cc/td/doc/product/lan/cat6000
hometocprevnextglossaryfeedbacksearchhelp
PDF

Table of Contents

Release Notes for Catalyst 6000 Family Multilayer Switch Module
Cisco IOS Release 12.0

Release Notes for Catalyst 6000 Family Multilayer Switch Module
Cisco IOS Release 12.0

September 21, 2000

Current Release: 12.0(10)W5(18d)

Caution ATM module requirement: The Multilayer Switch Module (MSM) software release version must be IOS release 12.0(5)W5(13a) or later to support the ATM module. These images, by default, will disable the ip proxy-arp functions on the MSM which cause ATM module download failures.

Previous Releases:
12.0(5)W5(13d)—Maintenance Release
12.0(5)W5(13c)—Maintenance Release
12.0(5)W5(13a)—Maintenance Release
12.0(4a)WX5(11)—Maintenance Release
12.0(1a)WX5(6g)—Maintenance Release
12.0(1a)WX5(6d)


These release notes describe the features, modifications, and caveats for the Catalyst 6000 family MSM. These release notes are applicable to the 12.0(10)W5(18d) Cisco IOS release and earlier releases. For features, modifications, and caveats for the Catalyst 6000 family supervisor engine software, refer to the Release Notes for Catalyst 6000 Family Software Release 5.x document.

Note The MSM ships with Cisco IOS software installed. However, before you can run this module in your Catalyst 6000 family switch, be sure that the Catalyst 6000 family supervisor engine is running supervisor engine software release 5.2(1)CSX or later. A copy of software release 5.2(1)CSX or later is provided in the enclosed DOS floppy disks. Software images are also available through Cisco Connection Online (CCO); see the "Cisco Connection Online" section for details.

Contents

This document consists of these sections:

MSM Memory Requirements

The following MSM default memory requirements are acceptable for all MSM images:

Software Ordering Information

Table 1 lists the images available for the MSM and their order numbers.


Table 1: Software Ordering Information
Filename Description Orderable Product Number
(Flash on System) 		Orderable Product Number
(Spare Upgrade—Floppy Media)

c6msm-in-mz.120-
10.W5.18d.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(10)W5(18d)

SFC6-MSM-12.0.5W

SFC6-MSMIN-12.0.5W=

c6msm-in-mz.120-
5.W5.13d.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(5)W5(13d)

SFC6-MSM-12.0.5W

SFC6-MSMIN-12.0.5W=

c6msm-in-mz.120-
5.W5.13c.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(5)W5(13c)

SFC6-MSM-12.0.5W

SFC6-MSMIN-12.0.5W=

c6msm-in-mz.120-
5.W5.13a.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(5)W5(13a)

SFC6-MSM-12.0.4W

SFC6-MSMIN-12.0.4W=

c6msm-in-mz.120-
4a.WX5.11.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(4a)WX5(11)

SFC6-MSM-12.0.3W

SFC6-MSM-12.0.3W=

c6msm-in-mz.120-
1a.WX5.6g.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(1a)WX5(6g)

SFC6-MSM-12.0.2W

SFC6-MSM-12.0.2W=

c6msm-in-mz.120-
1a.WX5.6d.bin

Catalyst 6000 MSM IP/IP-Multicast Routing Feature Set Release 12.0(1a)WX5(6d)

SFC6-MSM-12.0.1W

SFC6-MSM-12.0.1W=

Catalyst 6000 MSM IPX Feature Set License

FRC6-MSM-IPX

FRC6-MSM-IPX=

Features in Release 12.0(10)W5(18d)

There are no new features in Release 12.0(10)W5(18d).

Features in Release 12.0(5)W5(13d)

There are no new features in Release 12.0(5)W5(13d).

Features in Release 12.0(5)W5(13c)

There are no new features in Release 12.0(5)W5(13c).

Features in Release 12.0(5)W5(13a)

There are no new features in Release 12.0(5)W5(13a).

Features in Release 12.0(4a)WX5(11)

There are no new features in Release 12.0(4a)WX5(11).

Features in Release 12.0(1a)WX5(6g)

There are no new features in Release 12.0(1a)WX5(6g).

Features in Release 12.0(1a)WX5(6d)

Refer to the Catalyst 6000 and 6500 Series Multilayer Switch Module Installation and Configuration Note for detailed information on all Release 12.0(1a)WX5(6d) features.

