Table Of Contents
Caveats for Cisco IOS Release 12.1
Resolved Caveats—Cisco IOS Release 12.1(27b)
Resolved Caveats—Cisco IOS Release 12.1(27a)
Open Caveats—Cisco IOS Release 12.1(27)
Resolved Caveats—Cisco IOS Release 12.1(27)
Resolved Caveats—Cisco IOS Release 12.1(26)
Resolved Caveats—Cisco IOS Release 12.1(25)
Resolved Caveats—Cisco IOS Release 12.1(24)
Resolved Caveats—Cisco IOS Release 12.1(22c)
Resolved Caveats—Cisco IOS Release 12.1(22b)
Resolved Caveats—Cisco IOS Release 12.1(22a)
Resolved Caveats—Cisco IOS Release 12.1(22)
Resolved Caveats—Cisco IOS Release 12.1(21)
Resolved Caveats—Cisco IOS Release 12.1(20a)
Resolved Caveats—Cisco IOS Release 12.1(20)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(19)
Resolved Caveats—Cisco IOS Release 12.1(18)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(17)
Resolved Caveats—Cisco IOS Release 12.1(16)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(15)
Resolved Caveats—Cisco IOS Release 12.1(14)
Resolved Caveats—Cisco IOS Release 12.1(13a)
Resolved Caveats—Cisco IOS Release 12.1(13)
Resolved Caveats—Cisco IOS Release 12.1(12c)
Resolved Caveats—Cisco IOS Release 12.1(12b)
Resolved Caveats—Cisco IOS Release 12.1(12a)
Resolved Caveats—Cisco IOS Release 12.1(12)
Resolved Caveats—Cisco IOS Release 12.1(11b)
Resolved Caveats—Cisco IOS Release 12.1(11a)
Resolved Caveats—Cisco IOS Release 12.1(11)
Resolved Caveats—Cisco IOS Release 12.1(10a)
Resolved Caveats—Cisco IOS Release 12.1(10)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(9a)
Resolved Caveats—Cisco IOS Release 12.1(9)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(8a)
Resolved Caveats—Cisco IOS Release 12.1(8c)
Resolved Caveats—Cisco IOS Release 12.1(8)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(7c)
Resolved Caveats—Cisco IOS Release 12.1(7b)
Resolved Caveats—Cisco IOS Release 12.1(7a)
Resolved Caveats—Cisco IOS Release 12.1(7)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(6a)
Resolved Caveats—Cisco IOS Release 12.1(6)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(5e)
Resolved Caveats—Cisco IOS Release 12.1(5d)
Resolved Caveats—Cisco IOS Release 12.1(5c)
Resolved Caveats—Cisco IOS Release 12.1(5)
Resolved Caveats—Cisco IOS Release 12.1(4c)
Resolved Caveats—Cisco IOS Release 12.1(4a)
Resolved Caveats—Cisco IOS Release 12.1(4)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(3b)
Resolved Caveats—Cisco IOS Release 12.1(3)
Resolved Caveats—Cisco IOS Release 12.1(2b)
Resolved Caveats—Cisco IOS Release 12.1(2)
Novell IPX, XNS, and Apollo Domain
Resolved Caveats—Cisco IOS Release 12.1(1c)
Resolved Caveats—Cisco IOS Release 12.1(1)
Novell IPX, XNS, and Apollo Domain
Cisco Product Security Overview
Reporting Security Problems in Cisco Products
Obtaining Technical Assistance
Cisco Technical Support & Documentation Website
Definitions of Service Request Severity
Obtaining Additional Publications and Information
Caveats for Cisco IOS Release 12.1
November 2, 2005
Online Part Number OL-2897-11 Rev. D0
This document lists severity 1 and 2 caveats and select severity 3 caveats for Cisco IOS Release 12.1, up to and including Release 12.1(27b). Caveats describe unexpected behavior or defects in Cisco IOS software releases. Severity 1 caveats are the most serious caveats; severity 2 caveats are less serious.
To help us improve this document, please send us 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 contact relnote-feedback@cisco.com. For more information, see the "Documentation Feedback" section.
How to Use This Document
This document describes open and resolved severity 1 and 2 caveats and select severity 3 caveats:
•
The "Open Caveats" section lists open caveats that apply to the current release and may apply to previous releases.
•
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 IOS software documentation can be found on the web through Cisco.com.
For more information on caveats and features in Cisco IOS Release 12.1, refer to the following sources:
•
•
•
•
•
•
Note
The most recent release notes document when this caveats document was published were Release Notes for Cisco IOS Release 12.1 for Cisco IOS Release 12.1(26) on November 11, 2004.
Contents
•
•
•
Resolved Caveats—Cisco IOS Release 12.1(27b)
Cisco IOS Release 12.1(27b) is a rebuild release for Cisco IOS Release 12.1(27). The caveats in this section are resolved in Cisco IOS Release 12.1(27b) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
IP Routing Protocols
•
Symptoms: A router may reset its Border Gateway Protocol (BGP) session.
Conditions: This symptom is observed when a Cisco router that peers with other routers receives an Autonomous System (AS) path with a length that is equal to or greater than 255.
Workaround: Configure the bgp maxas limit command in such as way that the maximum length of the AS path is a value below 255. When the router receives an update with an excessive AS path value, the prefix is rejected and recorded the event in the log.
Miscellaneous
•
Symptoms: The Multiprotocol Label Switching (MPLS) forwarding table, label forwarding information base (LFIB), is not properly populated.
Conditions: This symptom is observed when reloading a Cisco router that is running Cisco IOS Release 12.1, Release 12.2, or Release 12.1 E.
Workaround: For every router (prefix/mask) with routes for which the LFIB entries are incorrect, execute the clear ip route prefix mask command.
Alternate Workaround: Execute the clear ip route * command.
•
Symptoms: A router may stop sending traffic.
Conditions: This symptom is observed after a Stateful Switchover (SSO) cutover is performed on a Cisco 7500 series that has label-controlled ATM (LC-ATM) incoming and outgoing interfaces and that is configured with distributed Cisco Express Forwarding (dCEF).
Workaround: There is no workaround.
•
Symptoms: A label may not be assigned for a peer provider edge (PE) router.
Conditions: This symptom is observed on a Cisco 7500 series and a Cisco 12000 series in a Virtual Private Network (VPN) configuration with multiple route reflectors (RRs) and label controlled ATM (LC-ATM) links between PE routers. The symptom may also occur on other platforms.
Workaround: There is no workaround.
•
Cisco IOS may permit arbitrary code execution after exploitation of a heap-based buffer overflow vulnerability. Cisco has included additional integrity checks in its software, as further described below, that are intended to reduce the likelihood of arbitrary code execution.
Cisco has made free software available that includes the additional integrity checks for affected customers.
This advisory is posted at http://www.cisco.com/warp/public/707/cisco-sa-20051102-timers.shtml.
Resolved Caveats—Cisco IOS Release 12.1(27a)
Cisco IOS Release 12.1(27a) is a rebuild release for Cisco IOS Release 12.1(27). The caveats in this section are resolved in Cisco IOS Release 12.1(27a) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
Miscellaneous
•
NetFlow Feature Acceleration has been deprecated and removed from Cisco IOS. The global command ip flow-cache feature-accelerate will no longer be recognized in any IOS configuration.
If your router configuration does not currently contain the command ip flow-cache feature-accelerate, this change does not affect you.
The removal of NetFlow Feature Acceleration does not affect any other aspects of Netflow operation, for example Access-list processing. The features are separate and distinct.
Cisco Express Forwarding (CEF) supercedes the deprecated NetFlow Feature Acceleration.
Additionally, the following MIB objects and OIDs have been deprecated and removed from the netflow mib (CISCO-NETFLOW-MIB):
cnfFeatureAcceleration 1.3.6.1.4.1.9.9.99999.1.3
cnfFeatureAccelerationEnable 1.3.6.1.4.1.9.9.99999.1.3.1
cnfFeatureAvailableSlot 1.3.6.1.4.1.9.9.99999.1.3.2
cnfFeatureActiveSlot 1.3.6.1.4.1.9.9.99999.1.3.3
cnfFeatureTable 1.3.6.1.4.1.9.9.99999.1.3.4
cnfFeatureEntry 1.3.6.1.4.1.9.9.99999.1.3.4.1
cnfFeatureType 1.3.6.1.4.1.9.9.99999.1.3.4.1.1
cnfFeatureSlot 1.3.6.1.4.1.9.9.99999.1.3.4.1.2
cnfFeatureActive 1.3.6.1.4.1.9.9.99999.1.3.4.1.3
cnfFeatureAttaches 1.3.6.1.4.1.9.9.99999.1.3.4.1.4
cnfFeatureDetaches 1.3.6.1.4.1.9.9.99999.1.3.4.1.5
cnfFeatureConfigChanges 1.3.6.1.4.1.9.9.99999.1.3.4.1.6Open Caveats—Cisco IOS Release 12.1(27)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(27). All the caveats listed in this section are open in Cisco IOS Release 12.1(27). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: When a Secure Shell (SSH) connection to a router that uses TACACS+ for authentication fails because of an unknown user name or incorrect password, a memory leak occurs and a TCP connection may hang in the CLOSEWAIT or ESTAB state. For an SSH2 connection, a memory leak occurs even if the authentication succeeds.
Conditions: This symptom is observed on an SSH connection to a Cisco router.
Workaround: There is no workaround. However, under normal circumstances, the memory leak is relatively small and should not interfere with normal operation.
Resolved Caveats—Cisco IOS Release 12.1(27)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(27). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(27). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: A system used for reverse connections, such as a console server or other "milking machine" applications, may unexpectedly restart due to a bus error.
Conditions: The conditions under which this occurs are not well understood, but it is likely that frequent, short-lived connections are more likely to cause the problem than environments where connections are either long-lived or rarely opened and closed.
Workaround: There is no workaround.
•
Symptoms: Although there are free tty lines, you cannot make a Telnet connection and a "No Free TTYs error" message is generated.
Conditions: This symptom is observed when there are simultaneous Telnet requests.
Workaround: There is no workaround.
Interfaces and Bridging
•
Symptoms: A Versatile Interface Processor 2-50 (VIP2-50) crashes because of a Cybus error with DMA receive errors.
Conditions: This symptom is observed on a Cisco 7500 series that runs Cisco IOS Release 12.1 and that is configured with a PA-2FE that is installed in a VIP2-50. The symptom may also occur in other releases.
Workaround: There is no workaround.
IP Routing Protocols
•
Symptoms: A router may reload when a static Enhanced Interior Gateway Routing Protocol (EIGRP) neighbor statement is being removed.
Conditions: If a static neighbor statement is removed from the EIGRP configuration at the same time the neighbor is going down, the router may reload.
Workaround: Do not use static neighbors in EIGRP.
•
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.
•
Symptoms: A router that is configured for OSPF may reload unexpectedly and reference the "ospf_build_one_paced_update" process.
Conditions: This is observed on a Cisco router that has a mixture of LSAs (of type 5 and 11) that travel throughout an autonomous system and LSAs (of any type other than type 5 and 11) that travel within a particular OSPF area. The symptom may occur at any time without any specific changes or configuration and is not specifically related to any type of LSA.
Workaround: There is no workaround.
Further Problem Description: The symptom is very unlikely to occur. The symptom does not occur on a router that has exclusively stub areas and NSSA areas. The symptom may occur when a router does not have exclusively stub areas and NSSA areas.
•
Symptoms: A router may reload with a bus error exception, the crashinfo file shows an address error (a load or instruction fetch), and there is a spurious access in the crashinfo file.
Condition: These symptoms are observed on a Cisco router that performs NAT on H.323 voice traffic.
Workaround: There is no workaround.
•
Symptoms: A router may reload when a subinterface with a certain configuration is deleted.
Conditions: This symptom is observed on a Cisco router that has multicast and the Hot Standby Routing Protocol (HSRP) configured.
Workaround: Remove the multicast configuration before deleting the subinterface.
•
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.
Miscellaneous
•
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.
•
Symptoms: A Cisco router that is configured for SNASw may reload because of a bus error.
Conditions: This symptom is observed when the downstream port is configured for VDLC (DLSw). The problem is more likely to happen in a large, busy SNASw environment.
Workaround: There is no workaround.
Further Problem Description: This issue is platform independent.
Wide-Area Networking
•
Symptoms: A memory leak may occur in the "Multilink Events" process, which can be seen in the output of the show memory summary command:
0x60BC47D0 0000000024 0000000157 0000003768 MLP bundle name
0x60BC47D0 0000000028 0000000003 0000000084 MLP bundle name
0x60BC47D0 0000000044 0000000001 0000000044 MLP bundle name
0x60BC47D0 0000000048 0000000001 0000000048 MLP bundle name
0x60BC47D0 0000000060 0000000001 0000000060 MLP bundle name
0x60BC47D0 0000000064 0000000013 0000000832 MLP bundle name
0x60BC47D0 0000000068 0000000008 0000000544 MLP bundle name
0x60BC47D0 0000000072 0000000001 0000000072 MLP bundle name
0x60BC47D0 0000000076 0000000001 0000000076 MLP bundle name
0x60BC47D0 0000000088 0000000018 0000001584 MLP bundle nameConditions: This symptom is observed when two interfaces are configured in the same multilink group or are bound to the same dialer profile.
