffic itself is not corrupted. The issue is that the Snoop interface does not provide reliable information to the monitoring station attached to that interface.
Workaround: None.
•
Symptom: You may see spurious memory access at atom_VCConn_nodecmp during a route processor switchover.
Workaround: None.
•
Symptom: Multiple Cisco products contain vulnerabilities in the processing of Simple Network Management Protocol (SNMP) messages. The vulnerabilities can be repeatedly exploited to produce a denial of service. In most cases, workarounds are available that may mitigate the impact. These vulnerabilities are identified by various groups as VU#617947, VU#107186, OUSPG #0100, CAN-2002-0012, and CAN-2002-0013.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-malformed-snmp-msgs-pub.shtml.
Workaround: None.
•
Symptom: SSH fails with CRC-check failed error message some times.
Workaround: None.
•
Symptom: A Simple Network Management Protocol version 3 (SNMPv3) user is created by using message digest 5 (MD5) authentication with the following commands:
–
–
An SNMP walk is performed, the configuration is saved, and the switch is reloaded.
newhope:~/src/wccp2# snmpwalk -v 3 -u abcdefghij -A abcdefghij -a MD5 -lAuthNoPriv 194.12.224.11It is working and a debug snmp header shows this:
Incoming SNMP packet: v3 packet security model: v3 security level: authusername: abcdefghijA second SNMP walk is performed:
newhope:~/src/wccp2# snmpwalk -v 3 -u abcdefghij -A abcdefghij -a MD5 -lAuthNoPriv 194.12.224.11After the second SNMP walk is performed, the command does not return any output and the debug snmp headers show this:
Incoming SNMP packet: v3 packet security model: v3 security level: noauth: username: abcdefghijWorkaround: None.
•
Symptom: There is no instance of ACL card in the MIB table.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(12c)E1
Cisco IOS Release 12.1(12c)E1 is a rebuild release for Cisco IOS Release 12.1(12c)E. The caveats in this section are resolved in Cisco IOS Release 12.1(12c)E1 but may be open in previous Cisco IOS releases.
•
Symptom: Upgrading from Cisco IOS Release 12.1(12c)E or 12.1(12c)EY, using the standard upgrade procedure, causes the switched connections (transiting or terminating) on the Catalyst 8540 to be torn down if ILMI keep-alives are enabled on the interface of the Catalyst 8540 and its peer through which the connections are transiting.
Workaround: Before upgrading, disable ILMI keep-alive on both the ends of the NNI link between the Catalyst 8540 and its peer switch.
•
Symptom: When the Catalyst 8540 MSR image is migrated from either Release 12.1(11b)E5 or 12.1(12c)E to Release 12.1(12.1.020719.) using the Route Processor switchover method, the new Primary Route Processor reloads while taking over.
Workaround: None.
•
Symptom: If an IMA group exists between a Catalyst 8540 and Catalyst 8510 switches with CBR soft VCs configured between the switches and a link of the IMA group is shutdown or fails the IMA interfaces still display the pervious bandwidth value. This bandwidth value is greater than the actual resources available on the interface and the soft VCs still appear as "up" on both the Catalyst 8540 and Catalyst 8510 switches. This continues to happen as more links are shutdown or fail until the 2 CBR soft VCs that require the resources of 7 E1 links still appear as "up" with only 1 E1 link actually "up" between the IMA interfaces and the other 6 links shutdown.
Note
Workaround: Use the shutdown and no shutdown commands on the IMA group and it will display the correct values of available and allocated bandwidth.
•
Symptom: Occasionally, after route processor switchover or reboot of the Catalyst 8540 MSR, there is very low usage in OC-12.The output rate of the show interface command is a 5-minute output rate 0 bits per sec, 0 packets per sec.
Workaround: Run the shutdown and no shutdown commands on the affected interface.
•
Symptom: The Catalyst 8540 MSR might encounter a memory leak in ATMSIG processes and this might render the switch unusable after one month of operation.
Workaround: None.
•
Symptom: A Catalyst 8540 MSR running Cisco IOS (tm) PNNI Software (cat8540m-WP-M), Release 12.0(20)W5(24a) RELEASE SOFTWARE, using the QUAD E3 interface module might experience a limitation to the number of soft-VCs that can be established to it's interfaces.
Output seen on the remote configured switch indicates "NOT CONNECTED" when using the show atm vc command.
The following similar output might appear when using the debug atm sig-soft interfaces atm card/subcard/port:
*May 14 06:57:09.511: ATMSOFT(ATM10/1/0 vpi = 249; vci = 32 ) Received SETUP*May 14 06:57:09.511: ATMSOFT(ATM10/1/0 vpi = 249; vci = 32 ) tx/rcv row indexs*May 14 06:57:09.511: RxQosIndex = 2147483647 TxQosIndex = 2147483647*May 14 06:57:09.511: ATMSOFT(ATM10/1/0 vpi = 249; vci = 32 ) Prepare for XConnect Installation*May 14 06:57:09.511: ATMSOFT(ATM10/1/0 vpi = 249; vci = 32 ) Call rej - xconnect fail*May 14 06:57:09.511: ATMSOFT(ATM10/1/0 vpi = 249; vci = 32 ) Call rejected*May 14 06:57:09.511: reason: connection installation failure received rel comp at dest, bad param,could be rel collision pvlIfCB_p:0 pvlVpi:0 pvlVci:0Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(12c)E
Cisco IOS Release 12.1(12c)E is a rebuild release for Cisco IOS Release 12.1(11b)E1. The caveats in this section are resolved in Cisco IOS Release 12.1(12c)E but may be open in previous Cisco IOS releases.
•
Symptom: When an ATM soft-VC or soft-VCP is configured between two interfaces of a Catalyst 8540 MSR using a particular VPI, and a soft-VC or soft-VCP is configured with the same VPI, the following error appears:
%ATMCORE-3-INTERNAL_ERROR: atmCore_connAllocateHalfLeg: return resource err.2The internal error message should not appear when the soft-VC or soft-VCP configuration fails.
Plus, the following expected message is seen more than once:
connection creation fails:vpi/vci has been usedThe internal error message is not seen when a PVC configuration on an interface is followed by a PVP configuration that uses the same VPI. Also, the internal error is not seen when the interface on which the configuration is being done does not have the resources required by the soft-VC or soft-VCP.
After the configuration is executed and then fails, this causes the allocation of the bandwidth requested by the CBR VC even though no VC actually exists. If, for example, N kbps is free on an interface, and a soft-VC is configured from that interface with a VPI-VCI already in use and uses a CTTR of N kbps, the configuration fails, but all N kbps that were free on that interface are lost. This bandwidth loss is seen only with the CBR service-category.
Note
Workaround: For the Catalyst 8540 MSR, if you switchover the route processor it recovers the lost bandwidth on the interface. No workaround exists on the Catalyst 8510 MSR.
No workaround exists for the following error message:
"%ATMCORE-3-INTERNAL_ERROR: atmCore_connAllocateHalfLeg: return resource err.2•
Symptom: If you configure OAM on a PVC while it is down, the OAM state of the VC is displayed as "not applicable" instead of "OAM-Down" and never changes. This means OAM cells are not sent even after the VC comes UP. This only occurs if AIS is not enabled for the VC.