Usage Guidelines and Restrictions

This section provides usage guidelines and restrictions for using the MSM.

Router (config)# router eigrp 109
Router (config-router)# maximum-paths 2
 

%AAL5-3-INTERNAL_ERROR: No more big aal5 pkts
 

%Error opening tftp://255.255.255.255/network-confg (Not enough space)
%Error opening tftp://255.255.255.255/cisconet.cfg (Not enough space)
%Error opening tftp://255.255.255.255/router-confg (Not enough space)
%Error opening tftp://255.255.255.255/ciscortr.cfg (Not enough space)
To disable the service configuration, go to enable mode and enter a no service config command and save to NVRAM.

Open and Resolved Caveats in Release 12.0(10)W5(18d)

These sections describe open and resolved caveats in Release 12.0(10)W5(18d):

Open Caveats in Release 12.0(10)W5(18d)

This section describes open caveats in Release 12.0(10)W5(18d) for the MSM.

The scheduler allocate command is not supported on the MSM.
Symptom: Unpredictable results might occur if upi change the CPU scheduler allocation in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. We recommend that you do not change the default values for this command.
The default spanning tree path cost is incorrect on Gigabit Ethernet interfaces.
Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.

Resolved Caveats in Release 12.0(10)W5(18d)

This section describes resolved caveats in Release 12.0(10)W5(18d) for the MSM.

A CPUHOG message appears when activating OSPF on a large port channel configuration.
Symptom: A CPUHOG message appears when OSPF is activated on a port channel interface with more than 30 subinterfaces.
Workaround: The operation succeeds even if this message appears. No action is necessary.
MSM will not boot from supervisory slot0 if a loopback interface is present.
Symptom: When a loopback interface is present the MSM does not boot..
Workaround: The MSM now boots from sup-slot0 even when a loopback interface exists.

Open and Resolved Caveats in Release 12.0(5)W5(13d)

These sections describe open and resolved caveats in Release 12.0(5)W5(13d):

Open Caveats in Release 12.0(5)W5(13d)

This section describes open caveats in Release 12.0(5)W5(13d) for the MSM.

The scheduler allocate command is not supported on the MSM.
Symptom: Unpredictable results might occur if you change the CPU scheduler allocation in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. We recommend that you do not change the default values for this command.
A CPUHOG message appears when activating OSPF on a large port channel configuration.
Symptom: A CPUHOG message appears when OSPF is activated on a port channel interface with more than 30 subinterfaces.
Workaround: The operation succeeds even if this message appears. No action is necessary.
The default spanning tree path cost is incorrect on Gigabit Ethernet interfaces.
Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.

Resolved Caveats in Release 12.0(5)W5(13d)

This section describes resolved caveats in Release 12.0(5)W5(13d) for the MSM.

Symptom: A defect in multiple versions of Cisco IOS software will cause a Cisco router or switch to stop and reload if the Cisco IOS http service is enabled and an attempt is made to browse to http://<router-ip. This defect can be exploited to produce a denial of service (DoS) attack. This defect has been discussed on public mailing lists and should be considered public information.
The vulnerability, identified as Cisco bug ID CSCdr36952, affects virtually all mainstream Cisco routers and switches running Cisco IOS Release 11.1 through Release 12.1. The vulnerability has been corrected and Cisco is making fixed versions available to replace all affected Cisco IOS releases. Customers are urged to upgrade to releases that are not vulnerable to this defect.
Workaround: Disable the Cisco IOS HTTP server by preventing access to the port in use by the HTTP server on the affected router or switch or apply an access-class option to the service itself. The IOS HTTP server is not enabled by default except on a small number of router models in specific circumstances.
Refer to http://www.cisco.com/warp/public/707/ioshttpserver-pub.shtml for the latest complete version of this security advisory.

Open and Resolved Caveats in Release 12.0(5)W5(13c)

These sections describe open and resolved caveats in Release 12.0(5)W5(13c):

Open Caveats in Release 12.0(5)W5(13c)

This section describes open caveats in Release 12.0(5)W5(13c) for the MSM.