Workaround: There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(26)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(26). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(26). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: A specifically crafted Transmission Control Protocol (TCP) connection to a telnet or reverse telnet port of a Cisco device running Internetwork Operating System (IOS) may block further telnet, reverse telnet, Remote Shell (RSH), Secure Shell (SSH), and in some cases Hypertext Transport Protocol (HTTP) access to the Cisco device. Telnet, reverse telnet, RSH and SSH sessions established prior to exploitation are not affected.
All other device services will operate normally.
Conditions: User initiated specially crafted TCP connection to a telnet or reverse telnet port results in blocking further telnet sessions. Whereas, services such as packet forwarding, routing protocols and all other communication to and through the device remains unaffected.
Workaround: The detail advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040827-telnet.shtml
Miscellaneous
•
Symptoms: When the second REQDACTPU is sent to VTAM by SNA Switching Services (SNASW), the FQPCID supplied is a new, unused FQPCID rather than that sent on the initial REQACTPU. This usually has no ill side-effects. However, under certain conditions, the fact that the REQDACTPU RSP cannot be correlated can allow the PU to become perpetually hung.
Conditions: This symptom occurs when the second REQDACTPU is sent to VTAM by SNA Switching Services (SNASW).
Workaround: There is no workaround.
•
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
TCP/IP Host-Mode Services
•
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.
Resolved Caveats—Cisco IOS Release 12.1(25)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(25). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(25). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Miscellaneous
•
Symptoms: A Cisco 3600 series with an NM-CT1/E1 network module that contains an NM-xDM network module may not allow incoming modem calls and generate the "no modem available" error message even though the output of the show modem command indicates that there is a free modem available.
Conditions: This symptom is observed when frequent retrains occur on the modems.
Workaround: There is no workaround.
•
Symptoms: On an RTR probe, an RSP does not report input or output packets for serial interfaces of PA-MC-8T1, PA-MC-8E1, and PA-MC-8TE1+ port adapters.
Conditions: This symptom is observed on a Cisco 7500 series that runs Cisco IOS Release 12.2(23a) or Release 12.3 and is more likely to occur when the number of channelized port adapters (such as the PA-MC-8T1, PA-MC-8E1, and PA-MC-8TE1+ port adapters) that are installed in the router is high.
Workaround: Reload the router.
Alternate Workaround: Enter the reload microcode router configuration command.
•
Symptoms: A Cisco 7500 series router may reload when multilink interface configured on the router comes up.
Conditions: This symptom would happen if service-policy is configured on the multilink interface and distributed switching is enabled.
Workaround: Not configuring service-policy on the router would prevent the router from reloading.
Wide-Area Networking
•
Symptoms: The IP Control Protocol (IPCP) times out in a Link Control Protocol (LCP) negotiation.
Conditions: The problem happens when "virtual-profile virtual-template" is configured without "virtual-profile if-needed" and an ASYNC call creates a Virtual-Access interface.
Workaround: Configure "virtual-profile if-needed" and use the ASYNC interface without a Vaccess.
Resolved Caveats—Cisco IOS Release 12.1(24)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(24). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(24). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: Many memory allocation failure (MALLOCFAIL) messages may occur for a Cisco Discovery Protocol (CDP) process:
%SYS-2-MALLOCFAIL: Memory allocation of -1732547824 bytes failed from x605111F0, pool Processor, alignment 0
-Process= "CDP Protocol", ipl= 0, pid= 42
-Traceback= 602D5DF4 602D78A0 605111F8 60511078 6050EC88 6050E684 602D0E2C 602D0E18Conditions: The symptom is observed on a Cisco 7513 that runs Cisco IOS Release 12.0(17)ST. The symptom may also occur on other Cisco 7500 series routers that run Release 12.0 S, 12.2 S, 12.3, or 12.3 T.
Workaround: To prevent the symptom from occurring again, disable CDP by entering the no cdp run global configuration command.
•
Symptoms: A router may reload unexpectedly due to memory corruption.
Conditions: Connection accounting is enabled and the router is handling a high volume of connections.
Workaround: Disable connection accounting.
Further Problem Description: The high volume scenario can occur when there are a lot of users (typically scripts) logging into the router through VTYs, and some of them cause a race condition where the system is trying to do accounting after the connection structures have been freed.
•
Symptoms: A Cisco device experiences a memory leak in the CDP process.
Conditions: The device sending CDP packets sends a hostname that is 256 or more characters. There are no problems with a hostname of 255 or fewer characters.
Workaround: Configure the neighbor device to use less than a 256 character hostname, or disable the CDP process with the global command no cdp run.
IBM Connectivity
•
Symptoms: A Cisco 2620 may reload because of a segmentation violation (SegV).
Conditions: This symptom is observed when you attempt to run X.25 (at packet level) over a Logical Link Control, type 2 (LLC2) (at frame level) from a third-party vendor workstation to the Cisco 2620. This problem was also seen when running DLSW (Data Link Switching).
Workaround: There is no workaround.
Interfaces and Bridging
•
Symptoms: A port adapter may stop receiving packets. When this symptom occurs, the output of the show interface EXEC command does not report any input or output drops. When the show controller EXEC command is issued on the Versatile Interface Processor (VIP) console of a router, the command output may display incrementing rx_no_buffer and virtual circuit connection (VCC) counts.
Conditions: This symptom is observed on an enhanced ATM Port Adapter (PA-A3) on a Cisco 7500 router.
Workaround: Bounce the port adapter interface by issuing the shutdown interface configuration command followed by the no shutdown interface configuration command.
IP Routing Protocols
•
Symptoms: A router flushes link-state advertisements (LSAs) that have not been refreshed for more than 50 minutes.
Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.0(19)ST1.
Workaround: There is no workaround.
•
Symptoms: If Network Address Translation (NAT) overload is configured, translation may not function properly. A packet may be translated twice, and the inside global address may be considered as the inside local address.
Conditions: This symptom is observed on a Multilayer Switch Feature Card (MSFC2) that is running Cisco IOS Release 12.1(2)E or Release 12.1(8b)E9.
Workaround: Configure an access list for the NAT dynamic mappings, that would permit inside hosts only.
•
Symptoms: NAT calculates an invalid UDP checksum for some checksum values.
Conditions: This symptom is observed in a very particular situation which depends on the NAT configuration and the UDP checksum value. After the translation, the new UDP checksum value of the translated packet is equal to zero. NAT ignores the new checksum value of zero and it uses the original checksum value, which causes a checksum error at the end device.
Workaround: There is no workaround.
Miscellaneous
•
Symptoms: When an available bit rate (ABR) permanent virtual circuit (PVC) is used with a Cisco 3600 or Cisco 2600 router on an ATM T1 Inverse Multiplexing over ATM (IMA) module, the maximum output possible on the interface is:
Approximately 50 kbps if the user configures ABR PCR = 1500
Approximately 1220 kbps if the user configures ABR PCR = 3000
Approximately 2500 kbps if the user configures ABR PCR = 4500
Output queue drops are shown on the ATM interface when the show interfaces interface x/y EXEC command is entered.
Conditions: This symptom is observed on a Cisco 3600 or Cisco 2600 series router on an ATM T1 Inverse Multiplexing over ATM (IMA) module.
Workaround: Use either a variable bit rate (VBR) or an unspecified bit rate (UBR).
•
Symptoms: Free memory in router, as shown by the show processor memory command, drops until ultimately the router may issue SYS-2- MALLOCFAIL messages and hang.
Conditions: Memory will be permanently leaked when a Cisco device receives duplicate per-user Authentication, Authorization, and Accounting (AAA) requests for TCP Header Compression for the same user.
Workaround: Disable per-user AAA TCP Header Compression.
•
Symptoms: SNMP does not return the ifAlias for the T1 controllers, even after the description command is configured on the controller.
Conditions: This symptom is observed on a Cisco AS5300 but is not platform dependent.
Workaround: There is no workaround.
•
Symptoms: A PRI time slot that receives an analog call may be stuck in a link-control-protocol (LCP) timeout for all consecutive ISDN PPP calls.
Conditions: This symptom is observed on a Cisco AS5200.
Workaround: Reload the router.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Symptoms: The show flash-filesystem EXEC command and the dir filesystem EXEC command may not work properly on a Cisco 2600XM, preventing you from seeing the flash images.
In addition, the copy destination url flash: EXEC command may fail when the erase option is not selected (that is, you type in no when you are asked if you want to erase the device). The copy destination url flash: EXEC command functions fine when you do select the erase option.
Conditions: These symptoms are observed on a Cisco 2600XM that is configured with a particular third-party vendor 16-MB SIMM. Note that the router is still functional with this SIMM; you can boot or reload the router, perform a TFTP download operation, and similar actions without any difficulty.
Workaround: There is no workaround.
•
Symptoms: The box may crash when configuring the SSH key.
Conditions: The crash happens when the key generation is done right after reload.
Workaround: Wait 30 seconds after reload before generating key pair.
•
Symptoms: The ringback tone is not heard on the originating phone in the call alerting phase of an H.323 slow start VoIP call using a Cisco IOS VoIP gateway as the terminating gateway (TGW).
Conditions: The TGW sends an H.225 Alerting message with a Progress Indicator value of 1 or 8, which means that the ringback tone is transported in-band from the TGW.
Workaround: There is no workaround.
•
Symptoms: A Cisco router may reload when a member is removed and then added back to a multilink interface.
Conditions: This symptom is observed on a Cisco 7500 series when distributed Multilink PPP (MLP) is enabled.
Workaround: Shut down the multilink and then add or remove the member links. This action can prevent this symptom, but is not a very acceptable workaround.
•
Symptoms: An enhanced ATM port adapter (PA-A3) may display an increasing "rx_no_buffer" counter in the output of the show controllers atm privileged EXEC command, and some PVCs configured on the PA-A3 port adapter may stop receiving traffic.
Conditions: This symptom is observed when there is a high-traffic load on the PA-A3. Certain types of PA-A3s are impacted by this problem (PA-A3-OC3/T3/E3 are impacted, but PA-A3-OC12 and PA-A3-8T1/8E1 IMA are not). Also, any platform supporting these types of PA-A3s may be impacted.
Workaround: Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the PA-A3.
Further Problem Description: The condition that triggers this problem is ignores on the ATM interface. The high-traffic load must be high enough to cause ignores on ATM interface in order for the problem to occur. However, it is important to note that ignores on the ATM interface does not always leads to this problem.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
TCP/IP Host-Mode Services
•
Symptoms: When prompts for Tool Command Language (TCL) and Voice Extensible Markup Language (VXML) applications are loaded using FTP, FTP does not return the proper size of the prompts.
Conditions: This symptom is observed on a Cisco AS5300 but is not platform dependent.
Workaround: Load the prompts using RAM, TFTP, or HTTP.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOSÆ software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
Wide-Area Networking
•
Symptoms: When a router running Cisco IOS Release 12.0S, 12.1, 12.2, or 12.2T receives a multilink packet with Protocol Field Compression (PFC) applied, the packet is not interpreted correctly, and is subsequently rejected. The following debug messages appear in the debug trace when the debug ppp negotiation command is enabled:
MLP: I UNKNOWN(192) [Not negotiated] id 0 len 0
LCP: O PROTREJ [Open] id 2 len 95 protocol MLPConditions: This symptom is observed when the router requests PFC during Link Control Protocol (LCP) negotiations and the peer applies PFC to its outbound packets. PFC is enabled by default on asynchronous serial interfaces, it is disabled by default on other interfaces.
Workaround: In Cisco IOS Release 12.2 and 12.2 T, PFC can be disabled using the ppp pfc local forbid interface configuration command. In Release 12.0 S and 12.1, there is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(22c)
Cisco IOS Release 12.1(22c) is a rebuild release for Cisco IOS Release 12.1(22). The caveats in this section are resolved in Cisco IOS Release 12.1(22c) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
Basic System Services
•
Symptoms: Many memory allocation failure (MALLOCFAIL) messages may occur for a Cisco Discovery Protocol (CDP) process:
%SYS-2-MALLOCFAIL: Memory allocation of
-1732547824 bytes failed from x605111F0, pool Processor, alignment 0
-Process= "CDP Protocol", ipl= 0, pid= 42
-Traceback= 602D5DF4 602D78A0 605111F8 60511078 6050EC88 6050E684 602D0E2C 602D0E18Conditions: The symptom is observed on a Cisco 7513 that runs Cisco IOS Release 12.0(17)ST. The symptom may also occur on other Cisco 7500 series routers that run Release 12.0 S, 12.2 S, 12.3, or 12.3 T.
Workaround: To prevent the symptom from occurring again, disable CDP by entering the no cdp run global configuration command.
•
Symptoms: A router may reload unexpectedly due to memory corruption.
Conditions: Connection accounting is enabled and the router is handling a high volume of connections.
Workaround: Disable connection accounting.
Further Problem Description: The high volume scenario can occur when there are a lot of users (typically scripts) logging into the router through vty's and some of them cause a race condition where the system is trying to do accounting after the connection structures have been freed.