Workaround: Configure OAM for a VC only when it is in UP state.
•
Symptom: When setting up SVCs over an NNI link, one system becomes the master in terms of deciding which VPI/VCIs are assigned to SVCs. When the slave signals a Connection ID in SETUP, it may specify a VPI/VCI value. The Catalyst 8540 MSR signals the lowest VCI in the address range.
When signaled over a VP tunnel, the Connection Identifier IE formulated by the slave specifies that the VCI signalled is to be ignored.
However, when connection management allocates the half-leg for the ingress interface, if the connection ID specified in the ID matches an already-existing connection, and the connection signaled is point-to-multipoint link, the Connection Manager assumes the signaled SVC is the same as the one already allocated. This causes two SVCs point to the same VPI/VCI. This, subsequently, results in the deletion of the existing half-leg, the bandwidth allocated by the SVC lost, and the existing connection turned into a half-leg connection (dangling).
Workaround: None. The dangling half-leg can be recovered for SVCs by using the shutdown and no shutdown commands on the interface.
•
Symptom: The values of input and output bytes on an ATM interface do not increment on successive SNMP queries. Increments in the values appear after 5 or 6 successive queries (approximately 25 seconds).
Workaround: None.
•
Symptom: After OAM detects a failure on a PVC terminating on the ATM router module (ARM), it fails to remove the adjacency, the directly connected route, and any additional route pointing to the next hop reachable over the failed PVC. This condition might stop traffic.
Workaround: There is no work-around, other than to perform an admin shutdown of the ARM subinterface (if the PVC is the last or only PVC on that subinterface). Alternatively, you could remove the PVC until the problem has been rectified.
•
Symptom: The Peak Cell Rate sent in the ATM Traffic Descriptor Information Element in the ATM Signalling SETUP messages for CES IWF circuit Soft PVCs does not conform to the standard.
The PCR signalled is always greater than or equal to that needed.
This has no effect on the set up of connections to other Catalyst 8540, Catalyst 8510, and LightStream 1010 ATM switches. These same calls may not be accepted by systems from other vendors.
•
Symptom: The IRB connection has no connectivity in a native VLAN across ISL/dot1q Gigabit Etherchannel in the same bridge-group. However, there is connectivity on other VLANs.
Workaround: Configure a dummy VLAN as native VLAN on the dot1q trunk.
•
Symptom: After deleting a connection, an immediate request to re-install with the same VPI/VCI results in connection-creation failure messages.
Workaround: Increase the bit map size from 32 to 64, to avoid reuse of the same VPI/VCI immediately after a call is released.
•
Symptom: Two adjacent ATM devices (for example, ATM port at Catalyst 8540, ATM NIC, or Cisco 7200 router) had problems performing ILMI negotiation. The result of the ILMI negotiation is VPI/VCI range fails and the following error message appears:
%SYS-2-MALLOCFAIL: Memory allocation of 1650891065 bytes failed from 0x6066352C,pool Processor, alignment 0-Process= "ILMI Response", ipl= 0, pid= 80-Traceback= 6009E1F4 6009FCA0 60663534 60661488 6065EF94 6065E934 6009944C 60099438Workaround: None.
•
Symptom: Spurious memory access at lss_arm_atm_adjacency_mac might occur after entering the epc portstuck-manual-reload command simultaneously on two ATM router module interfaces of two connected switch routers. There is spurious memory access to both switch routers. This has no functionality impact.
Workaround: None.
•
Symptom: After configuring an ipx network between a Catalyst 8540 MSR and a Router 7500, the Catalyst 8540 MSR displays old buffers detected messages. This does not impact functionality.
Workaround: None.
•
Symptom: When OAM management is configured on a PVC of an ATM router module in order to check the end-to-end connectivity, the subinterface might stay up/up even if the remote end is down.
Workaround: None.
•
Symptom: When you enter the show lane client command, part of the command output includes a statement such as "LEC up for 8 hours 41 minutes". The current MIBs, such as, interface mibs, ATM mibs, and the LAN-EMULATION-CLIENT-MIB do not display this information.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(11)E1
Cisco IOS Release 12.1(11)E1 is a rebuild release for Cisco IOS Release 12.1(11)E. The caveats in this section are resolved in Cisco IOS Release 12.1(11)E1 but may be open in previous Cisco IOS releases.
•
Symptom: LightStream 1010 switches running 12.0(X)W5 Cisco IOS may not completely adhere to the ITU-T Q.2971 "B-ISDN DSS2 UNI Layer 3 Specification for Point-to-Multipoint Call/Connection Control". This could affect interoperability.
Workaround: Configure UNI Version 3.1 on the LightStream 1010 ATM interfaces if the point-to-multipoint interoperability issue is observed.
•
Symptom: Occasionally after a system crash or route processor switchover you may see a PVC connection that exists in the interface software configuration but not does appear configured in the interface hardware.
Workaround: Perform an OIR on the interface module with the failed connection or perform a route processor switchover to restore the connection on the interface hardware.
•
Symptom: The Catalyst 8540 reboots if you try to add a point-to-multipoint leaf to the point-to-multipoint root from Ciscoview ADP fileset for Catalyst 8540 version 5.0.
Workaround: None.
•
Symptom: If you use the no negotiation auto command in interface config mode on a 2-port Enhanced Gigabit Ethernet interface module with GMAC-D2 (hw-version 6.0 onwards), any packets larger than 1518 bytes or less than 64 bytes are filtered by the GMAC (Gigabit Ethernet Media Access Controller).
Workaround: Reload the particular interface using the epc portstuck-manual-reload interface card/subcard/port command.
Resolved Caveats—Cisco IOS Release 12.1(11)E
Cisco IOS Release 12.1(11)E is a rebuild release for Cisco IOS Release 12.1(10)EY. The caveats in this section are resolved in Cisco IOS Release 12.1(11)E but may be open in previous Cisco IOS releases.
•
Symptom: If PNNI continuously receives a high rate of corrupted PTSP packets over a long period of time, it can cause a large number of internal node numbers to be generated for bogus Node IDs. If the number of internal node numbers reaches 1032, it can cause a crash.
Under normal conditions, PTSP packets should have a very low rate of being corrupted. But if PNNI interfaces are tunneled through a network where the connections are rate limited to a too low rate it can force cell drops which can corrupt packets.
Workaround: Troubleshoot and remove the cause of PTSP packet corruption.
•
Symptom: You might see a Bus Error exception at get_slot_ptr while trying to do the continuous snmpwalk on the ciscoLS1010SubModuleGroup object of CISCO-RHINO-MIB and simultaneously toggling the redundant power-supply after at least one route processor switchover.
Workaround: None.
•
Symptom: Once the route processor switches over the following error message is seen on new Primary if we have a Gigabit processor interface module:
%ONLINEDIAG-6-ACCESS_TEST_WARNING: Unable to read from card in slot 10 sub_card 0.Enter the show diag online access command to get more details about the failure.