The scheduler allocate command is not supported on the MSM.
Symptom: Unpredictable results might occur if you change the CPU scheduler allocation in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. We recommend that you do not change the default values for this command.
A CPUHOG message appears when activating OSPF on a large port channel configuration.
Symptom: A CPUHOG message appears when OSPF is activated on a port channel interface with more than 30 subinterfaces.
Workaround: The operation succeeds even if this message appears. No action is necessary.
The default spanning tree path cost is incorrect on Gigabit Ethernet interfaces.
Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.

Resolved Caveats in Release 12.0(5)W5(13c)

This section describes resolved caveats in Release 12.0(5)W5(13c) for the MSM.

Symptom: The show running-config command does not show anything. This problem is not related to the privilege level or TACACS configuration. When the running-config file is uploaded to a TFTP server, a 0-length file is created.
If snmpwalk is used against the MSM, snmpwalk returns 0.0.0.0 for the MSM module IPAddress.
After a switchover to the redundant supervisor engine, the MSM should re-register its IP address with the supervisor engine. However, if snmpwalk is used against the supervisor engine after the switchover, snmpwalk returns 0.0.0.0 as the management IP address.
Encapsulation dot1q native is not working. A subinterface configured as native sends untagged packets to the supervisor engine. The supervisor engine software specifically drops all untagged packets for the MSM if configured for dot1q. To make this work, a new version of supervisor engine software is required that forwards all untagged (native) dot1q packets to the supervisor engine.
Workaround: Send tagged packets (non-native packets) to the supervisor engine. Creating a subinterface on an unused VLAN and marking it as native will solve the problem.

Open and Resolved Caveats in Release 12.0(5)W5(13a)

These sections describe open and resolved caveats in Release 12.0(5)W5(13a):

Open Caveats in Release 12.0(4)WX5(13a)

This section describes open caveats in Release 12.0(5)W5(13a) for the MSM.

The scheduler allocate command is not supported on the MSM.
Symptom: Unpredictable results might occur if the CPU scheduler allocation is changed in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. We recommend that you do not change the default values for this command.
A CPUHOG message appears when activating OSPF on a large port channel configuration.
Symptom: A CPUHOG message appears when OSPF is activated on a port channel interface with more than 30 subinterfaces.
Workaround: The operation succeeds even if this message appears. No action is necessary.
The default spanning tree path cost is incorrect on Gigabit Ethernet interfaces.
Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.

Resolved Caveats in Release 12.0(5)W5(13a)

This section describes resolved caveats in Release 12.0(5)W5(13a) for the MSM.