•
Symptoms: A Cisco Catalyst 2950 experiences a memory leak in the CDP process.
Conditions: The device sending CDP packets sends a hostname that is 256 or more characters. There are no problems with a hostname of 255 or fewer characters.
Workaround: Configure the neighbor device to use less than a 256 character hostname, or disable the CDP process with the global command no cdp run.
Miscellaneous
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
Symptoms: A router may reload unexpectedly after it attempts to access a low memory address.
Conditions: This symptom is observed after ACLs have been updated dynamically or after the router has responded dynamically to an IDS signature.
Workaround: Disable IP Inspect and IDS.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
Symptoms: The ringback tone is not heard on the originating phone in the call alerting phase of an H.323 slow start VoIP call using a Cisco IOS VoIP gateway as the terminating gateway (TGW).
Conditions: The TGW sends an H.225 Alerting message with a Progress Indicator value of 1 or 8, which means that the ringback tone is transported in-band from the TGW.
Workaround: There is no workaround.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
Resolved Caveats—Cisco IOS Release 12.1(22b)
Cisco IOS Release 12.1(22b) is a rebuild release for Cisco IOS Release 12.1(22). The caveats in this section are resolved in Cisco IOS Release 12.1(22b) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
IP Routing Protocols
•
Symptoms: A router flushes link-state advertisements (LSAs) that have not been refreshed for more than 50 minutes.
Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.0(19)ST1.
Workaround: There is no workaround.
Miscellaneous
•
Symptoms: The show flash-filesystem EXEC command and the dir filesystem EXEC command may not work properly on a Cisco 2600XM, preventing you from seeing the flash images.
In addition, the copy destination url flash: EXEC command may fail when the erase option is not selected (that is, you type in no when you are asked if you want to erase the device). The copy destination url flash: EXEC command functions fine when you do select the erase option.
Conditions: These symptoms are observed on a Cisco 2600XM that is configured with a particular third-party vendor 16-MB SIMM. Note that the router is still functional with this SIMM; you can boot or reload the router, perform a TFTP download operation, and similar actions without any difficulty.
Workaround: There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(22a)
Cisco IOS Release 12.1(22a) is a rebuild release for Cisco IOS Release 12.1(22). The caveats in this section are resolved in Cisco IOS Release 12.1(22a) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
IP Routing Protocols
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
Miscellaneous
•
Symptoms: The H.323 protocol stack on Cisco IOS software uses a third-party Abstract Syntax Notation One (ASN.1) library, and Cisco does not have control over the source code. Under some situations, when there are encoding errors and no free memory is available, a Cisco router reloads inside the third-party library code. In the third-party library code, a function is called by the ASN encoder to free all allocated memory whenever an error is detected. However, the encoder does not free the memory correctly, and the router reloads.
Conditions: This symptom is observed on a Cisco AS5300 but may be observed on other Cisco platforms.
Workaround: Upgrade to Cisco IOS Release 12.1 T or a later release that uses Cisco-developed ASN.1 library code.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Symptoms: Cisco IOS software may cause a Cisco router to reload unexpectedly when the router receives a malformed H.225 setup message.
Conditions: This symptom is observed on a Cisco 1700 series that runs Cisco IOS Release 12.2(13c). The symptom occurs when the following debug privileged EXEC commands are enabled:
–
–
–
Workaround: There is no workaround.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
Resolved Caveats—Cisco IOS Release 12.1(22)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(22). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(22). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: After a Versatile Interface Processor (VIP) has reloaded, there does not seem to be a crashinfo file because the crashinfo file is not closed; therefore, it is not visible or accessible. If the same VIP reloads again, both the first and second crashinfo files are accessible.
Conditions: This symptom is observed on a Cisco 7500 series that is running Cisco IOS Release 12.2(6f).
Workaround: There is no workaround.
•
Symptoms: A Cisco router may fail to process Cisco Discovery Protocol (CDP) packets and update the IP process for On Demand Routing (ODR) routes.
Conditions: This symptom is mainly observed on WAN interfaces with traffic that is passing through the link when a Cisco router fails to update the hold-down timer and the IP process when it receives a CDP packet from a neighbor.
Workaround: There is no workaround.
•
Symptoms: A Cisco MC3810 router may reload when it makes Voice over Frame Relay (VoFR) calls.
Conditions: This symptom is most likely to be observed on a Cisco MC3810 router under stress conditions (24 calls). However, it may also occur (rarely) when a single call is made.
Workaround: There is no workaround.
IBM Connectivity
•
Symptoms: Ethernet redundancy may not function with Inter-Switch Link (ISL) trunking.
Conditions: This symptom is observed on a Cisco router or switch that is configured for data-link switching (DLSw) and Ethernet Redundancy (ER).
Workaround: There is no workaround.
•
Symptoms: A Cisco router that is configured for data-link switching (DLSw) may generate the following error messages and tracebacks:
%TCP-2-INVALIDTCPENCAPS: Invalid TCB encaps pointer: 0x0 -Process= "DLSw Peer Process", ipl= 0, pid= 81
-Traceback= 603BDCDC 603BEFC4 60AC5A24 60AC6E00 60AC4F54 60AB51D0 60AB4D04 60AB4 958 60223B44 60223B30
%TCP-2-INVALIDTCPENCAPS: Invalid TCB encaps pointer: 0x0 -Process= "IP Input", ipl= 0, pid= 29
-Traceback= 603BDCDC 603BEFC4 60AC5A24 60AC6E00 60AC4F54 60AB51D0 60ABCF44 603BD C28 60325EC0 60327C44 6035E49C 60346DCC 603452C8 603453C4 60345538 60223B44Conditions: This symptom is observed in a DLSw border peer network that uses DLSw priority peers. Note that the symptom does not affect the DLSw functionality.
Workaround: There is no workaround.
IP Routing Protocols
•
Symptoms: When next-hop-self is configured on a peer group, the next-hop calculation is only performed on the first member of the peer group, and the same next-hop value is replicated to the rest of the peers instead of calculating the next hop based on the next-hop-self configuration. The problem of wrong next-hop value on the peer group members occurs if the router is multihomed and if Border Gateway Protocol (BGP) uses those multiple interfaces to peer with the neighbors which are in the same peer group (or update group), then the same next-hop value of the leader of the peer group is used for all the members.
Conditions: This symptom is observed on a Cisco 7200 router that is running Cisco IOS Release 12.2, Release 12.3, or Release 12.3T.
Workaround: Remove the peer groups to allow the calculation to be run for each neighbor.
Alternate Workaround: Make sure that all the peers which are in the same peer group can be reached through a single interface and use that interface IP address, using the BGP update-source command, as the local peering address.
•
Symptoms: When a Border Gateway Protocol (BGP) process propagates routes that are learned from an internal BGP (iBGP) peer to an external BGP (eBGP) peer, the eBGP peer should see these routes with the next-hop address of the originator's address. However, the eBGP peer sees the routes with the next-hop address of the router that propagates the routes not the router that originates the routes.
Conditions: This symptom is observed in Cisco IOS Release 12.1(22).
Workaround: There is no workaround.
Miscellaneous
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Symptoms: A Cisco gatekeeper may reload if the show gatekeeper gw-type-prefix privileged EXEC command is entered on the gatekeeper.
Conditions: This symptom may be observed on all Cisco platforms when there is a large routing table on the gatekeeper.
Workaround: There is no workaround.
•
Symptoms: A router may reload.
Conditions: This symptom is observed when the show ip vrf EXEC command is entered for virtual private network (VPN) routing/forwarding (VRF) instances that are being unconfigured by entering the no ip vrf command script. This symptom affects releases that contain the Multiprotocol Label Switching (MPLS) VPN feature.
Workaround: Do not enter the show ip vrf EXEC command for VRF instances that are being processed by the no ip vrf command script.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Symptoms: A Cisco AS5300 may reload because of a software condition, and the following error message may be displayed when you enter the show version EXEC command:
System returned to ROM by error - a Software forced crash, PC 0xXXXXXXXX"0xXXXXXXXX" represents the program counter on which the reload occurred.
Conditions: This symptom is observed on a Cisco AS5300 that is running Cisco IOS Release 12.1(16) but may also occur in Release 12.2 T.
Workaround: There is no workaround.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Symptoms: A router may reload because of a bus error.
Conditions: This symptom is observed on a Cisco router if a VPN routing/forwarding (VRF) instance is deleted while the show ip vrf vrf-name EXEC command executes.
Workaround: Do not enter the show ip vrf vrf-name EXEC command while any VRF is being deleted.
•
Symptoms: A Cisco router may generate a "SYS-2-GETBUF" message during the "Tag Input" process and may reload unexpectedly.
Conditions: This symptom is observed when the router fragments a Multiprotocol Label Switching (MPLS) packet.
Workaround: There is no workaround.
•
Symptoms: Cisco IOS software may cause a Cisco router to reload unexpectedly when the router receives a malformed H.225 setup message.
Conditions: This symptom is observed on a Cisco 1700 series that runs Cisco IOS Release 12.2(13c). The symptom occurs when the following debug privileged EXEC commands are enabled:
–
–
–
Workaround: There is no workaround.
•
Symptoms: CPUHOG messages and tracebacks may occur on a Cisco router when you attempt to register more than 10,000 gateways.
Conditions: This symptom is observed on a Cisco 7200 series that is configured with a Network Processing Engine G1 (NPE-G1).
Workaround: There is no workaround.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.Wide-Area Networking
•
Symptoms: A router may reload because of a watchdog timeout if the no dialer pool-member interface configuration command is entered on the D channel of the router.
Conditions: This symptom is observed on a Cisco router when the command is entered on the D channel and there is more than one link that is bound to the dialer profile with Multilink PPP (MLP).
Workaround: Shut down the dialer interfaces and physical interfaces that are relevant to the dialer pool. After the interfaces are completely down, enter the no dialer pool-member interface configuration command.
Resolved Caveats—Cisco IOS Release 12.1(21)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(21). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(21). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: A Cisco 7500 series router may reload because of a software condition after an online insertion and removal (OIR) of a Versatile Interface Processor (VIP) that is configured with an ATM OC-3c/STM-1 port adapter (PA-A3- OC3) and after the following error message has been generated:
%SYS-6-STACKLOW: Stack for process OIR Handler running low, 12/3000Conditions: This symptom is observed on a Cisco 7500 series router that is running Cisco IOS Release 12.2(7) and occurs with a VIP2-50, VIP4, and VIP6. The symptom is related to the PA-A3-OC3 and occurs only during an OIR.
Workaround: Do not perform an OIR on any VIP that is configured with a PA-A3- OC3.
IP Routing Protocols
•
Symptoms: A router may experience frequent Multicast Source Discovery Protocol (MSDP) session resets with the MSDP peers of the router.
Conditions: This symptom is observed on Cisco router that is configured for MSDP. This situation is often caused by excessive source, group (S, G) information that should be contained in a domain being passed to the outside, resulting in additional entries in the Source-Active (SA) cache.
Workarounds:
–
–
–
•
Symptoms: A Cisco router that is using a static Network Address Translation (NAT) outside configuration is forced to reload.
Conditions: The static NAT ip nat inside source static network command or ip nat outside source static network command must be present to reload the router.
Workaround: There is no workaround.
•
Symptoms: Using the show ip sdr command may result in the printing of garbage characters.
Conditions: This symptom occurs if the multicast Session Description Protocol (SDR) session expires while the output is being printed.
Workaround: There is no workaround.
•
Symptoms: A router may reload because of a bus error after the show ip sdr EXEC command is entered.
Conditions: This symptom is observed on a Cisco 7513 router that is running either Cisco IOS Release 12.0(22)S1 or Release 12.0(22)S2. This symptom occurs because an invalid SDR Session announcement message is received without the name of the session.
Workaround: Avoid the use of the show ip sdr command.
•
Symptoms: A summarized entry may remain in the Enhanced Interior Gateway Routing Protocol (EIGRP) topology table after manual summarization is disabled.
Conditions: This symptom is observed when manual summarization is enabled and subsequently disabled on a network that is also being redistributed into EIGRP.
Workaround: Restart the EIGRP process.
•
Symptoms: A Cisco router may reload by bus error when ip accounting is configured on the router.
Conditions: This symptom is observed on a Cisco 7206VXR router that is running Cisco IOS Release 12.2(11)T2 after entering the clear ip accounting EXEC command.
Workaround: Do not use the clear ip accounting EXEC command or the show ip accounting EXEC command.
•
Symptoms: A Cisco router may reload with a bus error.
Conditions: This symptom is observed on a Cisco 7200 series router when the ip accounting EXEC command has been used.
Workaround: There is no workaround.
Miscellaneous
•
Symptoms: While running IP security (IPSec), a Cisco 2600 series router reloads with the following error message:
ALIGN-3-SPURIOUS: Spurious memory access made at 0x810FF844 reading 0x0Conditions: This symptom is observed on a Cisco 2600 series router that is running Cisco IOS Release 12.1(5)T5.
Workaround: There is no workaround.