Workaround: None.
•
Symptom: Catalyst 8540 CSR eight port Gigabit Ethernet interface modules fails to run CLNS.
Workaround: Update to Cisco IOS release 12.1(11b)E.
•
Symptom: A Catalyst 8540 MSR locks up when configuring ATM port snooping. The switch lockup can occur if ATM port snooping is configured without an analyzer connected to the snoop port.
Workaround: 1.- Load the most recent version of Cisco IOS.
or
2.- Have the analyzer connected to the snoop port prior to configuring ATM port snooping. Keep the analyzer connected until the snoop configuration is removed.
•
Symptom: In a network running cell-mode MPLS (tag switching) with a large number of IP routes (and consequently, with many TVCs being set up), a 16-port OC-3 interface module might fail to allocate a new TVC on a bind request and print out the below error message. This occurs even though the number of TVCs currently set up does not reach the limit of VCs supported on that interface or the cross-connect interface.
Jun 22 20:49:02.011: %TCATM-4-RESOURCE_LIMIT: VC resource exhausted on ATMx/y/z
Workaround: Enter the shutdown command followed by the no shutdown command on the interface for which the error message appears.•
Symptom: A switch router with an eight-port Gigabit Ethernet interface module might forward duplicate broadcasts when bridging is configured on a trunk.
Workaround: None.
•
Symptom: The hierarchical VP tunnel configuration fails on a WAI-OC3-1S3M mixed mode port adapter module when it is in slot 0 subslot 1 of the C85MS-SCAM-2P carrier module. Slot 0 subslot 0 of the carrier module can either be empty or have another card in it, and hierarchical VP tunnel configuration will still fail. However, if the mixed mode port adapter is inserted in slot 0 subslot 0 of the carrier module, then the hierarchical VP tunnel can be configured.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(10)EY
Cisco IOS Release 12.1(10)EY is a rebuild release for Cisco IOS Release 12.1(10)E. The caveats in this section are resolved in Cisco IOS Release 12.1(10)EY but may be open in previous Cisco IOS releases.
•
Symptom: Excessive CLP0 discards on TSCAM for PD enabled VCs.
Workaround: None.
•
Symptom: Multicast packets of some specific lengths (121 bytes to 240 bytes) cause a portstuck condition on the ATM router module interfaces.
Workaround: None.
•
Symptom: Unusual delays in Frame Relay VC counters update on Frame Relay ATM interface modules.
Workaround: None.
•
Symptom: Globally changing TDP to LDP or changing LDP to TDP sometimes causes a loss of neighbors.
Workaround: Following is a four-step work-around:
1. Configure no tag-switching ip on both sides.
2. Configure the desired protocol at each end.
3. Wait two minutes.
4. Configure tag-switching ip at each end.
•
Symptom: Ping fails when one of the two parallel soft VCs is missing.
Workaround: Use clear bridge command to reset bridge forwarding cache.
•
Symptom: When Frame Relay/ATM Network Interworking (FRF.5) is configured on a Catalyst 8540, traffic above the peak information rate is dropped, and the INPUT ERROR counter is incremented, rather than the INPUT DROPS counter.
Workaround: None.
•
Symptom: After the simultaneous rebooting of switches terminating the active and passive side of Soft-VC for FRF5, randomly Soft-VPC stays up but, traffic might be dropped on the Frame Relay interface module. This might occur either on the active side of Soft-PVC or on the passive side of the call.
Workaround:
The Frame Relay interface module starts forwarding traffic only after you use the shutdown and no shutdown commands on the serial interface.
You might see the following error message when you reset the Frame Relay interface module:
Error Message: Channel-Group 1 State Change on FR-PAM failed 1For additional details, refer to DDTS CSCdv17817.
•
Symptom: The Internetwork Packet Exchange (IPX) statistics sometimes fluctuate when no traffic is flowing through the interface. The only affected functionality is IPX statistics.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(10)E
Cisco IOS Release 12.1(10)E is a rebuild release for Cisco IOS Release 12.1(7a)EY1. The caveats in this section are resolved in Cisco IOS Release 12.1(10)E but may be open in previous Cisco IOS releases.
•
Symptom: A crash occurs while configuring BVI/eigrp, or while powering off the redundant power supply.
Workaround: None.
•
Symptom: When running Cisco IOS Release 12.0(18)W5(22) on a Catalyst 8540 MSR, or on a LightStream 1010, interoperability problems occur when peers run an older Cisco IOS version, and tunnels and SVC/Soft VCs are used.
Workaround: Upgrade all switches to the most current Cisco IOS release.
•
Symptom: When a Catalyst 8540 MSR has a large number of inverse multiplexing over ATM (IMA) port adapter modules, a message similar to the following is generated:
-Process= "IMAPAM Onesec", ipl= 6, pid= 70-Traceback= 600D8280 600D8644 6049F928 600C03CC 600C03B8%SCHED-3-THRASHING: Process thrashing on watched message event.Workaround: None.
•
Symptom: When using a Packet Over SONET card, the Open Shortest Path First (OSPF) routes do not come up after a route processor switchover, even though they were up before the switchover.
Workaround: None.
•
Symptom: The switch might reload when VP tunnels are created and removed.
Workaround: None.
•
Symptom: On a Catalyst 8540 MSR, an SNMP query on the table ciscoLS1010PortTable, or on the object ciscoLS1010PortIfIndex, might result in a crash.
Workaround: None.
•
Symptom: The Management Information Base (MIB) object "ifAdminStatus" shows "UP" for an ADMIN DOWN E1 Controller after a reload or a route processor switchover.
Workaround: Enter the no shutdown/shutdown command sequence on the controller.
•
Symptom: With a Frame Relay-ATM soft-VC set up in FRF5, UPC on the passive side is always set to PASS, regardless of the hard setting on the serial interface with the frame-relay upc-intent tag-drop command. The default is PASS. When this is changed to tag-drop, the configuration does not reflect the change and the show vc int serial command always shows the UPC intent as PASS.
Workaround: None.
•
Symptom: The SNMP TRAP object ccrCpuStatusChange returns an invalid value.
Workaround: None.
•
Symptom: A port of a 4-port DS3 port adapter for LightStream 1010s and for Catalyst 8540s is in a yellow or red alarm state, and stops sending DS3 framing for a brief time when the alarm is cleared from the interface and the interface transitions to an UP/UP state. The loss of frame (LOF) can be seen when the Acterna tool TBERD is connected to the TX port of the DS3 port.
Workaround: None.
•
Symptom: The EIGRP Summary Routes are not properly populated in the adjacency table when a default route also exists. This affects the Fast Ethernet and ATM router module interfaces, but does not impact the Gigabit Ethernet interfaces.
Workaround: None.
•
Symptom: A Cisco edge services router (ESR) that is running Cisco IOS Release 12.1(2)E1 experiences interoperability problems when an ATM traffic descriptor information element is included in the CONNECT message on a non-negotiated link. Under such conditions, a STATUS message is sent with cause "invalid information element contents (100)" and call will not succeed. This problem is observed when Cisco routers are connected to other vendor switches like FORE, and occurs only when UNI version is 4.0.