No reassembly buffers to receive pkt, vpi 0, vci 36.
Symptom: On a system with over 10,000 IP multicast routes, multiple "No reassembly buffers to receive pkt, vpi 0, vci 36" messages are displayed.
Multicast configuration on MSM breaks IP connectivity.
Symptom: If you enable multicast routing on an MSM configured with four separate trunk ports, and any multicast group has two or more subinterfaces in its outgoing list, then IP connectivity between any subinterfaces that share the same physical interface with this multicast traffic will be intermittent. The problem also occurs if the MSM interfaces are configured as a port channel trunk.
Symptom: If you configure BVI, but do not want to do IP routing at the BVI level, and you have IP addresses for the interfaces in the bridge group, then routing protocols might not work.
A blocked physical port shows the learned entry.
Symptom: On a router with a rev-B1 EPIF, there is no way you can to turn off time-stamp refresh during a destination address lookup. A source address is learned on a port in the initial spanning tree state. After the port goes into blocking state, a destination address lookup on the received packet keeps the entry alive. This situation occurs when the source and destination address are the same, as in the case of keepalive packets.
Large bridge group configurations with IRB lead to AAL5 buffer exhaustion.
Symptom: When you configure a large number of bridge groups or bridge group members and enable IRB on the devices, reloading one of the devices or configurations might lead to high CPU utilization. The side effect is temporary spanning tree loops, leading to AAL5 buffer exhaustion problems.
Removing adjacencies when bridge entry ages might lead to high CPU utilization.
Symptom: Adjacencies are removed when a bridge table entry ages out. This situation causes traffic to be routed to the CPU, which might lead to high CPU utilization.
CPUHOG messages appear on a 16 EtherChannel subinterfaces configuration.
Symptom: With 16 or more subinterfaces on a port channel, the following configuration might cause CPUHOG messages to appear:
The show controller command can bring down an interface.
Symptom: Occasionally, the show controller interface-name command can bring down an interface and the line protocol. The condition that leads to this scenario is random and infrequent. The end result is that routing/bridging stops on this interface.
The interface packet counter incorrectly doubles packet count.
Symptom: In some cases, the interface packet counter counts each terminating IP packet (a packet sent to the SRP) twice. When you enter a show interface command, the reported number of total packets received is double the actual packet count.
A warning message appears when you enter the ospf disable command.
Symptom: If you enable the OSPF protocol and enter an ospf disable command, you receive a warning message that the OSPF process is taking too much CPU (SRP) processing power.
Warning message when issuing the clear IP route command.
Symptom: If the Catalyst 6000 family switch has more than 1,024 routes in its routing tables when you issue a clear ip route command, you receive a warning message that the process is taking too much CPU (switch route processor [SRP]) processing power.
The Cisco Discovery Protocol (CDP) does not work when Inter-Switch Link (ISL) over Fast EtherChannel (FEC) is configured on the system.
Symptom: If you have CDP configured on a FEC interface that has ISL configured on it, you cannot see the neighbor information when you enter a show cdp neighbor command.
Symptom: Entering a clear ip route network command deletes routes to directly connected networks.
IP multicast TTL threshold does not affect packet forwarding.
Symptom: An IP multicast TTL threshold will have no effect on a port-channel interface with an ISL encapsulation; the IP multicast TTL threshold is not analyzed and the packet is forwarded.
CPU utilization is very high (up to 100 percent) for no obvious reason.
Symptom: Certain types of traffic, including routing protocol updates and traffic terminating at the switch route processor, can generate extremely high CPU traffic.
Spurious INTERNAL_ASSERT messages appear when the system boots.
Symptom: These messages are produced by the Cisco IOS Release 12.0(0.6)W5(1) boot loader.
CPUHOG messages appear after a shutdown and no shutdown command sequence on an interface.
Symptom: CPUHOG messages appear if more than 1,024 routes are tied to an interface when you shut it down. Because of the large number of routes, the CPU is occupied by the IPX process longer than normal.
%ALIGN-3-CORRECT: Alignment correction made at 0x6017C6C4.
Symptom: These alignment errors show up when a Telnet session is opened that uses TCP. The errors are harmless.
The clear arp command does not delete adjacency entries.
Symptom: The Catalyst 6000 family MSM uses adjacency entries internally, so it does not zero out the MAC addresses for the corresponding IP addresses when clear arp is executed. This causes a problem if a MAC address corresponding to an IP address changes and the new host does not initiate an ARP request.
BVI does not forward packets after a node is moved.
Symptom: When a node is moved to a different interface, the bridge-group virtual interface (BVI) does not forward packets intended for that node.
Catalyst 6000 family switch does not accept untagged frames for 802.1Q trunking.
Symptom: When the Catalyst 6000 family switch is configured with subinterfaces to use encapsulation dot1q vlan-tag trunking, the native VLAN assigned to the directly connected Catalyst switch (such as the Catalyst 4000 or Catalyst 5000), will not function. The Catalyst 6000 family switch currently does not process untagged packets received on an 802.1Q trunked interface, so all packets are dropped.
A device might lose IP OSPF neighbors and connectivity over BVI interfaces.
Symptom: A device might lose its IP OSPF neighbor information and connections over a BVI interface after a few days of running, requiring a system reboot.
A device might lose IPX connectivity over a BVI interface.
Symptom: A device might lose IPX connectivity over a BVI interface, requiring the use of the clear ipx route command to reestablish connectivity.
Subsequent IPX pings fail after a ping with a timeout value of zero.
Symptom: After executing an IPX ping with a timeout value of zero, subsequent IPX pings on the device will fail.
An HSRP MAC appears as a "remote" entry in the CAM table for an active router.
Symptom: When a peer switch with Uplink Fast enabled is connected to a Catalyst 6000 family MSM bridge group interface, the virtual MAC address of the HSRP for an active device in one of the member interfaces of the bridge group is programmed as a remote entry.
%IPX-3-BADINSERT: Duplicate SAP entry insert attempted.
Symptom: The router returns a %IPX-3-BADINSERT message when a duplicate SAP entry is made.