•
Symptoms: After a few weeks of normal operation, the interface on a Cisco PA- MC-8E1 begins flapping and finally pauses with the output queue stuck as follows:
Serial1/1:1 is up, line protocol is up
Encapsulation HDLC, crc 16, Data non-inverted
Keepalive set (120 sec)
Last input 00:00:03, output 04:14:23, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 21952
Queueing strategy: weighted fair
Output queue: 30/4000/64/21855 (size/max total/threshold/drops)
30 second input rate 0 bits/sec, 0 packets/sec
30 second output rate 0 bits/sec, 0 packets/sec
43903807 packets input, 3646461183 bytes, 0 no buffer
Received 0 broadcasts, 321 runts, 0 giants, 0 throttles
5160 input errors, 4 CRC, 0 frame, 0 overrun, 0 ignored, 2945 abort
42026998 packets output, 2185017012 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 output buffer failures, 0 output buffers swapped out
31 carrier transitions
no alarm present
Timeslot(s) Used:1-31, subrate: 64Kb/s, transmit delay is 0 flagsThe following traceback is observed in the log:
%LINK-4-TOOBIG: Interface Serial60:1, Output packet size of
1526 bytes too big
Traceback= 0x604007F8 0x604A927C 0x6084E4D4 0x6057425C 0x60CE921C
0x60CE55EC
%LINK-4-TOOBIG: Interface Serial20:1, Output packet size of
1526 bytes too big
Traceback= 0x604007F8 0x604A927C 0x6084E4D4 0x6057425C 0x60CE921C
0x60CE55ECConditions: This symptom is observed on a Cisco router that is configured with a PA-MC-8E1 interface.
Workaround: There is no workaround.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.•
Symptoms: REQACTPU is rejected with an 08060000 sense code. SNA Switching Services (SNASw) may not stop the link station so that the end device can try another data-link switching (DLSw) peer. The SNASw link and the DLSw TCP/IP circuit stay intact so the physical unit (PU) continues to retry on an invalid host, which affects sites that peer to multiple hosts.
Conditions: This symptom is observed in Cisco IOS Release 12.1(15) or Release 12.2(12) and later releases. A design change was introduced via CSCdw93088 to cause the circuit not to break.
Workaround: Manually break the circuit so that DLSw can use the other DLSw peer.
•
Symptoms: It may be difficult to make an Inverse Multiplexing over ATM (IMA) link between a Cisco router and other vendor equipment.
Conditions: This symptom is observed on Cisco 2600 series and Cisco 3600 series routers. When an IMA link is configured between the Cisco 2600 series and the Cisco 3600 series and other vendor equipment, the Cisco routers keep sending the test link command (set to 1) in the IMA Control Protocol (ICP) cell regardless of the ima test interface configuration command. Both the Cisco 2600 series and Cisco 3600 series platforms need the fix for the caveat CSCds55768 to eliminate this symptom.
Workaround: There is no workaround.
•
Symptoms: An ATM output subinterface may pause indefinitely. When this defect is observed, the pxmt counter in the show controllers command also pauses with nonzeroentries as shown in the following log:
Router#show controllers atm | in pxmt
pxmt: 196 queued: 59Conditions: This symptom is observed on a Cisco 7206 router with a PA-A2 ATM CES port adapter that is running Cisco IOS Release 12.1(17).
Workaround: Use the shutdown interface configuration command followed by the no shutdown interface configuration command to disable and restart the ATM subinterface.
•
Symptoms: IP Security (IPSec) connections from Cisco Virtual Private Network (VPN) 1.1 clients may fail when connecting to a Cisco router. The crypto debugs will indicate a problem negotiating DH phase I values as shown in the following display:
*Aug 6 06:44:49.639: ISAKMP (0:1): Unable to generate DH phase I values!Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1(20).
Workaround: There is no workaround.
•
Symptoms: Leaks may be observed for some Virtual Private Network routing and forwarding (VRF) routes in the global Forwarding Information Base (FIB) table when a VRF is deleted and recreated.
Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.0 S or Release 12.2 T.
Workaround: There is no workaround.
Wide-Area Networking
•
Symptoms: A Cisco Catalyst 2900 series XL switch sends Bridge Protocol Data Units (BPDUs) over a permanent virtual circuit (PVC), which is in the DOWN state because of an Operation, Administration, and Maintenance (OAM) failure.
Conditions: This symptom is observed on a Cisco Catalyst 2900 series XL switch.
Workaround: There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(20a)
Cisco IOS Release 12.1(20a) is a rebuild release for Cisco IOS Release 12.1(22). The caveats in this section are resolved in Cisco IOS Release 12.1(20a) but may be open in previous Cisco IOS Releases. This section describes severity 1 and 2 caveats and select severity 3 caveats.
Miscellaneous
•
Symptoms: The H.323 protocol stack on Cisco IOS software uses a third-party Abstract Syntax Notation One (ASN.1) library, and Cisco does not have control over the source code. Under some situations, when there are encoding errors and no free memory is available, a Cisco router reloads inside the third-party library code. In the third-party library code, a function is called by the ASN encoder to free all allocated memory whenever an error is detected. However, the encoder does not free the memory correctly, and the router reloads.
Conditions: This symptom is observed on a Cisco AS5300 but may be observed on other Cisco platforms.
Workaround: Upgrade to Cisco IOS Release 12.1 T or a later release that uses Cisco-developed ASN.1 library code.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Symptoms: Cisco IOS software may cause a Cisco router to reload unexpectedly when the router receives a malformed H.225 setup message.
Conditions: This symptom is observed on a Cisco 1700 series that runs Cisco IOS Release 12.2(13c). The symptom occurs when the following debug privileged EXEC commands are enabled:
- debug h225 asn1
- debug h225 events
- debug h225 q931
Workaround: There is no workaround.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
A vulnerability in the Transmission Control Protocol (TCP) specification (RFC793) has been discovered by an external researcher. The successful exploitation enables an adversary to reset any established TCP connection in a much shorter time than was previously discussed publicly. Depending on the application, the connection may get automatically re-established. In other cases, a user will have to repeat the action (for example, open a new Telnet or SSH session). Depending upon the attacked protocol, a successful attack may have additional consequences beyond terminated connection which must be considered. This attack vector is only applicable to the sessions which are terminating on a device (such as a router, switch, or computer) and not to the sessions that are only passing through the device (for example, transit traffic that is being routed by a router). In addition, this attack vector does not directly compromise data integrity or confidentiality.
All Cisco products which contain TCP stack are susceptible to this vulnerability.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-ios.shtml, and it describes this vulnerability as it applies to Cisco products that run Cisco IOS® software.
A companion advisory that describes this vulnerability for products that do not run Cisco IOS software is available at http://www.cisco.com/warp/public/707/cisco-sa-20040420-tcp-nonios.shtml.
•
Cisco products running IOS contain vulnerabilities in the processing of H.323 messages, which are typically used in packetized voice or multimedia applications. Features such as NAT and IOS Firewall must inspect H.323 messages and may be vulnerable as well. A test suite has been developed by the University of Oulu to target this protocol and identify vulnerabilities.
Support for the H.323 protocol was introduced in Cisco IOS Software Release 11.3T, and all later Cisco IOS releases are affected if configured for various types of Voice/Multimedia Application support. The vulnerabilities can be exploited repeatedly to produce a denial of service (DoS).
There are workarounds available that may mitigate the impact, but these techniques may not be appropriate for use in all customer networks.
This advisory is available at http://www.cisco.com/warp/public/707/cisco-sa-20040113-h323.shtml.
Resolved Caveats—Cisco IOS Release 12.1(20)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(20). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(20). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms: A serial interface may show an interface as down/down even when all signals (request to send [RTS], clear to send [CTS], data terminal ready [DTR], data carrier detect [DCD]) are present.
Conditions: This symptom is observed on the serial interface of a Cisco router that is running Cisco IOS Release 12.0(15).
Workaround: Reload the microcode.
•
Symptoms: A router may reload because of a segmentation violation exception error or a bus error and display either of the following two error messages:
System was restarted by error - a SegV exception, PC 0x8042B0ACSystem returned to ROM by bus error at PC 0x605A9970, address 0x64
Conditions: This symptom is observed on a Cisco router that has generic traffic shaping configured when the router reloads because of a segmentation violation exception error or a bus error.
Workaround: Remove the traffic shaping configuration from the interfaces that have traffic shaping configured.
IBM Connectivity
•
Symptoms: Data-link switching (DLSw) may cause a buffer leak in the small buffer pool of a router.
Conditions: This symptom is observed when DLSw Ethernet redundancy is used. This behavior does not occur when DLSw is used with source bridging, transparent bridge groups, Synchronous Data Link Control (SDLC), or Qualified Logical Link Control (QLLC).
This symptom occurs when the end system begins to communicate by sending an Exchange Identification (XID) frame to a destination service access point (DSAP) other than DSAP 0. This symptom does not occur if the end system begins communicating by sending a TEST frame or an XID frame to DSAP 0.
If the dlsw timer explorer-wait-time time global configuration command is configured, the rate of the buffer leak is proportional to the value of the time argument. An increase in the configured value of the time argument leads to an increase in the rate of the buffer leak.
Workaround: To minimize the impact of the buffer leak, the user may consider configuring the time argument of the dlsw timer explorer-wait-time time global configuration command to a value of "1" or removing the dlsw timer explorer-wait-time time global configuration command from the configuration.
Note
Interfaces and Bridging
•
Symptoms: A router may not recognize an ATM WAN OC-3 port adapter.
Conditions: This symptom is observed when an ATM WAN OC-3 port adapter is installed in slot 1 of a Cisco 7200 series router that has a Network Processing Engine 150 (NPE-150).
Workaround: There is no workaround.
IP Routing Protocols
•
Symptoms: A router that is directly connected to a source may not start registering when the source becomes active, and the (S,G) state may time out on the rendezvous point (RP).
Conditions: This symptom is observed on a router that is configured for Protocol Independent Multicast (PIM) and that has an (S,G) entry with the F flag reset.
Workaround: There is no workaround.
•
Symptoms: A directly-connected interface on a router that is covered by an Enhanced Interior Gateway Routing Protocol (EIGRP) network statement may not be displayed on the EIGRP topology table.
Conditions: This symptom is observed after a router that has the directly-connected interface is reloaded.
Workaround: Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the interface.
•
Symptoms: When an area border router receives type-4 link-state advertisements (LSAs) via the nonbackbone, the router may incorrectly generate type-4 LSAs into the backbone. This situation may cause a routing loop to occur.
Conditions: This symptom is observed in Cisco IOS Release 12.0(22)S or a later release or in Release 12.2(10) or a later release when the following conditions occur:
–
–
–
In this situation, Router 1 generates type-4 LSAs into the backbone area for the ASBR. This situation should not occur and may lead to a routing loop.
Workaround: Reset the Open Shortest Path First (OSPF) process by entering the clear ip ospf process privileged EXEC command.
•
Symptoms: The Multilayer Switch Feature Card (MSFC) of a Cisco Catalyst 6000 may reload with the following error message:
System was restarted by bus error at PC 0x40DFEE54, address 0xB0D0B7DConditions: This symptom is observed on a Cisco Catalyst 6000 that is configured for Network Address Translation (NAT).
Workaround: There is no workaround.
Miscellaneous
•
Symptoms: The boot Flash file system of a router may become corrupted.
Conditions: This symptom is observed when Cisco IOS software writes a crashinfo file after a router reloads.
Workaround: Configure the no exception crashinfo global configuration command to prevent the router from writing a crashinfo file after the router reloads.
•
A Cisco device running IOS and enabled for the Border Gateway Protocol (BGP) is vulnerable to a Denial of Service (DOS) attack from a malformed BGP packet. 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. BGP MD5 is a valid workaround for this problem.
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-20040616-bgp.shtml.
•
Symptoms: A Cisco Versatile Interface Processor 2-50 (VIP 2-50) may reload after a software upgrade.
Conditions: This symptom is observed on the Cisco VIP 2-50 of a Cisco 7500 series that has two single-port Fast Ethernet port adapters after the Cisco 7500 series is upgraded from Cisco IOS Release 12.1(2) to Release 12.1(16).
Workaround: This behavior can be avoided by setting a single interface or both interfaces to be administratively shut down while the router boots up with the new Cisco IOS release. The interfaces can be brought back up individually after the software is loaded and the router is stable.
•
Symptoms: The amount of memory that is held up by the Simple Network Management (SNMP) process increases constantly. This behavior may cause the router to eventually run out of memory and start displaying memory allocation (MALLOC) failure messages.
Conditions: This symptom is observed when the cieEngineStatusTable table is requested by SNMP.
Workaround: Stop collecting information about the cieEngineStatusTable table using SNMP.
•
Symptoms: A Cisco 7200 series router that is configured with Systems Network Architecture Switching Services (SNASw) and enterprise extender uplinks to a mainframe permanently pauses during the mainframe initial program load (IPL). The show process cpu user EXEC command indicates that the router is at 99 percent CPU utilization during the IPL.
Conditions: This symptom is observed on a Cisco 7200 series router.
Workaround: There is no workaround.
•
A Cisco device running IOS and enabled for the Border Gateway Protocol (BGP) is vulnerable to a Denial of Service (DOS) attack from a malformed BGP packet. 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. BGP MD5 is a valid workaround for this problem.
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-20040616-bgp.shtml.