Workaround: None.
•
Symptom: Ping might fail through an 8-port Gigabit interface module with dot1q/vlan 1.
Workaround: Save the configuration and reload the switch.
•
Symptom: The following error messages might occur on a reload:
Mar 22 10:43:43: %SYS-2-WATCHDOG: Process aborted on watchdog timeout, process = TCP Timer.-Traceback=3D 600B6498 600B9D68 600B35B8 602D2EF4 602D2DB0 6032E26C 60347EB8
Workaround: None.•
Symptom: IP traffic is not sent over a PVC if aal5mux encapsulation is used on a point-to-point subinterface on the Catalyst 8540 MSR. If the deb atm errors command is enabled, the following error messages will appear:
point-to-point interface does not have a VCDWorkaround: None.
Resolved Caveats—Cisco IOS Release 12.1(7a)EY1
Cisco IOS Release 12.1(7a)EY1 is a rebuild release for Cisco IOS Release 12.1(7a)EY. The caveats in this section are resolved in Cisco IOS Release 12.1(7a)EY1 but may be open in previous Cisco IOS releases.
•
Symptom: Point-to-multipoint PVCs that are configured by using SNMP are not preserved across route processor switchovers.
Workaround: Disable the dynamic synchronization feature.
•
Symptom: After disabling the synchronization of both the running configuration and the startup configuration, a switchover results in connection installation failure, and the failure of all Layer 3 and ATM router module card downloads.
Workaround: None.
•
Symptom: In a primary/secondary redundant router configuration, virtual-circuit cross-connections and virtual channel link (VCLs) that are in the primary router and are deleted by using the Simple Network Management Protocol (SNMP) are still present in the secondary router. Minutes after the configuration change, the deleted elements are still present in the secondary router. If SNMP is used to write the running configuration to the startup configuration, a message is generated that indicates the configuration was received on the secondary router. When the primary router is reloaded, the secondary takes over, but does not reflect the change in its running configuration, even though the startup configuration correctly indicates the cross-connects/VCLs are deleted.
Workaround: None.
•
Symptom: After a reload of the primary router in a primary/secondary redundant Catalyst 8540 configuration, the secondary router cannot fetch as many of the elements of the atmVcCrossConnectTable as are known to exist. The following message appears when a get is attempted, and fails, on the table:
00:22:13: %ATMCORE-3-INTERNAL_ERROR: ATOM ERROR: ifCB_p=NULL for highIfWorkaround: None.
•
Symptom: Cisco Discovery Protocol (CDP) does not work over Portchannel platforms.
Workaround: None.
•
Symptom: After switch processor switchover, interfaces that were up prior to the switchover become administratively down.
Workaround: After switch processor switchover, enter the no shutdown command to make the interface up.
•
Symptom: Use of the epc portstuck-manual-reload command on a port channel causes a Catalyst 8540 to crash.
Workaround: The epc portstuck-manual-reload command should be used only on physical ports; it should not be used on logical ports. If a port-channel member needs to be reloaded, individually reload each physical port in the port channel. Allow approximately 1 minute for the port to reload/recreate the VCs, before using this CLI on the next port.
•
Symptom: The unframed mode is not preserved when transiting across route processor switchovers.
Workaround: Delete the serial interface, and use the CLI to recreate it.
•
Symptom: When the shutdown/no shutdown command sequence is entered for a port in a CES card, a burst of bit errors can be observed on the circuit on some other port of the same port adapter.
Workaround: None.
•
Symptom: Under stressful conditions such as large topology changes, TVC cross-connections might fail to be removed from the ATM switching fabric, causing the TC-ATM function to repeat attempts to remove the TVCs. This might lead to the TC-ATM state process entering a bad state, and to an error message being displayed.
Workaround: None.
•
Symptom: Not all of the contents of the atmVcCrossConnectTable can be seen after the removal and re-insertion of a module. The message:
%ATMCORE-3-INTERNAL_ERROR: ATOM ERROR: ifCB_p=NULL for lowIfis seen on the console during some Simple Network Management Protocol (SNMP) get operations.
Workaround: None.
•
Symptom: During the Interim Local Management Interface (ILMI) batch synchronization process, only the last LAN Emulation Configuration Server (LECS) address can be successfully synchronized from the active route processor to the standby route processor. This affects redundancy support for the service registry feature only when more than one LECS address has been configured on the active route processor.
Workaround: None.
•
Symptom: When RFC 1483 bridged frames are sent with PID 0x0001 to the controller of a Catalyst 8540 enhanced ATM router module, the frame is corrupted. This might show up as "dispose ip.formaterr" from "debug ip error output".
Workaround: If the equipment that sends the frames can be configured so that it sends RFC 1483 bridged frames with PID 0x0007 only, make that change. In all other cases, there is no workaround.
•
Symptom: The system crashes when changing a Bridge Group Virtual Interface (BVI) to a non-BVI using the same IP address.
Workaround: None.
•
Symptom: Some point-to-multipoint crossconnnect entries are missing from the atmVcCrossConnectTable when the point-to-multipoint root if-index is higher than the if-index of the leaves.
Workaround: None.
•
Symptom: Interfaces with Tag enabled get stuck in the TDP not ready state.
Workaround: None. The only way to bring up the Tag on such interfaces is via system reload.
•
Symptom: During TagVC installation, if the switch driver returns an error, the Connection Manager leaves the TagVC in the wrong FSM state. This can lead to TVCs existing on an interface when there are no tag bindings.
Workaround: None; reload the switch to clear such unused TVCs.
•
Symptom: When running Cisco IOS Release 12.0(13)W05(19), a bus error causes the system to restart. This occurs when an 8-port Gigabit Ethernet card with control traffic is at or near its traffic capacity.
Workaround: None.
•
Symptom: If a large frame or packet (more than 6096 Bytes) is received on the tag control virtual circuit (0/32 of an interface enabled for tag switching), the frame is not purged from the switch fabric. Subsequent packets received on this virtual circuit are dropped, causing the Tag Distribution Protocol (TDP) and the routing session (OSPF) to time out.
Workaround: Disable, then re-enable, tag switching on the affected interface; subsequent occurrences can be avoided by increasing the queue depth, using the following entry:
atm threshold 2 max-queue-limit 511•
Symptom: The ATM router module portstuck recovery code does not always work.
Workaround: None.
•
Symptom: Channel stuck detection/recovery doesn't work sometimes for ATM router module interface.
Workaround: Enter the sh epc portstuck arm command to determine whether "stat_sent" or "stat_rcvd" counters are incrementing up on only one channel of any ATM router module port. If the counters are incrementing up on only one channel, this might indicate the problem is a stuck channel; enter the epc portstuck-manual-reload interface command to manually reload the port.
•
Symptom: Over an 8-port Gigabit Ethernet interface, a ping to a Bridge Group Virtual Interface (BVI) fails over a dot1q subinterface that has a VLAN ID of 1 with native VLAN.
Workaround: Configure the dot1q VLAN ID to a value other than 1 for a native VLAN (for example, configure dot1q vlan id 3 as native VLAN).