Open and Resolved Caveats in Release 12.0(4a)WX5(11)

These sections describe open and resolved caveats in Release 12.0(4a)WX5(11):

Open Caveats in Release 12.0(4a)WX5(11)

This section describes open caveats in Release 12.0(4a)WX5(11) for the MSM.

Symptom: Changing the MAC address for a bridge group member causes CPU-bound packets on that interface to be discarded. For a port channel, the problem occurs when the first member is removed from the port channel.
Workaround: When an interface's MAC address is changed, remove the interface from its bridge group and then add it back into the bridge group.
Symptom: If you configure BVI, but do not want to do IP routing at the BVI level, and you have IP addresses for the interfaces in the bridge group, then routing protocols might not work.
Workaround: Remove the interfaces from the bridge group and then add them back into the bridge group.
Symptom: When a FEC is configured as a part of a bridge group and a topology change occurs, the bridge table may get out of sync with the Layer 2 CAM information.
Workaround: Use the clear bridge command to keep the IOS bridge table and the Layer 2 CAM in sync.

Resolved Caveats in Release 12.0(4a)WX5(11)

This section describes resolved caveats in Release 12.0(4a)WX5(11) for the MSM.

Symptom: Although the VTY login password is configured in the startup configuration of the MSM, the MSM does not prompt for the VTY login password when you access the MSM console through either session or telnet.

Open and Resolved Caveats in Release 12.0 (1a) WX5 (6g)

These sections describe open and resolved caveats in Release 12.0(1a)WX5(6g):

Open Caveats in Release 12.0(1a)WX5(6g)

This section describes open caveats in Release 12.0(1a)WX5(6g).

Symptom: Although the VTY login password is configured in the startup configuration of the MSM, the MSM does not prompt for the VTY login password when you access the MSM console through either session or Telnet.
Workaround: After the MSM comes up, you can manually get into the MSM console and enter a copy-startup-config running-config command. This has to be done each time the MSM is rebooted.
Workaround: Enter the clear ipx eigrp neighbor command to recover.
Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.
Symptom: A CPUHOG message appears when OSPF is activated on a port-channel interface with more than 30 subinterfaces.
Workaround: The operation will succeed even if this message appears. No action is necessary.
Symptom: Unpredictable results might occur if the CPU scheduler allocation is changed in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. Do not issue any scheduler allocate commands; using the scheduler allocate command could cause problems with network interrupt throttling.
Symptom: Occasionally, the show controller interface-name command can bring down an interface and the line protocol. The condition that leads to this scenario is random and infrequent. The end result is that routing/bridging stops on this interface.
Workaround: None.
Symptom: With 16 or more subinterfaces on a port channel, the following configuration might cause CPUHOG messages to appear:
Workaround: No workaround is needed; the CPUHOG messages should not affect the functionality of the device.
Symptom: Adjacencies are removed when a bridge table entry ages out. This situation causes traffic to route to the CPU, which might lead to high CPU utilization.
Workaround: Set the ARP timeout value to be less than the bridge age timer.
Symptom: When a peer switch with Uplink Fast enabled is connected to an MSM bridge group interface, the virtual MAC address of the HSRP for an active device in one of the member interfaces of the bridge group is programmed as a "remote" entry.
Workaround: Enter the clear bridge command to update the CAM table.
Symptom: When a port channel is added to an existing bridge group, the Cisco IOS software does not program the existing MAC addresses already in that bridge group as "remote" entries in the newly added port channel members. This leads to the port channel members going out of sync with regard to the "remote" entries, causing flooding in one direction.
Workaround: None.
Symptom: When a large number of bridge groups or bridge group members are configured and IRB is enabled on the devices, reloading one of the devices or configurations may lead to high CPU utilization. The side effect of this situation is temporary spanning tree loops, leading to AAL5 buffer exhaustion problems.
Workaround: In this case, if IRB is enabled, you may need to increase the number of buffers on the device using the aal5 buffers command.
Symptom: When a packet generator sends heavy traffic to the MSM and does not have a route to the next hop router or destination, the console can be sluggish or even freeze and the CPU utilization might be very high. Also related to heavy traffic is the problem that occurs when using the shut interface and no shut interface commands.
Both caveats occur when a packet generator is used to blast traffic at the MSM, and either there is a misconfiguration or the routes have been cleared for some other reason. In such a case, the packet processing engines forward these unknown-destination packets to the main CPU causing high CPU utilization.
Workaround: Reduce or stop packet generator traffic and correct the misconfiguration. Resume traffic only after the MSM has learned the routes.
Symptom: In some cases, if you have two IP multicast routers within the same IP subnet and the MSM stays in Prune State, IP multicast traffic received by the MSM might be delivered to the MSM CPU causing high CPU utilization.
We recommend that you use PCMCIA slot0 on the supervisor engine for storing multiple versions of system images and configuration files for the MSM as follows:

Resolved Caveats in Release 12.0(1a)WX5(6g)

This section describes resolved caveats in Release 12.0(1a)WX5(6g).