•
Symptoms: The "giaddr" field is not updated for flooded Dynamic Host Configuration Protocol (DHCP) broadcasts.
Conditions: When the ip forward-protocol spanning-tree any-local-broadcast global configuration command is configured, DHCP broadcasts are forwarded to all interfaces in a bridge group, but the "giaddr" field is not set. This behavior causes issues in a network because the DHCP server uses the "giaddr" field in order to properly allocate addresses when the client is not in the local network.
Workaround: Configure the following commands on each interface that is connected to the DHCP or BOOTstrap Protocol (BOOTP) clients to forward the DHCP or BOOTP packets to the DHCP or BOOTP server and to properly set the "giaddr" field for locally-attached clients:
–
–
–
•
Symptoms: Systems Network Architecture Switching Services (SNASw) physical units may pause indefinitely in the "PendActPu" state.
Conditions: This symptom is observed after a host initial program load (IPL) occurs.
Workaround: Stop and restart SNASw.
•
Symptoms: A Systems Network Architecture Switching Services (SNASw) router that is configured with the snasw ipstrace global configuration command may reload.
Conditions: This symptom is observed when the show snasw ips user EXEC command is entered immediately after the show snasw link user EXEC command is entered.
Workaround: Do not configure the snasw ipstrace global configuration command. Always use the snasw start ipstrace user EXEC command and the snasw stop ipstrace user EXEC command in conjunction with the snasw ipstrace global configuration command.
•
Symptoms: A downstream physical unit (PU) may pause indefinitely in the "Pend ACTPU" state.
Conditions: This symptom is observed on a Systems Network Architecture (SNA) switch after a host initial program load (IPL) occurs when the SNA switch is busy activating PUs. The SNA switch does not send a REQACTPU response to the dependent logical unit server (DLUS).
Workaround: Redirect the PU to a redundant SNA switch.
Alternate Workaround: Stop and restart the SNA switch.
Novell IPX, XNS, and Apollo Domain
•
Symptoms: A router may reload unexpectedly.
Conditions: This symptom is observed when Internetwork Packet Exchange (IPX) routing is configured on a Cisco 3640.
Workaround: There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(19)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(19). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(19). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms The show bootflash: chips EXEC command may cause subsequent commands such as the show bootflash all EXEC command to fail.
Conditions This symptom is observed on a Cisco router that has a Route Switch Processor (RSP8). This symptom occurs because the bootflash module is flawed.
Workaround Enter the show version EXEC command to restore the router to normal operating condition. Alternatively, you may reseat or replace the Flash single in-line memory module (SIMM).
•
Symptoms: A software-forced reload may occur on a router.
Conditions: This symptom is observed on a Cisco router when the hunting process takes over the local plain old telephone service (POTS) dial peers. Hunting is a process that occurs on the router when it searches for a dial peer that it can use to send a voice call.
Workaround: There is no workaround.
Interfaces and Bridging
•
Symptoms: A Cisco router may pause indefinitely in rare traffic conditions.
Conditions: This symptom is observed on a Cisco 7200 series router that is using an MPA-CE1 port adapter. This condition occurs if weighted fair queueing (WFQ) or Multilink PPP (MLP) is mixed with pure FIFO queueing on the configured channels.
Workaround: There is no workaround.
•
Symptoms: A software-forced reload may occur on a 1-port Fast Ethernet 100BASE-TX (PA-1FE-TX) port adapter because of a block overrun.
Conditions: This symptom is observed on the PA-1FE-TX port adapter when it is running on a Cisco router. The likelihood of the port adapter reloading increases when the number of interface resets increases. The interface reset counter can be displayed by entering the show interface EXEC command on the PA-1FE-TX port adapter.
Workaround: There is no workaround.
•
Symptoms: A Versatile Interface Processor (VIP) may reload and restart after a Fast Ethernet port adapter (PA-FE) is installed.
Conditions: This symptom is observed on a VIP that is installed in a Cisco 7500 series router.
Workaround: There is no workaround.
•
Symptoms: X.25 encapsulation may not come up on interfaces that are installed on a port adapter.
Conditions: This symptom is observed on the 2-port channelized T1 ISDN port adapter (PA-2CT1) or 2-port channelized E1 port adapter (PA-2CE1) of a Cisco 7500 series. The cbus complex process may be executed if messages such as "output frozen" and "not transmitting" are displayed on interfaces that are installed on the port adapter.
Workaround: There is no workaround.
•
Symptoms: An enhanced ATM port adapter (PA-A3) may display an incrementing rx_no_buffer counter in the output of the show controllers atm EXEC command, and the PA-A3 port adapter may subsequently pause indefinitely and stop receiving traffic.
Conditions: This symptom is observed when there is traffic on the PA-A3 port adapter.
Workaround: Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the PA-A3 port adapter.
IP Routing Protocols
•
Symptoms: A universal access server may reload because of a bus error.
Conditions: This symptom is observed on a Cisco AS5300 series universal access server that is running Cisco IOS Release 12.1(15).
Workaround: There is no workaround.
•
Symptoms: Enhanced Interior Gateway Routing Protocol (EIGRP) neighbors may be reset unnecessarily on an interface that is configured with summarization. The following output is displayed when this symptom occurs:
%DUAL-5-NBRCHANGE: IP-EIGRP 111: Neighbor 10.108.255.244 (FastEthernet4/0) is down: Summary up, remove external
%DUAL-5-NBRCHANGE: IP-EIGRP 111: Neighbor 10.108.255.245 (FastEthernet4/0) is down: Summary up, remove externalConditions: The EIGRP neighbors are reset unnecessarily, and the summary is regenerated on an interface if all components of the summary are lost and at least one component is relearned.
Workaround: Remove the EIGRP summary statements from the interface.
•
Symptoms: A router may reload because of a bus error at the ipnat_unlock_parent_entry process.
Conditions: This symptom is observed on a Cisco router.
Workaround: There is no workaround.
Miscellaneous
•
Symptoms: A router may reload if a T1 or E1 channel-associated signaling (CAS) controller goes down before it receives more than a certain number of CAS or ISDN analog calls.
Conditions: This symptom is observed if CAS is configured on the T1 or E1 controllers of a Cisco router.
Workaround: There is no workaround.
•
Symptoms: A router may transmit bit errors.
Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1 and that has a serial 1DS3 or 2DS3 port adapter (PA-T3 or PA- 2T3) when the signal is below 0.5 decibels (dB).
Workaround: Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the interface.
•
Symptoms: A router may reload and display the following error message:
%SYS-6-STACKLOW: Stack for process Crypto Support running low, 0/1000Conditions: This symptom is observed on a Cisco 2500 series or Cisco 7500 series that has IP Security (IPSec) enabled.
Workaround: There is no workaround.
•
Symptoms: A Cisco 7500 series router experiences spurious memory accesses and reloads with a bus error.
Conditions: This symptom is observed on a Cisco 7500 series router that is running the Cisco rsp-pv-mz image and that has Multiprotocol Label Switching (MPLS) enabled. This situation is related to the use of the aggregate-address Border Gateway Protocol (BGP) command.
Workaround: There is no workaround.
•
Symptoms: A window condition may occur if an initial program load (IPL) is performed on a mainframe while it is processing a primary logical unit secondary logical unit (PLU-SLU) session over a dependent logical unit requester (DLUR). A logical unit (LU) may hang and may cause a downstream physical unit (DSPU) to pause indefinitely. The DSPU state is shown as "reset" even though the finite state machine (FSM) history shows the DSPU state as "reset link inactive." The state of the LU is also reset, and the DSPU cannot be used. An IPL has to be performed on the Systems Network Architecture (SNA) switch to recover the DSPU.
Conditions: These symptoms are observed only in a small window condition when an IPL is performed on a host while DLUR is used.
Workaround: Stop and then restart the SNA switch.
•
Symptoms: A software-forced reload may occur on a router, and the following messages may be displayed:
System was restarted by error - a Software forced crash, PC 0x60396E7C at 4500 Software (C4500-A3JS-M), Version 12.2(8.1), MAINTENANCE INTERIM SOFTWARE Compiled (current version) Image text-base: 0x60008948, data-base: 0x61116000 Stack trace from system failure: FP: 0x618A8458, RA: 0x60396E7C FP: 0x618A8458, RA: 0x603952F4 FP: 0x618A8480, RA: 0x6039D584 FP: 0x618A84A0, RA: 0x603A0CC8 FP: 0x618A84C0, RA: 0x60398BDC FP: 0x618A8558, RA: 0x6037E1F0 FP: 0x618A85A0, RA: 0x6174B1F0Conditions: This symptom is observed on a Cisco 4500 router.
Workaround: There is no workaround.
•
Symptoms: The data terminal ready (DTR) signal in a serial connection does not pulse for the configured length of time when the pulse-time interface configuration command is configured and the interface is reset.
Conditions: This symptom is observed on a Cisco router that has a serial interface and that supports the pulse-time interface configuration command. When the pulse-time interface configuration command is not working properly, the Cisco router may lose the capability to resynchronize external encryption equipment, or to cause dial-on-demand equipment to hangup current calls.
Workaround: There is no workaround.
•
Symptoms: A Cisco 7200 series may reload because of a bus error.
Conditions: This symptom is observed on a Cisco 7200 series that is running Cisco IOS Release 12.1(16) and that is configured for IP Security (IPSec) crypto.
Workaround: There is no workaround.
•
Symptoms: The ifIndex index is given as an index to the subinterface of the failing permanent virtual circuit (PVC) when the atmIntfPvcFailuresTrap notification is received.
Conditions: This symptom is observed on a Cisco 7200 series that is running Cisco IOS Release 12.1(5).
Workaround: Issue a Simple Network Management Protocol (SNMP) query on the atmCurrentlyFailingPVclTable table to identify the failing PVC.
•
Symptoms: Upstream control point-to-control point (CP-CP) sessions may flap and very high CPU utilization may occur on a Systems Network Architecture switching services (SNASw) router.
Conditions: This symptom is observed when two servers of a specific third- party vendor have the same logical unit (LU) name for a specific software application of the same third-party vendor. The symptom does not occur when other servers or other applications are used.
The SNASw router registers the LU under the first server. When the second server comes up, it sends a registry message to the SNASw router, specifying resource type end node control point (ENCP). The SNASw router replies with sense code 0x08890000 (LU-to-LU session sense code 0x1014023C). The second server then sends a second registry message to the SNASw router without including the duplicate LU name, but specifying in the resource type in control vector 3C that it is the network node control point (NNCP). Because the SNASw router does not expect a network node (NN) downstream, this second registry message causes the SNASw router to send an incorrect registry message to a virtual telecommunications access method (VTAM). The VTAM replies with sense code 0x08890100 and unbinds the CP-CP session.
The SNASw router then sends the incorrect registry message again, causing the CP-CP session to unbind again. This situation will repeat itself over and over again, causing very high CPU utilization in the SNASw router.
Reloading the SNASw router alleviates the situation, but only until the second server comes up again.
Workaround: There is no workaround. The duplicate LU name needs to be corrected in the servers of the third-party vendor.
•
Symptoms: As a dependent logical unit requester (DLUR), a Systems Network Architecture Switching Services (SNASw) switch must report all links to upstream nodes (end nodes or network nodes) in the initial topology database update (TDU) that it sends to the dependent logical unit sender (DLUS). Currently, the SNASw switch reports only uplinks to network nodes (NNs) and links to virtual routing nodes (VRNs). Links to upstream end nodes (ENs) must be included.
Conditions: This symptom is observed on an SNASw switch that is running Cisco IOS Release 12.2 T.
Workaround: Topology updates other than the initial TDU include ENs. Therefore, the link can be bounced to the EN to cause the DLUS to learn about the link after the initial topology has been reported.
•
Symptoms: Websites that require authentication may not be reachable.
Conditions: This symptom is observed when the content engine (CE) has the Authentication Bypass feature enabled and the CE is sending traffic (that the router has redirected to the CE) back to the router. When this behavior occurs, the router does not handle the authentication bypass return traffic correctly.
Workaround: Ensure that Cisco Express Forwarding (CEF) is enabled. If it is not possible to enable CEF, disable the bypass auth-traffic enable global configuration command on the CE.
•
Symptoms: The following error conditions may occur when an adjacent node connects to a Systems Network Architecture (SNA) switch:
–
–
–
If any of these errors occur, the SNA switch may not recover and all subsequent CP-CP session activations from that adjacent node may be rejected with a "08120010" sense code.
Conditions: This symptom occurs when an adjacent node generates one of the three errors that are specified in the symptoms statement above.
Workaround: Perform an initial program load (IPL) of the SNA switch.
•
Symptoms: The following error message may be displayed when a multilink interface is shut down:
CPUHOG. %SYS-3-CPUHOG: Task ran for 2480 msec (3/2), process = MultilinkConditions: This symptom is observed when the state of a multilink interface changes, such as when the multilink interface comes up or goes down.
Workaround: There is no workaround for the CPU hog condition. However, if any application or routing protocol is affected by this CPU hog condition in the form of timeouts, the timers for the application or routing protocol can be incremented to workaround the CPU hog condition.
•
Symptoms: A Systems Network Architecture Switching Services (SNASw) router may stop forwarding BIND and UNBIND requests.