•
Symptom: When operating at 75 to 100% of wire speed with a packet-size range of 64 to 84 bytes, a simple marking-action in an IPQoS service policy causes a port-stuck condition in an Enhanced Gigabit Ethernet processor interface card.
Workaround: None.
•
Symptom: Ping might fail through an 8-port Gigabit interface module with dot1q/vlan 1.
Workaround: Save the configuration and reload the switch.
•
Symptom: Under a heavy load, the switch router mis-orders cells going out of single-port OC-12 port-adapter modules placed in a carrier module.
Workaround: Whenever a heavy load is expected on single-port OC-12 modules placed in a carrier module, reprogram the carrier module with Carrier Module FPGA Image 1.9, and upgrade the Cisco IOS image to the latest version.
•
Symptom: The ifTable indexes do not match the MIB-II table indexes.
Workaround: None.
•
Symptom: Under stressful conditions, tag virtual-circuit cross-connections fail to be set up, even though the input and output legs have been allocated. This can cause the input and/or output legs to remain allocated, but be in the NOT CONNECTED state.
Workaround: None.
•
Symptom: A LightStream 1010 stops switching some VCs, and log error messages similar to the following example are generated:
Jan 9 12:37:34.793: ATMSIG(12/1/0:0 0,37 - 274610/00): (vcnum:0) XConnected failed ToJan 9 12:37:34.793: ATMSIG(12/0/1:0 0,268 - 398903/00): (vcnum:0)Jan 9 12:37:34.793: ATMSIG: CONNECTION MANAGER/CAC Xconn install requestPort A - (12/1/0) : 0 vpi = 0; vci = 37Port B - (12/0/1) : 0 vpi = 0; vci = 268; new vpi = 0; new vci = 268 existing vpi/vci for leg BJan 9 12:37:34.793: ATMSIG: CONNECTION MANAGER/CAC Xconn install resultCONN_INSTALL_RESULT_HW_FAILWorkaround: Reload the switch.
Resolved Caveats—Cisco IOS Release 12.1(7a)EY
Cisco IOS Release 12.1(7a)EY is a rebuild release for Cisco IOS Release 12.1(6)EY. The caveats in this section are resolved in Cisco IOS Release 12.1(7a)EY but may be open in previous Cisco IOS releases.
•
Symptom: If the well known VC`s of a tunnel are deleted and configured manually and a route processor switchover occurs then these manually configured VC`s might not be recognized by the new primary route processor. Instead, they are replaced with the default well known VC.
Workaround: None.
•
Symptom: In Cisco IOS Release 12.1(6)EY and earlier, only two equal cost IP paths were supported on Ethernet processor interface based Fast Ethernet and Gigabit Ethernet interface modules.
Workaround: Upgrade to Cisco IOS Release 12.1(7a)EY. In this release, four path load balancing for IP is available through the epc epif-4-path-lbal command.
•
Symptom: On the ATM router module, the HyBridge Input P might cause high route processor utilization.
Workaround: For every bridge group on the ATM router module, add a map-group with a corresponding map-list. Under the map-list configuration add a "bridge atm-vc XXX broadcast," where XXX is the VCI value created on the ATM router module subinterface.
•
Symptom: The no negotiation auto configuration might not be preserved in the running configuration after an OIR of an 8-port Gigabit Ethernet interface module.
Workaround: Reconfigure no negotiation auto.
•
Symptom: You might not be able to change the spanning tree protocol of a bridge-group.
Workaround: Remove the current bridge-group configuration and then reconfigure it with the new spanning tree protocol.
•
Symptom: When SVCs go down and OSPF recalculates IP routes and updates the routing/cef/epc tables, if the switch does not have an arp entry, it will switch by using the route processor. This occurs when the network behind that "empty" arp entry is put into the epc-table, it finds no adjacency, and switches via the route processor.
Workaround: Configure 'arp timeout 0' under the ATM interfaces that are configured with SVCs.
•
Symptom: When entering the show atm ilmi-status command the switch shows the 0.0.0.0 as a peer IP address even though a loopback address is configured.
Workaround: None.
•
Symptom: Cell rejects and cell drops might occur because of the nature of UBR connections and inconsistent programming of scheduler on connections going out of OC48.
•
Symptom: When running cell-mode MPLS on a network of Catalyst 8500s with redundant paths, conversion for MPLS after a routing change might take up to 4 1/2 minutes. OSPF converges normally.
Workaround: None.
•
Ports on the Two-port Enhanced Gigabit Ethernet interface module might get stuck after an OIR.
Symptom: If a 256K Two-port Enhanced Gigabit Ethernet interface module is hot swapped with a 64K Enhanced Gigabit Ethernet interface module, the ports in 64K module might not come up.
Workaround: None.
•
Symptom: ATM-SOFT-PVC-MIB atmSoftPVccRetryThreshold has an incorrect default value 0 instead of 1. This incorrect value turns on atmSoftPvcCallFailuresTrap by default and impacts the usability of the feature.
Workaround: None.
•
Symptom: A ROMMON configuration might not work properly on a slot after performing an OIR on the interface module in that slot.
Workaround: Reload the switch router.
•
Symptom: The CMPM carrier module shows the same serial numbers for the carrier module and the interface modules in it; however, when the show hardware command is entered, different serial numbers from the carrier module are shown for both the carrier module and interface modules.
Workaround: None.
•
Symptom: After rebooting the switch router, Ethernet interfaces might have invalid MAC addresses, causing connectivity problems over that interface.
Workaround: Enter the shutdown command and then the no shutdown command on the affected Ethernet interfaces.
•
Symptom: When a static route to a connected interface is added and deleted when another static route is present in the same network, the packets might be process-switched.
Workaround: None.
•
Symptom: This condition occurs when asymmetrical routing, bridge groups, and LANE clients are configured on ATM router module sub interfaces. The bridge groups allow bridging to ELANs on non ATM interfaces and sometimes packets destined to an outside ELAN or VLAN might be sent or received over the LANE broadcast bus for a particular LANE client. The packets should instead go over a data direct vc. This is caused by a sender's default gateway being the remote BVI instead of the local BVI.
•
Symptom: In Front Panel View, if one member is removed from the Cluster, a particular device will be removed successfully, but others will show ports with cyan color, not reflecting the actual status of the ports.
Workaround: Let the Front Panel View refresh automatically or manual by pressing the refresh button or refresh menu item or close and reopen the Front Panel View window. Have to refresh the Cluster LED also along with the Cluster Data.
•
Symptom: The physical status might show that it is down while the IMA shows that it is up. This happens when there is a loss of frame (LOF) on the link. IMA dictates the state of the Receive Fault or Tranceive Fault on the link looking at the alarms OOCD, OOF, LOS and AIS. However, it does not need to look at LOF as a persisting LOF results in OOCD.
The physical status blindly looks for all the alarms on links to be 0. This is a cosmetic problem which will not affect the traffic on the link.
Workaround: None.
•
Symptom: IP addresses on shutdown interfaces might be programmed as invalid host addresses after entering the epc portstuck-manual-reload command. This does not impact functionality.