This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: In the MSM, the EIGRP protocol computations are done by the CPU and the computed routes are fed to the switching engines. For a large number of EIGRP neighbors and routes, this process is extremely computation intensive.
Solution: The optimizations in this release substantially improve this performance and convergence as compared to the previous MSM release. See the "Usage Guidelines and Restrictions" section for EIGRP configuration guidelines.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: If multiple subinterfaces of a port channel are outgoing interfaces of the same multicast group, and one subinterface has been disabled for multicasting, the data packets might continue to be forwarded to that subinterface. If this subinterface is reenabled for multicasting, it might receive duplicate multicast data packets. This situation is due to an improper checking of the subinterface type being disabled for multicasting. As a result, the oldest member subinterface is disabled; this interface might not be the one being modified.
Solution: The software that checks the type of the outgoing interface or subinterface being deleted has been fixed. The correct port channel subinterface will be removed from the outgoing interface list and the multicast data packets are stopped from being forwarded to this member.
The network interrupt throttling enhancement added in Release 12.0(1a)WX5(6g) fixes caveat CSCdm21851 provided you use the following implementation procedure. You must perform this procedure prior to doing any configuration of the MSM.
Symptom: If the MSM continues to receive heavy traffic that has to be processed by the CPU, most of the CPU time is used to handle the interrupts generated by the switch fabric, leaving very limited time for the processes to process the packets. As a result, the routing table cannot be updated. You might lose control of the console and the system appears to hang at the time when the CPU is busy handling the interrupts. The HSRP might also fail to work.
Solution: Network interrupt throttling limits the time spent on handling interrupts generated by the switch fabric so that the processes get a chance to receive and process the packets and update the routing table. You will be able to get the console prompt and enter commands even when the MSM is receiving heavy traffic that needs to be processed by the CPU.
To implement network interrupt throttling: The previous release of the MSM software does not contain this enhancement and the system, on bootup, writes a command line "no sched alloc" into the configuration file and stores it in NVRAM. You need to remove the command line "no sched alloc" in order to turn on network interrupt throttling.
Perform these steps to implement network interrupt throttling:

Step 1  Enter the show startup command from the console in enable mode to check if the command line "no sched alloc" appears near the bottom of the configuration file. If this command line is absent, nothing needs to be done, and you can skip the following steps. Otherwise, continue with the following steps.

Step 2  Add the command line "sched alloc" to the running configuration and save the configuration to NVRAM.

Step 3  Reload the MSM with the new release of the MSM software.

Step 4  After the system boots up, enter the show startup and show running commands from the console in enable mode to verify that the "no sched alloc" command line is not there. Whenever you upgrade the MSM software from Release 12.0(1a)WX5(6d) to Release 12.0(1a)WX5(6g), you need to perform the above steps once in order to turn on network interrupt throttling. When you have turned on network interrupt throttling, it will automatically turn on upon power up and reload. When you copy an old configuration file onto the MSM running the new release, make sure that you remove any "no sched alloc" command lines that might be in the configuration file before you copy it to the running or the startup configuration.

This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: When the MSM is configured with subinterfaces to use encapsulation dot1q (VLAN-tag) trunking, the native VLAN assigned to the switch will not function. The MSM currently does not process untagged packets received on a dot1q trunked interface, so all packets are dropped.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: When a node is moved to a different interface, the bridge-group virtual interface (BVI) does not forward packets intended for that node.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: The MSM uses adjacency entries internally, so it does not zero out the MAC addresses for the corresponding IP addresses when the clear arp command is executed. This situation causes a problem if a MAC address corresponding to an IP address changes and the new host does not initiate an ARP request.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: After executing an IPX ping with a timeout value of zero, subsequent IPX pings on the device will fail.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: A device may lose IPX connectivity over a BVI interface, requiring the use of the clear ipx route command to reestablish connectivity.
This problem is fixed in software Release 12.0(1a)WX5(6g).
Symptom: The device may lose its IP OSPF neighbor information and connections over a BVI interface after a few days of running, requiring a system reboot.