Conditions: This symptom is observed after the SNASw router receives an activate logical unit (ACTLU) request that is forwarded to a downstream physical unit (DSPU). This symptom will occur if the DSPU sends the ACTLU response back to the SNASw router without setting the response bit in the SNA request and resources header (RH).
Workaround: Restart the SNASw router.
•
Cisco routers and switches running Cisco IOS software and configured to process Internet Protocol version 4 (IPv4) packets are vulnerable to a Denial of Service (DoS) attack. A rare sequence of crafted IPv4 packets sent directly to the device may cause the input interface to stop processing traffic once the input queue is full. No authentication is required to process the inbound packet. Processing of IPv4 packets is enabled by default. Devices running only IP version 6 (IPv6) are not affected. A workaround is available.
Cisco has made software available, free of charge, to correct the problem.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20030717-blocked.shtml
•
Symptoms: A router may run out of crypto_epa_blk memory blocks. The router may prevent the further establishment of crypto connections and the reestablishment of existing connections that have timed out.
Conditions: These symptoms are observed on a Cisco router that is operating under stress conditions with Cisco Encryption Technology (CET) tunnels. These symptoms are specific to hardware Encryption Service Adapters (ESAs) and do not affect software crypto engines.
Workaround: Reboot the router.
•
Symptoms: An Encryption Service Adapter (ESA) card (hardware crypto accelerator) may run out of high memory area (HMA) memory blocks.
Conditions: This symptom is observed when there are repeated failed attempts to establish a Cisco Encryption Technology (CET) connection.
Workaround: Use software encryption.
Protocol Translation
•
Symptoms: A router may reload with a bus error at a null point.
Conditions: This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1(6) or Release 12.1(9) and that is configured for protocol translation.
Workaround: There is no workaround.
For further information about bus errors, refer to the Troubleshooting Bus Error Crashes document at the following location:
http://www.cisco.com/warp/public/122/crashes_buserror_troubleshooting.shtml
Wide-Area Networking
•
Symptoms: A Cisco 7200 series that is acting as a Layer 2 Tunneling Protocol (L2TP) network server (LNS) may reload with an error interrupt.
Conditions: This symptom is observed when the Cisco 7200 series is receiving malformed packets that have invalid payload data (such as User Datagram Protocol [UDP] packets).
Workaround: There is no workaround.
•
Symptoms: A router that is used as a Layer 2 Tunneling Protocol (L2TP) network server (LNS) may reload because of redzone I/O memory corruption.
Conditions: This symptom is observed on a Cisco router when the LNS receives PPP packets that have invalid lengths.
Workaround: There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(18)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(18). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(18). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms A router that is configured for TACACS+ may run out of memory because of a buffer leak in the middle buffer pool that is caused by TACACS+ packets. The occurrence of this symptom can be verified by entering the show buffers EXEC command or the show tcp brief all EXEC command on the router.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.0(15).
Workaround Reload or power-cycle the router to free the buffers and memory.
•
Symptoms The cache error recover function (CERF) is disabled after a Cisco 7200 series router is reloaded. This symptom is observed after CERF is enabled, written into the startup configuration, and the router is reloaded.
The output of the show memory cache error-recovery EXEC command may indicate that the commands are disabled after the router is reloaded:
no memory cache error-recovery L3 data
no memory cache error-recovery options nvram-report
no memory cache error-recovery options parity-check
memory cache error-recovery options window 0
memory cache error-recovery options max-recoveries 0Conditions This symptom is observed on a Cisco 7200 series router that is using a Network Processing Engine (NPE-300) that has 32 MB of memory in the dual in-line memory module (DIMM2).
Workaround Install 64 MB of memory in the DIMM2.
•
Symptoms TACACS+ Microsoft Challenge Handshake Authentication Protocol (MS-CHAP) authentication may reload.
Conditions This symptom is observed when an incorrect key is configured while the debug tacacs EXEC command is enabled.
Workaround Ensure that the correct key is configured or avoid enabling the debug tacacs EXEC command.
•
Symptoms If a short and nonspecific dnis-number is specified in the aaa dnis map dnis-number authorization network group server-group-name global configuration command to map a dialed number identification service (DNIS) number to a particular authentication, authorization, and accounting (AAA) server group after a longer dnis-number (that shares the same first digits with the short nonspecific dnis-number) has been previously configured by entering the same command, the aaa dnis map dnis-number authorization network group server-group-name global configuration command that has the short and nonspecific dnis-number may not work as expected.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1, Release 12.2, or Release 12.2 T.
Workaround Enter the aaa dnis map dnis-number authorization network group server-group-name global configuration command that has the short and nonspecific dnis-number before entering the same command for the configuration that has the longer dnis-number.
IBM Connectivity
•
Symptoms Logical link control (LLC) circuits that are owned by an "active" data-link switching (DLSw) services Ethernet Redundancy (ER) router may be disconnected when service is restored on a failed DLSw ER router.
Conditions This symptom is observed in a failover scenario when DLSw ER is used on a router that is running Cisco IOS Release 12.2(10b). This symptom is observed only when the same MAC address is mapped to different Ethernet interfaces that are in different DLSw ER segments on two separate routers.
Workaround Map both of the MAC addresses on the same router to different Ethernet interfaces.
Interfaces and Bridging
•
Symptoms A single-port Fast Ethernet 100BASETX port adapter (PA-FE-TX) on a Cisco 7206VXR router that has a Network Processing Engine (NPE-300) may stop receiving burst traffic packets.
Conditions This symptom is observed on a PA-FE-TX of a Cisco 7206VXR that has an NPE-300.
Workaround This symptom can be cleared by entering the shutdown interface configuration command followed by the no shutdown interface configuration command on the PA-FE-TX interface.
•
Symptoms Packet drops may be observed.
Conditions This symptom is observed on a channelized T1 (CT1) interface between a provider edge router (PE) and a customer edge (CE) router.
Workaround There is no workaround.
•
Symptoms An Ethernet driver on an Ethernet interface may receive and forward packets that are not destined for itself.
Conditions This symptom is observed on an Ethernet interface that has the promiscuous mode enabled in a network that has multiple Hot Standby Router Protocol (HSRP) groups. This symptom is also observed when no transparent bridging is occurring.
Workaround There is no workaround.
•
Symptoms A router may reload after a channelized T3 (CT3) port adapter that is configured as part of a Multilink PPP (MLP) bundle is removed, and the MLP bundle interface is shut down.
Conditions This symptom is observed in a network in which two Cisco 7200 series routers are connected back-to-back via channelized T3 (CT3) port adapters. Channel groups are created and configured for MLP, and a bundle interface multilink is created on both of the routers in this setup.
Workaround There is no workaround.
•
Symptoms Open Shortest Path First (OSPF) multicast packets are not received on a 1-port Fast Ethernet 100BASE-TX port adapter (PA-FE-TX).
Conditions This symptom is observed on a PA-FE-TX port adapter on a Cisco 7500 router that is configured with OSPF. The PA-FE-TX does not receive OSPF multicast traffic because MAC multicast entries are not added to the MAC table.
Workaround There is no workaround.
IP Routing Protocols
•
Symptoms Enhanced Interior Gateway Routing Protocol (EIGRP) neighbors may be reset unnecessarily on an interface that is configured with summarization. The following output is displayed when this symptom occurs:
%DUAL-5-NBRCHANGE: IP-EIGRP 111: Neighbor 10.108.255.244 (FastEthernet4/0) is down: Summary up, remove external
%DUAL-5-NBRCHANGE: IP-EIGRP 111: Neighbor 10.108.255.245 (FastEthernet4/0) is down: Summary up, remove externalConditions The EIGRP neighbors are reset unnecessarily, and the summary is regenerated on an interface if all components of the summary are lost and at least one component is relearned.
Workaround Remove EIGRP summary statements from the interface.
Miscellaneous
•
Symptoms Committed Access Rate (CAR) limits do not function properly.
Conditions This symptom is observed when Cisco Express Forwarding (CEF) is enabled on a router. When CEF is disabled on a router, CAR limits function properly. (This configuration is not supported.) This behavior has been observed on routers that use the same interface for input and output traffic.
Workaround There is no workaround.
•
Symptoms If the write memory EXEC command is issued simultaneously with the show config privileged EXEC command or the show running-config EXEC command via two individual Telnet sessions by two different users, output similar to the following may be displayed:
bGc nx ^@^@^@^A^A^A^@^@^A^@^@^E^@^@^@^@^@^@^@^@^@^@^^@^@^@^@^@^@^@^A^@^@^@^@^@^@^@^@^ @^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^Conditions This symptom is observed on a Cisco 7500 router that is running Cisco IOS Release 12.1(10)E.
Workaround There is no workaround.
•
Symptoms Several "RX FIFO was stuck - forced to reset MAC" messages may be logged on the console of a router. This message is specific to port adapters and I/O cards that use a vendor-specific chipset.
Conditions This symptom is observed on a Cisco 7200 router that is operating in the normal mode. The following is a list of the affected port adapters and I/O cards:
–
–
–
–
Workaround There is no workaround.
•
Symptoms The Web Cache Communication Protocol (WCCP) stops on some VLANs.
Conditions This symptom is observed when Cisco Express Forwarding (CEF) is enabled on a Cisco 5000 series Route Switch Feature Card (RSFC). WCCP redirection does not occur on interfaces that have CEF enabled even if a WCCP redirect statement is present.
Workaround Disable CEF on interfaces that have a redirect statement. This workaround may impact the performance of the Cisco 5000 series and should be considered carefully before it is applied.
•
Symptoms If two active lines that are connected to ISDN BRI interfaces on a 4-port ISDN BRI network module (NM-4B-S/T) are unplugged simultaneously, one of the ISDN BRI interfaces may display the Layer 1 status as "DEACTIVATED" when the show isdn status EXEC command is entered.
Conditions This symptom is observed on a Cisco 2600 series router that has a NM-4B-S/T network module and that is running Cisco IOS Release 12.1(16).
Workaround Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the ISDN BRI interface that exhibits this symptom.
•
Symptoms A Systems Network Architecture Switching Services (SNASw) boundary function does not unbind a primary logical unit-secondary logical unit (PLU-SLU) session when a cold activate logical unit (ACTLU) response (RSP) is received. The PLU-SLU session on the downstream physical unit (DSPU) cannot be restarted because the virtual telecommunications access method (VTAM) and SNASw are not in agreement about the state of the PLU-SLU session with the DSPU.
Conditions This symptom is observed on a SNASw switch. When a DSPU sends a cold ACTLU RSP to a dependent logical unit requester (DLUR), it indicates that a PLU-SLU session is over and the cold ACTLU response should not be used.
Workaround Restart SNASw or the DSPU.
•
Symptoms Physical units remain in the Pend Activate Physical Unit (ACTPU) state, and the show snasw pu EXEC command shows several downstream physical units (DSPUs) that have the same IDBLK/IDNUM physical unit identifier. The virtual telecommunications access method (VTAM) may be slow to send a response to the Request Activate Physical Unit (REQACTPU), and the DSPU disconnects and reconnects before the response arrives from VTAM. When the DSPU reconnects, Systems Network Architecture (SNA) Switching Services (SNASw) may treat the DSPU as a new DSPU. When VTAM sends the REQACTPU, SNASw may assume that there are two DSPUs with the same name and rejects the request with a 082C 002 sense code.
Conditions This symptom is observed when the user is waiting for the REQACTPU from the VTAM.
Workaround There is no workaround.
•
Symptoms A Cisco 3662-AC versatile multiservice access platform returns a MIB value of "c3660(92)" for the chassis type in the OLD-CISCO-CHASSIS-MIB. The "3660(92)" chassis type is not recognized in the Cisco Element Management Framework (CEMF), and the chassis is not recognized and cannot be deployed using CEMF.
Conditions This symptom is observed on a Cisco 3662-AC that is running Cisco IOS Release 12.1(16).
Workaround There is no workaround.
•
Symptoms An interface on a 2-port Fast Ethernet port adapter (PA-2FE) may stop transmitting if this interface or the other interface on the same port adapter goes down or flaps under a heavy traffic load. The interface that stops transmitting may display the following messages:
%RSP-3-RESTART: interface FastEthernet3/0/0, not transmitting %RSP-3-RESTART: interface FastEthernet3/0/0, output frozen %RSP-3-RESTART: cbus complexConditions This symptom is observed on the 2-port Fast Ethernet port adapter (PA-2FE) on a Cisco 7500 series Versatile Interface Processor 4 (VIP4) that is configured with Fast Ether Channel (FEC). This symptom is observed when the port adapter is carrying a heavy traffic load and when part of the traffic is originating from a port adapter (PA-A3) that is located on the same VIP.
Workaround There is no workaround.
•
Symptoms Packets that are switched from an incoming X.21 interface to an E1 channelized interface may not be sent. Packets that are switched the other way around from an E1 channelized interface to an X.21 interface are sent.
Conditions This symptom is observed on a Cisco 7200 series router.
Workaround There is no workaround.
•
Symptoms A loss of connectivity may be observed on an RSETUP High-Performance Routing (HPR) pipe to a vendor-specific open systems adapter.