Workaround: None.
•
Symptom: The ATM router module might get channel stuck. This stops traffic from passing through MSR. This is often initially described as a "hung" condition.
Workaround: OIR the ATM router module.
•
Symptom: NCDP uses default address of the switch for uniquely identifying the root. The default address might get corrupted on a route processor switchover. If more than one redundant Catalyst 8540 is running NCDP then NCDP might not work as expected after a route processor switchover.
Workaround: None.
•
Symptom: ATM router module might set the CLP of all cells to 1 when bridging is configured over a 1483 PVC. This happens for cells being transmitted over the 1483 PVC only.
Workaround: None.
•
Symptom: After the switch has been running for over two hours it might start to lose traffic on its LANE/1483/1577 interfaces on the ATM router module. Frame interfaces (like Fast Ethernet) or non-LANE interfaces on the ATM router module (with configured pvc's) traffic passes without any problems. It doesn't matter if the traffic is going to the route processor not, part of the traffic is lost.
Workaround: OIR the module or reload the switch. Shutting down the ATM router module interfaces does not help.
•
Symptom: When using the Internetwork Packet Exchange (IPX), the output of the traceroute command skips the middle hop when the middle router is a Catalyst 85xx MSR, Catalyst 85xx CSR, or LightStream 1010.
Workaround: None.
•
Symptom: When global switch ATM address prefixes are added or deleted via the atm address command, addresses registered for existing CES circuits are not changed. Therefore, the goal of changing the switch address might not be achieved (in that CES does not change its addressing, and therefore will not accept calls using the new prefix).
Workaround: This workaround is for the primary switch address only. When the address is changed, the existing circuits can be deleted and created again, and the circuits will register addresses using the new primary switch address.
•
Symptom: The ATM router module might not function correctly over 20 M. When this occurs, the ATM router module loses packets or port gets stuck. This occurs even when multicast is not configured.
Workaround: None.
•
Symptom: When an enhanced Gigabit Ethernet interface module sends out ISL packets with packet sizes 1531 to 1548, the output error counter on the egress port increases.
Workaround: None.
•
Symptom: An 8-port Gigabit Ethernet interface module port that is shut down might come up after an OIR.
Workaround: Enter the shutdown command on the affected port.
Resolved Caveats—Cisco IOS Release 12.1(6)EY
Cisco IOS Release 12.1(6)EY is a rebuild release for Cisco IOS Release 12.1(5)EY1. The caveats in this section are resolved in Cisco IOS Release 12.1(6)EY but may be open in previous Cisco IOS releases.
•
Symptom: When video traffic is sent through an eight-port Gigabit Ethernet interface module traffic might be bursty because of packets being delivered out of order occasionally.
Workaround: None.
•
Symptom: Sending a crafted control NTP packet might cause a buffer overflow.
Workaround: Include the "ntp access-group serve-only" line in the configuration of the router.
•
Symptom: On an OC-48 interface module, the show controllers command might show the following:
"Port status (0x80001) - ".While normally it should report:"Port status (0x1) - no alarms"
This does not affect functionality.Workaround: Enter the shutdown command and then the no shutdown command on the interface.
•
Symptom: The Online Diagnostic Snake test, which is enabled by default, might send ILMI traps when an interface changes state to UP. This can cause clients on other interfaces to reset ILMI.
Workaround: Disable the snake test by entering the no diag online snake command.
•
Symptom: An ATM router module interface configured for transparent bridging and in blocking state might forward certain frames.
Workaround: Depending on topology, a work-around might be to ensure that a non-ATM router module interface is in blocking state. Where this is not feasible, power cycle the ATM router module interface that is in blocking state.
•
Symptom: ILMI might stay in the "waitdevtype" state after a redundancy failover. This is because on an OC3 port adapter, each of the 4 ports are served by one processor interface.
Workaround: The first port needs to be set for the other three ports to be able to receive. Sometimes this does not occur on a redundancy failover. Bring up port 0, if problem is seen on ports 1-3, port 4, if problem is seen on ports 5-7, port 8, if problem is seen on ports 9-11, and port 12, if problem is seen on ports 14-15. The ports can be brought up by connecting a loopback cable, or by simply using that port.
•
Symptom: When an IOS configuration for the boot loader is set to an image on a SanDisk PC Card, the subsequent reboot with a configuration register setting of 0x2102 fails with the following message:
cs22_c84_01#reloadSystem configuration has been modified. Save? [yes/no]: noProceed with reload? [confirm]System Bootstrap, Version 12.0(14)W5(19.29) INTERIM TEST SOFTWARECopyright (c) 2001 by cisco Systems, Inc.other cpu not present....global_lock not workingThis CPU is the PRIMARYCatalyst 8540 platform with 262144 Kbytes of main memoryopen: read error...requested 0x4 bytes, got 0xfffffff8trouble reading device magic numberloadprog: error - on file openopen: failed to find and/or load the bootloader: "disk1:cat8540c-in-mz.120-13.W5.19.bin"loadprog: error - on file openboot: cannot load "cisco2-Catalyst 8540"Workaround: None.
•
Symptom: When access is made to PNNI tables, a memory leak might occur. This only happens when an Network Management System (NMS) is running SNMP on the switch router.
Workaround: Turn off NMS polling of PNNI tables and use the commands instead for determining any PNNI information.
•
Symptom: An IP address might not be learned by peer switch when using ILMI-status.
Workaround: None.
•
Symptom: The port stuck message might not appear into system log.
Workaround: None.
•
Symptom: After performing an online insertion and removal of an OC48 port adapter, entering the show command and the no show commands on the OC-48 might generate the following error messages:
%ATMCORE-3-INTERNAL_ERROR: connAllocateHalfLeg: tag neighbor leg present%ATMCORE-3-INTERNAL_ERROR: atmCore_connSetupUniDirXconn: i/p Connection Leg ExistsWorkaround: Perform one more online insertion and removal and ensure that at least two minutes elapse between any two online insertions and removals.
•
Symptom: Telnet responses between two switch routers might be very slow.
Workaround: Replace one switch router with another that can accept ISL-frame less than 94B (for instance, the Catalyst 5000).
•
Symptom: When there is a lot of signaling svc set up and teardown on an nni tunnel interface, calls will fail with a vpi/vci collision.
Workaround: Use a non-tunnel interface.
•
Symptom: The switch router might lose traffic on its LANE or 1483 or 1577 interfaces on ATM router module. Frame interfaces or non-lane interfaces on ATM router module with configured pvc's should pass traffic without any problems. It doesn't matter if the traffic is going to RP or not, part of the traffic is lost.
Workaround: The switch router will recover from this state only after an OIR of the module or a reload.
•
Symptom: IP routing and IPX routing over BVI on an 8-port Gigabit Ethernet interface module might fail when the interface module is removed from the bridge group.
Workaround: None.
•
Symptom: When entering the show atm interface traffic command on an ATM subinterface, the total number of cells that has passed might show as zero. This rate is incorrect.
Workaround: None.