Open and Resolved Caveats in Release 12.0(1a)WX5(6d)

These sections describe open and resolved caveats in Release 12.0(1a)WX5(6d).

Open Caveats in Release 12.0(1a)WX5(6d)

This section describes open caveats in Release 12.0(1a)WX5(6d).

Symptom: Path costs on Gigabit Ethernet interfaces shown by the show span command do not match the expected default value of 100.
Workaround: Set the path costs manually. One option would be to set the path cost for Gigabit EtherChannel (GEC) ports to 1 and set the path cost for the non-GEC ports to 2.
Symptom: A CPUHOG message appears when OSPF is activated on a port-channel interface with more than 30 subinterfaces.
Workaround: The operation will succeed even if this message appears. No action is necessary.
Symptom: Unpredictable results might occur if you change the CPU scheduler allocation in the configuration file.
Workaround: Remove the scheduler allocate command from the configuration file. Do not issue any scheduler allocate commands; using the scheduler allocate command could cause problems with network interrupt throttling.
Symptom: When a packet generator sends heavy traffic to the MSM, and it does not have a route to the next hop router or destination, the console can be sluggish or even freeze and the CPU utilization might be very high. Also related to heavy traffic is the problem that occurs when using the shut interface and no shut interface commands.
Both caveats occur when a packet generator is used to blast traffic at the MSM, and either there is a misconfiguration or the routes have been cleared for some other reason. In such a case, the packet processing engines forward these unknown-destination packets to the main CPU causing high CPU utilization.
Workaround: Reduce or stop packet generator traffic and correct the misconfiguration. Resume traffic only after the MSM has learned the routes.
Symptom: In some cases, if you have two IP multicast routers within the same IP subnet and the MSM stays in Prune State, IP multicast traffic received by the MSM might be delivered to the MSM CPU causing high CPU utilization.
We recommend that you use PCMCIA slot0 on the supervisor engine for storing multiple versions of system images and configuration files for the MSM as follows:

Resolved Caveats in Release 12.0(1a)WX5(6d)

This section describes resolved caveats in Release 12.0(1a)WX5(6d).

This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: When the MSM is configured with subinterfaces to use encapsulation dot1q (VLAN-tag) trunking, the native VLAN assigned to the switch will not function. The MSM currently does not process untagged packets received on a dot1q trunked interface, so all packets are dropped.
Workaround: Assign the native VLAN on the switch to a nonexistent VLAN number. Make sure that the sc0 interface is not part of this VLAN. For example, if VLAN 999 does not exist anywhere in the network, create VLAN 999 and assign it as the native VLAN for the trunked port. (A native VLAN is the VLAN the port would be in when it is not trunking.) You can verify the trunk state by entering the show port trunk mod_num command. Note that if the port is already trunking, you must bring the trunk down to change the native VLAN.
This problem is fixed in Release 12.0(1a)WX5(6g).
Symptom: The MSM uses adjacency entries internally, so it does not zero out the MAC addresses for the corresponding IP addresses when the clear arp command is executed. This causes a problem if a MAC address corresponding to an IP address changes and the new host does not initiate an ARP request.
Workaround: Enter the clear adjacencies command and the clear arp command to clear aged ARP entries in the adjacency table.

Additional Documentation

For additional information on Catalyst 6000 family switches and command-line interface (CLI) commands, refer to the Catalyst 6000 Family Software Configuration Guide and the Catalyst 6000 and 6500 Family Command Reference publication.

For detailed hardware configuration and maintenance procedures, refer to the Catalyst 6000 Family Installation Guide.

For additional information on Cisco IOS commands, refer to the Configuration Fundamentals Command Reference publication.

Cisco Connection Online

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

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

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

You can access CCO in the following ways:

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

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

Documentation CD-ROM

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

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





hometocprevnextglossaryfeedbacksearchhelp
Posted: Thu Sep 21 11:18:41 PDT 2000
Copyright 1989-2000©Cisco Systems Inc.