Conditions This symptom is observed in a network in which Systems Network Architecture (SNA) switch routers are connected to Catalyst 6500 series switches via Fast Ethernet ports. The SNA switch routers are connected to the vendor-specific open systems adapter via an enterprise extender.
The virtual telecommunications access method (VTAM) on the vendor-specific open systems adapter terminates the pipe, but the SNA switch does not terminate the pipe. This behavior causes the pipe to enter into an invalid state and prevents anything that is sent over it from making it to the VTAM. Consequently, session setup requests will hang.
Workaround There is no workaround.
Novell IPX, XNS, and Apollo Domain
•
Symptoms On a router, the following message is displayed:
%IPX-3-TOOMANYNETS: Too many networksConditions This symptom is observed if the number of interfaces that are running the Internetwork Packet Exchange (IPX) protocol exceeds 200. This combination may include a variety of interfaces that are running the Routing Information Protocol (RIP), the Enhanced Interior Gateway Routing Protocol (EIGRP), or the NetWare Link Services Protocol (NLSP). However, if an interface is running both RIP and EIGRP simultaneously, it is considered to be running two protocols instead of one. This means that the 200 limit would be reached if there are 100 interfaces running both RIP and EIGRP.
Workaround On an interface that is running both EIGRP and RIP, remove either one of the two protocols. Enter the no network network-number DHCP pool configuration command immediately after the ipx router rip global configuration command in the startup-config file of the router where the interface is installed.
TCP/IP Host-Mode Services
•
Symptoms When two Cisco routers are running data-link switching (DLSw) and are peered to the same 190 remote sites over a Frame Relay network through a High-Speed Serial Interface (HSSI), about 80 of the 190 DLSw peers disconnect and reconnect at irregular intervals.
Some peers stay up for several hours, while others disconnect and reconnect frequently. The DLSw peers are disconnecting because the TCP stack has reached its retransmit threshold. When this condition occurs, extended pings (pings that are sourced by the IP address of the DLSw peer) to the IP address of the remote DLSw peer that is experiencing connectivity issues are consistently successful.
During successive retransmission timeout, the timer receives a negative timeout value. This situation causes the packets to be on the retransmit queue for as long as the new round-trip timeout (KRTT [using the Karn algorithm]) value even though the packets have been acknowledged.
If TCP transaction debugging is enabled, the following error message is displayed when this symptom occurs:
Received a negative sleep value:<value>Conditions This symptom is observed in a configuration in which two Cisco 7507 routers are running Cisco IOS Release 12.1(13). Each Cisco 7507 router has a Channel Interface Processor 2 (CIP 2) that is connected to a mainframe. This symptom may also be observed on Cisco 3600 series, Cisco 4700 series, or Cisco 7200 series routers. This symptom is observed only on Cisco IOS releases that contain the fix for CSCdu18397 but not the fix for CSCdw58350.
Workaround Use a Cisco IOS release such as Cisco IOS Release 12.2(10b).
Wide-Area Networking
•
Symptoms A universal access server may reload because of a memory corruption.
Conditions This symptom is observed on a Cisco AS5300 universal access server that is running Cisco IOS Release 12.1, Release 12.2, or Release 12.2(2)XB. The memory corruption occurs only on a virtual private dial-up network (VPDN) network access server (NAS) when Layer 2 Forwarding (L2F)-encapsulated IP packets are reencapsulated in another L2F tunnel (VPDN packets that are switched using the Stack Group Bidding Protocol (SGBP) in a multichassis-Multilink PPP [MLP] environment).
Workaround Use the sgbp protocol l2tp global configuration command to configure the router to use the Layer 2 Tunneling Protocol (L2TP) as the encapsulation protocol for packets that are forwarded by SGBP.
Resolved Caveats—Cisco IOS Release 12.1(17)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(17). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(17). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms A Cisco 7200 series router reloads because of a bus error, a watchdog timeout, or an unspecified error (that is, no reload error message is provided).
Conditions This symptom is observed on a Cisco 7200 series router that is configured with a Network Processing Engine 200 (NPE-200), a PA-A1 port adapter, and a PA-FE port adapter when a medium to high traffic load occurs and the traffic load consists of large packet sizes.
Workaround options
–
–
–
•
Symptoms A CPU hog condition may be observed on a router, and the router may reload.
Conditions These symptoms are observed on a Cisco router that is running Cisco IOS Release 12.1 and that has several thousands of logical entities configured, when the snmp-server community global configuration command is executed.
Workaround There is no workaround.
•
Symptoms Simple Network Management Protocol (SNMP) version 2c security-to-group mappings may remain after an SNMP community is deleted.
Conditions This symptom is observed on a Catalyst 6000 series switch that is running Cisco IOS Release 12.1(3).
Workaround Delete the stale mapping using SNMP.
IBM Connectivity
•
Symptoms A router may reload when the dlsw transparent redundancy-enable interface configuration command is removed from and reapplied to the Ethernet interface.
Conditions This symptom is observed when the dlsw transparent redundancy-enable interface configuration command is removed from and reapplied to the Ethernet interface on a Cisco router that is using data-link switching (DLSw) Ethernet redundancy while there may be multiple circuits between the same pair of MAC addresses that are on different service access points (SAPs).
Workaround Use DLSw with transparent bridging instead of using the DLSw Ethernet redundancy feature.
•
Symptoms Cisco Express Forwarding (CEF) does not route packets to a Channel Interface Processor (CIP) Common Link Access for Workstations (CLAW) backup host.
Conditions This symptom is observed on a Cisco 7500 series router that is running Cisco IOS Release 12.1(15). The CIP in the setup is configured using CLAW backup.
This symptom does not occur when there is an active CLAW connection to the primary host; rather, occurs when there is an active CLAW connection to the backup host. IP connectivity to the host IP address fails when an attempt is made to traverse an ingress interface that has CEF enabled. Full IP connectivity is restored if CEF is disabled on the ingress interface. The host IP address can be pinged from the CIP router but not from another device that is attached to an interface on the CIP router that has CEF enabled.
Workaround Disable CEF by entering the no ip route-cache cef global configuration command on the ingress interface.
•
Symptoms A Synchronous Data Link Control (SDLC) session enters the down state (no response and no polling) after an I-Frame is sent. SDLC sessions pause indefinitely if the simultaneous keyword is used in the interface configuration for the serial SDLC interface. No polling is observed when this symptom occurs.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.2(10).
Workaround Do not use the simultaneous keyword in the configuration of the SDLC serial interface.
•
Symptoms A router may unexpectedly remove unrelated data-link switching (DLSw) circuits on other serial interfaces.
Conditions This symptom is observed on a router that is using DLSw with Synchronous Data Link Control (SDLC) on attached serial interface controllers.
Workaround There is no workaround.
Interfaces and Bridging
•
Symptoms A Versatile Interface Processor (VIP) may reload because of an error at the ct3sw_check_tx process.
Conditions This symptom is observed on a Cisco 7000 series VIP that is running Cisco IOS Release 12.0(16)S4.
Workaround There is no workaround.
•
Symptoms An autoinstall feature may not function properly.
Conditions This symptom is observed when a Cisco 7204VXR router is autoinstalled with a T3 connection.
Workaround There is no workaround.
•
Symptoms A single-port Fast Ethernet 100BASE-TX port adapter (PA-FE-TX) on a Cisco 7206VXR router that has a Network Processing Engine (NPE-300) may stop receiving burst traffic packets.
Conditions This symptom is observed on a PA-FE-TX on a Cisco 7206VXR that has a Network Processing Engine (NPE-300).
Workaround This symptom can be cleared by entering the shutdown interface configuration command followed by the no shutdown interface configuration command on the PA-FE-TX interface.
•
Symptoms The following error message may be displayed on a router when a Cisco IOS software upgrade is performed:
%SYS-6-STACKLOW: Stack for process ATM PA Helper running low, 0/3000Conditions This symptom is observed on a Cisco 7500 series router when the Cisco IOS software is upgraded from Cisco IOS Release 12.0(7)T to Release 12.1(14).
Workaround There is no workaround.
•
Symptoms A router that is configured with a multichannel port adapter reloads because of a bus error exception.
Conditions This symptom is observed when link flaps occur or interfaces are reset on a router that is configured with PA-MC-T1, PA-MC-E1, PA-MC-E3, or PA- MCX port adapters.
Workaround There is no workaround.
•
Symptoms An Ethernet driver on an Ethernet interface may receive and forward packets that are not destined for itself.
Conditions This symptom is observed on an Ethernet interface that has the promiscuous mode enabled in a network that has multiple Hot Standby Router Protocol (HSRP) groups. This symptom is also observed when no transparent bridging is occurring.
Workaround There is no workaround.
•
Symptoms A router may reload after a channelized T3 (CT3) port adapter that is configured as part of a Multilink PPP (MLP) bundle is removed, and the MLP bundle interface is shut down.
Conditions This symptom is observed in a network in which two Cisco 7200 series routers are connected back-to-back via channelized T3 (CT3) port adapters. Channel groups are created and configured for MLP, and a bundle interface multilink is created on both of the routers in this setup.
Workaround There is no workaround.
•
Symptoms Open Shortest Path First (OSPF) multicast packets are not received on a 1-port Fast Ethernet 100BASETx port adapter (PA-FE-TX).
Conditions This symptom is observed on a PA-FE-TX port adapter on a Cisco 7500 series router that has OSPF configured. The PA-FE-TX does not receive OSPF multicast traffic because MAC multicast entries are not added to the MAC table.
Workaround There is no workaround.
IP Routing Protocols
•
Symptoms An Address Resolution Protocol (ARP) entry is not removed if the no arp ip-address mac-address global configuration command is issued. The MAC address of the Ethernet interface that has the IP address may be changed.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.2(7a).
Workaround Enter the clear arp EXEC command after removing a static ARP entry.
•
Symptoms A high CPU utilization condition may be observed on a router.
Conditions This symptom is observed on a Cisco router that is running the Enhanced Interior Gateway Routing Protocol (EIGRP) and that is running Cisco IOS Release 12.1(13). The high CPU utilization condition may trigger other symptoms such as the loss of EIGRP neighbor relations on the router. This symptom may rectify itself without any user intervention.
Workaround There is no workaround.
Miscellaneous
•
Symptoms On a router that has static Address Resolution Protocol (ARP) entries configured, the router may fail to use the static ARP entries immediately.
Conditions This symptom is observed on a Cisco router if CEF is enabled manually or if the router is reloaded with Cisco Express Forwarding (CEF) enabled.
Workaround Disable CEF on the router by entering the no ip cef global configuration command.
•
Symptoms A router may stop sending traffic if the microcode is reloaded while the router is forwarding traffic.
Conditions This symptom is observed on a Cisco 7500 series Versatile Interface Processor (VIP4) that has a 2-port Fast Ethernet port adapter (PA-2FE). This symptom can be resolved temporarily by entering the shutdown interface configuration command followed by the no shutdown interface configuration command on the PA-2FE interface.
Workaround Reload the microcode while there is no egress traffic on the router.
•
Symptoms The E1 controller displays inaccurate statistics after the show controllers [e1 | t1] EXEC command is issued. The following command output shows that the elapsed seconds and the unavailable counters are do not advance:
Timestamp - 00:00 E1 3/4 is up. Applique type is Channelized E1 - balanced Framing is UNFRAMED, Line Code is HDB3, Clock Source is Line. 0 Errored Secs, 0 Bursty Err Secs, 0 Severely Err Secs, 270 Unavail Secs 0 Errored Secs, 0 Bursty Err Secs, 0 Severely Err Secs, 86400 Unavail Secs
Timestamp - 01:50 E1 3/4 is up. Applique type is Channelized E1 - balanced Framing is UNFRAMED, Line Code is HDB3, Clock Source is Line. 0 Errored Secs, 0 Bursty Err Secs, 0 Severely Err Secs, 270 Unavail Secs 0 Errored Secs, 0 Bursty Err Secs, 0 Severely Err Secs, 86400 Unavail SecsConditions This symptom is observed when an E1 controller is configured for unframed operations using the controller {t1 | e1} slot/port channel-group 0 unframed command. The counters are correct when the controller is configured for a Frame Relay operation (CRC-4 or No-CRC4).
Workaround There is no workaround.
•
Symptoms A router may reload with the following error message: System was restarted by bus error at:
PC 0x60A9DBF8, address 0xD0D0D2D at Image text-base: 0x60008950, data-base: 0x61542000 0x60A9DBF8 x60A9DF38 0x60A8AC18 0x60A8B1D8 0x60A8B374 0x603FC5B4 0x603FC5A0Conditions This symptom is observed on a Cisco 7140 router that is running the c7100-io3s-mz.121-12 image of Cisco IOS Release 12.1(12).
Workaround There is no workaround.
•
Symptoms Modem ISDN channel aggregation (MICA) technologies modules may become bad and later recover without either the use of modem recovery or any user intervention.
Conditions This symptom occurs under normal conditions in which calls are received and disconnected at a regular 30-minute intervals.
Workaround There is no workaround.
•
Symptoms A Versatile Interface Processor (VIP4-80) may reload with a bus error when distributed Multilink PPP (dMLP) is configured.