•
Symptom: On a Catalyst 8500 MSR performing an ATM Frame Relay internetworking function, when the Local Management Interface (LMI) indicates a permanent virtual connection (PVC) is inactive, this information is not propagated by the ATM Frame Relay code.
Workaround: None.
•
Symptom: Entering the shutdown command on the ATM router module interface might cause CPU hogs if a large number of PVCs are configured on that interface.
Workaround: None.
•
Symptom: ROMMON does not work on the Gigabit Ethernet interface module.
Workaround: None.
•
Symptom: When entering the show atm interface traffic command on an ATM subinterface, the total number of cells that has passed might show as zero. This rate is incorrect.
Workaround: None.
•
Symptom: HVPT: Incorrect scheduling values installed for low PCR. When a HVPT with a PCR of 2 kbps is configured on a OC-3 interface, this results in the cell starving of WKVC on the main interface.
Workaround: None.
•
Symptom: A Catalyst 8540, which correctly boots as a fully redundant system, does not complete a route processor failover. The secondary route processor, as it resumes function as the primary after a failover, displays the following error:
%Secondary route processor Detected remove_conn_from_fabric: Error close party failed for ...Workaround: None.
•
Symptom: The router reloads when handling internetworking packet exchange (IPX) access-logging-messages. This situation may still occur if access-logging is not enabled.
You may exceed the 200 IPX network limit when the hardware is activated after being removed. When the hardware is reactivated, the interfaces that were not counted while the hardware was removed cause the limit to be exceeded.
Workaround: None.
•
Symptom: The static bridging command disappears from the running configuration.
Workaround: None.
Resolved Caveats—Cisco IOS Release 12.1(5a)EY
Cisco IOS Release 12.1(5a)EY is a new build of a Cisco IOS Release. The caveats in this section are resolved in Cisco IOS Release 12.1(5a)EY but may be open in previous Cisco IOS releases.
•
Symptom: The switch experiences high route processor utilization when IPX networks are added to BVI interfaces.
Workaround: This is load related; as more IPX networks are added, route processor utilization increases. If configuring bridging over ATM router module subinterfaces, then configure a map-list for the bridged VC with the keyword bridge.
•
Symptom: The no ip route-cache cef command might disappear from the running configuration on BVI interfaces after reloading the switch.
Workaround: None.
•
Symptom: After performing a route processor switchover several times, and then reloading only the secondary route processor and booting up again, the primary route processor might display the following error messages:
00:45:25: %LINK-5-CHANGED: Interface ATM11/0/ima1, changed state to going down00:45:29: %LINK-3-UPDOWN: Interface ATM11/0/ima1, changed state to up00:48:56: wrong response present for the cmd given cmd: 0x35 resp : 0x3D00:49:37: %SUNI622: Read diagnostics for ATM3/0/1 failed(1)This does not impact functionality.
Workaround: None.
•
Symptom: If a static ARP is configured for one interface and then the IP address for that interface is moved to a subinterface, the static ARP entry remains associated with the original interface instead of with the subinterface.
Workaround: When the IP address is reconfigured to a subinterface off of the original interface, delete and reconfigure the ARP address also.
•
Symptom: An active Fast Ethernet interface might become administratively down after the route processor automatically switches over following entry of the test crash command.
Workaround: None.
•
Symptom: After switching over the route processor several times, the following message might display:
imapam_send_stored_config: config_status is UNIimapam_send_stored_config: config_status is UNIimapam_send_stored_config: config_status is UNIThis happens when an IMA port adapter is present in the chassis. Once the switch router boots up no functionality impact is seen.
Workaround: None.
•
Symptom: A system running PNNI with Cisco IOS Release 12.0(7)W5(15c) will reject an ATM call set up if the set up includes the 5a information element coded with associated signaling, explicit VPCI, and any VCI.
If another ATM switch sends a set up that includes the 5a information element coded with associated signaling, explicit VPCI, and any VCI, the system will release the call with cause code 35: requested VPCI/VCI not available.
Call set ups that include the 5a information element coded with the associated signaling, explicit VPCI, and explicit VCI signaling, including those made by Cisco ATM switches, are not affected by this problem. This interoperability issue exists between Cisco ATM switches and other vendor switches that do not explicitly request the VCI value in the call set up.
Workaround: None.
•
Symptom: The show hardware detail command does not show the actual version of the FPGA for any of the daughter cards. This shows the version of the FPGA motherboard which is misleading.
Workaround: None.
•
Symptom: If you save the running configuration, you may see the following error:
ROUTER# copy run startDestination filename [startup-config]?startup-config file open failed (File table overflow)This happens when you enter a write mem command or a copy running-config startup-config command. Enter a show file command to show the 127 entries of private configuration.
Workaround: Save the running configuration in a file, reload the router, and then enter a write mem command or copy running-config startup-config command.
•
Symptom: The Xconnect set up fails for unidirectional TVC if there is another one with a different direction and the same vpi/vci on a particular interface of the Catalyst 8540 MSR. This can be seen only if VC merge is disabled.
Workaround: None.
•
Symptom: If you set the ARP timers on an interface and replace the module, the startup-configuration is not loaded into the card. All ARP timers are reset to default.
Workaround: None.
•
Symptom: When the MRP reset button is pushed, the switch router might not boot the main image. If this happens, the switch router will eventually go into ROMMON mode.
Workaround: None.
•
Symptom: The output errors counter is incorrect on Gigabit Ethernet show interface command. They have a high value 4xxxxxxxxx and a clear counter command does not change it back to 0. This value can increase or decrease over time.
Workaround: None.
•
Symptom: The route processor redundancy might fail if you have both PVC and bridging configured on the subinterface of the ATM router module.
Workaround: None.
•
Symptom: The ATM router module does not strip the Ethernet pad when switching data from Ethernet to ATM. This might occasionally lead to connectivity issues since some end systems do not expect to see the pad.
Workaround: None.
•
Symptom: The ATM router module microcode for RFC1483 did not have the ability to understand CLNS topology updates, forcing the card to drop packets. This also affected the IS-IS routing updates.
Workaround: Upgrade to the Cisco IOS release 12.0(11)W5(19).
•
Symptom: There is no support for SONET MIB objects in an ATM uplink module.
Workaround: None.
•
Symptom: When reloading a new image into the enhanced 2-port Gigabit Ethernet interface module, the ATM uplink enhanced Gigabit interface appears to have been reset. The enhanced two-port Gigabit Ethernet interface module download shows no effect on the ATM uplink.
Workaround: None.
•
Symptom: With an encapsulation change on a 2-port Gigabit Ethernet interface module, the interface no longer sends packets. It seems that it is receiving packets and sending packets to the route processor when necessary, but the packets are not sent.
Workaround: None.
•
Symptom: When two 8540 CSR switch routers are directly connected by a 2-port Gigabit Ethernet interface module, a reset of one switch router does not cause the remote switch router's interfaces to go down too. The problem is caused by the reset switch router's interfaces not going down, so remote interfaces do not go down.
Workaround: Physically remove the cable from the ports or enter the shutdown command.
•
Symptom: In some instances, an ATM interface on a Catalyst 8540 MSR might get stuck in a going down state.
Processor utilization stays high for a long period of time. For example, for an 8K VC, the route processor stays high for approximately 720 seconds, and for 4K VCs, approximately 300 seconds.
Workaround: None.
•
Symptom: If configuring region sizes, you are not considering the ACL region size when checking for the total size. Configuration allows regions sizes exceeding total TCAM size. No error message is printed at the time of configuration but upon next boot you get the following error messages:
ACL cannot be enabled on GigabitEthernet0/0/0 - insufficient TCAMAvailable: 24 Configured for ACL: 2048 Use sdm access-list command to reconfigWorkaround: None.
•
Symptom: The ip cef command does not re-enable the CEF switching on the Catalyst 8540 when it was previously disabled due to lack of memory. The Catalyst 8540 requires CEF switching but this can become disabled if insufficient memory is available. When insufficient memory is available, the following message appears:
"%FIB-2-FIBDOWN: CEF has been disabled due to a low memory condition.Workaround: None.
•
Symptom: An ima_failure_trap might not be generated when an IMA group gets deleted and the group state change might not be sensed by the switch processor.
Workaround: None.
•
Symptom: On rare occasions the switch fails SVC set ups even when physical connectivity is established.
Workaround: Reload the switch.
•
Symptom: The root cause of this is that the HSRP specific structures are not properly updated when an interface goes down or when a card is removed.
Workaround: Remove HSRP from the configuration before removing the card.
Obtaining Documentation
Cisco documentation and additional literature are available on Cisco.com. Cisco also provides several ways to obtain technical assistance and other technical resources. These sections explain how to obtain technical information from Cisco Systems.
Cisco.com
You can access the most current Cisco documentation on the World Wide Web at this URL:
http://www.cisco.com/univercd/home/home.htm
You can access the Cisco website at this URL:
International Cisco websites can be accessed from this URL:
http://www.cisco.com/public/countries_languages.shtml
Ordering Documentation
You can find instructions for ordering documentation at this URL:
http://www.cisco.com/univercd/cc/td/doc/es_inpck/pdi.htm
You can order Cisco documentation in these ways:
•
http://www.cisco.com/en/US/partner/ordering/index.shtml
•
Documentation Feedback
You can submit e-mail comments about technical documentation to bug-doc@cisco.com.
You can submit comments by using the response card (if present) behind the front cover of your document or by writing to the following address:
Cisco Systems
Attn: Customer Document Ordering
170 West Tasman Drive
San Jose, CA 95134-9883We appreciate your comments.
Obtaining Technical Assistance
For all customers, partners, resellers, and distributors who hold valid Cisco service contracts, the Cisco Technical Assistance Center (TAC) provides 24-hour-a-day, award-winning technical support services, online and over the phone. Cisco.com features the Cisco TAC website as an online starting point for technical assistance. If you do not hold a valid Cisco service contract, please contact your reseller.
Cisco TAC Website
The Cisco TAC website provides online documents and tools for troubleshooting and resolving technical issues with Cisco products and technologies. The Cisco TAC website is available 24 hours a day, 365 days a year. The Cisco TAC website is located at this URL:
Accessing all the tools on the Cisco TAC website requires a Cisco.com user ID and password. If you have a valid service contract but do not have a login ID or password, register at this URL:
http://tools.cisco.com/RPF/register/register.do
Opening a TAC Case
Using the online TAC Case Open Tool is the fastest way to open P3 and P4 cases. (P3 and P4 cases are those in which your network is minimally impaired or for which you require product information.) After you describe your situation, the TAC Case Open Tool automatically recommends resources for an immediate solution. If your issue is not resolved using the recommended resources, your case will be assigned to a Cisco TAC engineer. The online TAC Case Open Tool is located at this URL:
http://www.cisco.com/tac/caseopen
For P1 or P2 cases (P1 and P2 cases are those in which your production network is down or severely degraded) or if you do not have Internet access, contact Cisco TAC by telephone. Cisco TAC engineers are assigned immediately to P1 and P2 cases to help keep your business operations running smoothly.
To open a case by telephone, use one of the following numbers:
Asia-Pacific: +61 2 8446 7411 (Australia: 1 800 805 227)
EMEA: +32 2 704 55 55
USA: 1 800 553-2447For a complete listing of Cisco TAC contacts, go to this URL:
http://www.cisco.com/warp/public/687/Directory/DirTAC.shtml
TAC Case Priority Definitions
To ensure that all cases are reported in a standard format, Cisco has established case priority definitions.
Priority 1 (P1)—Your network is "down" or there is a critical impact to your business operations. You and Cisco will commit all necessary resources around the clock to resolve the situation.
Priority 2 (P2)—Operation of an existing network is severely degraded, or significant aspects of your business operation are negatively affected by inadequate performance of Cisco products. You and Cisco will commit full-time resources during normal business hours to resolve the situation.
Priority 3 (P3)—Operational performance of your network is impaired, but most business operations remain functional. You and Cisco will commit resources during normal business hours to restore service to satisfactory levels.
Priority 4 (P4)—You require information or assistance with Cisco product capabilities, installation, or configuration. There is little or no effect on your business operations.
Obtaining Additional Publications and Information
Information about Cisco products, technologies, and network solutions is available from various online and printed sources.
•
http://www.cisco.com/go/marketplace/
•
http://cisco.com/univercd/cc/td/doc/pcat/
•
•
•
http://www.cisco.com/go/iqmagazine
•
•
http://www.cisco.com/en/US/learning/index.html
This document is to be used in conjunction with the documents listed in the "Related Documentation" section.
CCSP, CCVP, the Cisco Square Bridge logo, Follow Me Browsing, and StackWise are trademarks of Cisco Systems, Inc.; Changing the Way We Work, Live, Play, and Learn, and iQuick Study are service marks of Cisco Systems, Inc.; and Access Registrar, Aironet, ASIST, BPX, Catalyst, CCDA, CCDP, CCIE, CCIP, CCNA, CCNP, Cisco, the Cisco Certified Internetwork Expert logo, Cisco IOS, Cisco Press, Cisco Systems, Cisco Systems Capital, the Cisco Systems logo, Cisco Unity, Empowering the Internet Generation, Enterprise/Solver, EtherChannel, EtherFast, EtherSwitch, Fast Step, FormShare, GigaDrive, GigaStack, HomeLink, Internet Quotient, IOS, IP/TV, iQ Expertise, the iQ logo, iQ Net Readiness Scorecard, LightStream, Linksys, MeetingPlace, MGX, the Networkers logo, Networking Academy, Network Registrar, Packet, PIX, Post-Routing, Pre-Routing, ProConnect, RateMUX, ScriptShare, SlideCast, SMARTnet, StrataView Plus, TeleRouter, The Fastest Way to Increase Your Internet Quotient, and TransPath are registered trademarks of Cisco Systems, Inc. and/or its affiliates in the United States and certain other countries.
All other trademarks mentioned in this document or Website are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (0502R)
Copyright © 2005 Cisco Systems, Inc. All rights reserved.
Posted: Fri Apr 15 17:24:31 PDT 2005
All contents are Copyright © 1992--2005 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.