Conditions This symptom occurs when traffic is passed through the dMLP bundle and occurs just after the interface comes up. This symptom is specific to the dMLP feature and will occur only if dMLP is configured on a platform. The dMLP feature is independent of other features and does not affect other features.
Workaround There is workaround.
•
Symptoms The first information frame of the user logical unit-logical unit (LU-LU) session is sent with both the source MAC address and the destination MAC address in the non-canonical (Token Ring) format.
Conditions This symptom is observed if Systems Network Architecture Switching Services (SNASw) is configured with High Performance Routing (HPR) Logical Link Control (LLC1) frame over data-link switching (DLSw) for the uplinks and Ethernet is used for the downstream connection. This symptom will occur regardless of whether the downstream port uses the address of the Ethernet interface or the address of the Hot Standby Router Protocol (HSRP) interface. This symptom does not occur if the downstream port is a Token Ring, virtual Token Ring, or a virtual data-link control (VDLC) port. This symptom does not occur if the uplink uses LLC2 port definition.
Workaround There is no workaround.
•
Symptoms A physical unit (PU) that is connected downstream to a Systems Network Architecture Switching Services (SNASw) router may enter the reset state when another PU connects to it using the same cpname and same IDBLK/IDNUM physical unit identifier. The reset state of the PU is indicated in the output of the show snasw pu EXEC command. The problem determination log (PDLOG) will display an "invalid internal state detected" message when this symptom occurs.
Conditions This symptom is observed on a Cisco router. The SNASw router has to be restarted to clear this symptom.
Workaround Use the dyncplen connection type.
•
Symptoms Sessions that have a destination logical unit that is a low entry networking (LEN) control point (CP) downstream from a Systems Network Architecture Switching Services (SNASw) branch network node (BrNN) may fail with a sense code of 087D0001. The LEN CP does not show up in the directory database on the host network node server or on the BrNN.
Conditions This symptom is observed on a SNASw switch.
Workaround Add a location statement for the LEN CP to the SNASw configuration.
•
Symptoms A Systems Network Architecture Switching Services (SNASw) boundary function does not unbind a primary logical unit-secondary logical unit (PLU-SLU) session when a cold activate logical unit (ACTLU) response (RSP) is received. The PLU-SLU session on the downstream physical unit (DSPU) cannot be restarted because the virtual telecommunications access method (VTAM) and SNASw are not in agreement about the state of the PLU-SLU session with the DSPU.
Conditions This symptom is observed on a SNASw switch. When an old Downstream Physical Unit (DSPU) sends a cold ACTLU RSP to a dependent logical unit requester (DLUR), it indicates that a PLU-SLU session is over and the cold ACTLU response should not be used.
Workaround Restart SNASw or the DSPU.
•
Symptoms A Cisco 3662-AC versatile multiservice access platform returns a MIB value of "c3660(92)" for the chassis type in the OLD-CISCO-CHASSIS-MIB. The "3660(92)" chassis type is not recognized in the Cisco Element Management Framework (CEMF), and the chassis is not recognized and cannot be deployed using CEMF.
Conditions This symptom is observed on a Cisco 3662-AC that is running Cisco IOS Release 12.1(16).
Workaround There is no workaround.
•
Symptoms A back-to-back ping for multilink fails.
Conditions This symptom is observed when you have the multilink min- links links command configured on the multilink interface.
Workaround Enter the shutdown interface configuration command on the multilink interface, wait for the multilink and the member links to go down, and enter the no shutdown interface configuration command on the multilink interface. After the multilink comes up, the back-to-back ping for multilink will work.
•
Symptoms Pings cannot be sent between routers.
Conditions This symptom is observed when a ping is sent from a customer edge (CE) router to another CE router via a provider edge (PE) router. This symptom occurs if the routers are configured using scripts.
Workaround There is no workaround.
•
Symptoms When member links are removed from a multilink bundle (M2) and configured as members of another multilink bundle (M5), the M5 multilink bundle does not come up.
Conditions This symptom is observed only with the distributed Multilink PPP (MLP) feature when a member link is reconfigured to be a member link of another multilink bundle. This symptom is observed only on Cisco 7500 series and Cisco 7600 series routers.
Workaround There is no workaround.
•
Symptoms A FlexWAN module may reload if the member link of a multilink bundle is removed by entering the no channel-group interface configuration command.
Conditions This symptom is observed on a Cisco 7600 series router that has a FlexWAN module when distributed Multilink PPP (MLP) is configured.
Workaround Do not remove the member link using the no channel-group interface configuration command when the member link is in the UP state. Instead shutdown the member link first, remove the multilink configuration for the member link, and remove the member link by entering the no channel-group interface configuration command.
TCP/IP Host-Mode Services
•
Symptoms TCP processing fails in the data repacketized process and creates inaccurate packets. A data-link switching (DLSw) circuit disconnects suddenly, and the following error message is displayed if you enter the debug dlsw core command on one for the DLSw routers:
DLSW: Invalid dlsw version 78The number 78 is an example and may be any other number.
Conditions These symptoms are observed during an attempted TCP retransmission of a message and TCP data packet reconstruction on a DLSw router.
Workaround There is no workaround.
•
Symptoms A memory leak may occur on a router after TCP-to-X.25 translation is configured.
Conditions This symptom is observed if a user attempts to use TCP-to-X.25 translation while a router is already performing translation for the maximum number of configured users. The additional user will not be able to use translation, and the router will leak memory.
Workaround There is no workaround.
Wide-Area Networking
•
Symptoms Inbound data packets that are reassembled from multilink fragments may not be processed properly on Multilink PPP (MLP) interfaces that are receiving encrypted IP Security (IPSec) traffic that is terminated locally when a hardware accelerator is used for decryption.
Conditions This symptom affects all inbound reassembled data frames that are received by the bundle and not just those data frames that are carrying encrypted IP datagrams. Most significantly, inbound Internet Security Association and Key Management Protocol (ISAKMP) keepalives are not processed, leading to the eventual failures of the associated IPSec sessions.
The IPSec sessions are reestablished after each failure, but traffic drops will occur until the session is renegotiated via the Internet Key Exchange (IKE). Thus, the observable symptoms are an intermittent failure of IPSec sessions combined with high loss rates in the encrypted data traffic.
Workaround Disable hardware crypto acceleration, and use software crypto acceleration instead.
•
Symptoms The Layer 2 Tunneling Protocol (L2TP) network server (LNS) reloads after the L2TP access concentrator (LAC) sends a Call Disconnect Notification (CDN).
Conditions This symptom is observed on an LNS that is configured to call back the dial-in user when the string that is sent by the LNS during a callback is the correct string, but the LAC uses a different string to dial out.
Workaround There is no workaround.
•
Symptoms Calls that have unsupported traffic parameters are released when a Cisco router is running User-to-Network Interface (UNI) version 4.0 software with a vendor-specific switch.
Conditions This symptom is observed when a Cisco router is connected to a vendor-specific switch and if the router is configured to operate UNI version 4.0 software across the unnegotiated router-switch link. The calls are released by the Cisco router if any traffic parameter is modified on the unnegotiated link.
Workaround Use UNI version 3.1.
Resolved Caveats—Cisco IOS Release 12.1(16)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(16). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(16). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms A router may reload with a bus error.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 11.2(18)P, 12.0(10), 12.0(5)XK1, 12.1(3)T, or 12.1(4).
Workaround There is no workaround.
•
Symptoms A router may reload after the frame-relay payload-compress frf9 stac interface configuration command is configured on a port adapter interface.
Conditions This symptom is observed on a Cisco 7500 router that has a 4-port serial port adapter (PA-4T+) and that is running the Route Switch Processor (RSP) software RSP-ISV-M of Cisco IOS Release 12.1(10)E. This symptom does not occur if a Fast Serial Interface Processor (FSIP) is used.
Workaround Use a FSIP or enter the frame-relay payload-compression frf9 stac software interface configuration command on the serial port adapter interface.
•
Symptoms A router may reload when the Simple Network Management Protocol (SNMP) is used to set the smonVlanIdStatsTable element.
Conditions This symptom is observed on a Cisco router only when the 64-bit counter is set to certain values.
Workaround Disable SNMP.
•
Symptoms An ISDN interface may remain in the "ESTABLISH_AWAITING_TEI" state.
Conditions This symptom is observed on a Cisco 1604 router that has a BRI (U) interface. The BRI U interface will not come up nor will it pass any traffic.
Workaround There is no workaround.
•
Symptoms Service Assurance Agent (SAA) latency measurements may show unrealistic spikes.
Conditions This symptom is observed on a Cisco router when the Border Gateway Protocol (BGP) is operating at moderate to full capacity.
Workaround Enable the rtr responder global configuration command on the remote Cisco router and use the jitter probe.
•
Symptoms Packets may be dropped on a Fast Ethernet port.
Conditions This symptom is observed on a Fast Ethernet port that is installed on a Catalyst 6000 switch that is running Cisco IOS Release 12.1 and that is connected to the Internet. This symptom does not occur when NetFlow is enabled on a Gigabit Ethernet interface. In the affected setup, a Content Switching Module (CSM) that is on the Catalyst 6000 is used to perform Network Address Translation (NAT). This symptom may affect only traffic that is sent through the CSM for NAT when NetFlow is enabled.
Workaround Disable Cisco Express Forwarding (CEF) on the interface that has the ip route-cache flow interface configuration command and the ip nat outside interface configuration command configured. Packets are not dropped when fast switching is used; therefore, fast switching may be used in place of CEF.
IBM Connectivity
•
Symptoms A small buffer leak may occur on a router.
Conditions This symptom is observed on a Cisco router that is running data-link switching (DLSw) local conversion from Synchronous Data Link Control (SDLC) to Logical Link Control, type 2 (LLC2). This condition was seen on a router only when several multidropped SDLC controllers were configured.
Workaround Ensure that there are no Physical Unit 2.1 type (PU 2.1) devices that have an exchange identification (XID) configured but that are not installed. If a specific controller is not installed, remove the configuration for the controller from the router.
•
Symptoms A router may reload and display the following error message:
%SYS-2-MALLOCFAIL: Memory allocation of 140 bytes failed from 0x602EAAA4, alignment 0 Pool: Processor Free: 0 Cause: Not enough free memoryConditions These symptoms are observed on a Cisco router that is running Cisco IOS Release 11.2(18)P but has recently been upgraded to Cisco IOS Release 12.1 (12b).
Workaround Reload the router.
•
Symptoms A software-forced reload may occur on a router.
Conditions This symptom is observed on a Cisco 7200 router and is specific to a configuration in which a central router that is running data-link switching plus (DLSw+) is receiving a unnumbered information frame (UI-frame) such as destination service access point (DSAP) AA or source service access point (SSAP) AA from a DLSw+ remote peer. The reload occurs under certain conditions such as when the central router is computing an internal variable incorrectly. The occurrence of this symptom is specific to a DLSw+ configuration.
Workaround Configure the dlsw icannotreach saps aa global configuration command on the central DLSw+ router. This command will eliminate UI-frames on service access points (SAPs) AA from the DLSW network.
Interfaces and Bridging
•
Symptoms An IP ping does not go through on the bridging and the bridging-to-routing path.
Conditions This symptom is observed in an integrated routing and bridging (IRB) environment. Pings can be sent through the routing and the routing-to-bridging path, but pings cannot be sent through on the bridging and the bridging-to-routing path.
Workaround There is no workaround.
•
Symptoms A router may reload immediately after the card type is configured for a port adapter.
Conditions This symptom is observed on a Cisco 7500 series router that has an 8-port channelized T1/E1 PRI, 256-channel port adapter.
The Cisco 7500 reloads if an online insertion and removal (OIR) is performed on the Versatile Interface Processor (VIP) on which the port adapter is installed after the card type is configured on the port adapter.
Workaround There is no workaround.
IP Routing Protocols
•
Symptoms CPU utilization on a router may reach 100 percent.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1 when Network Address Translation (NAT) is configured and there are 7000 NAT entries at 3 packets per second (pps).
Workaround Use Cisco IOS Release 12.2.
•
Symptoms A switch feature card may reload with a bus error.
Conditions This symptom is observed when a link-state advertisement (LSA) that has an incontiguous mask is sent to a Catalyst 6000 switch that has a Multi-layer Switch Feature Card 2 (MSFC2). The MSFC2 reloads with a bus error in the Not-So-Stubby Area (NSSA) of the Open Shortest Path First (OSPF) code.
Workaround Avoid sending LSAs that have an illegal mask to the Catalyst 6000.
Miscellaneous
•
Symptoms Opening the port configuration (port > Configure > Interface) shows all fields as "N/A" or as having incorrect values. Also, port > Monitor > Interface charts do not get updated and show errors in the status bar.
Conditions These symptoms are observed in CiscoView on a MultiChannel DS3 port adapter in the WS-X6182-PA module.
Workaround There is no workaround.
•
Symptoms A bus error may cause a router to be returned to the ROM mode.
Conditions This symptom is observed on a Cisco 7204VXR router that is running Cisco IOS Release 12.1(7). The Cisco 7204VXR returns to the ROM mode if the snasw dump all privileged EXEC command (or the snasw dump summary-ipstrace privileged EXEC command) and the snasw stop privileged EXEC command are entered simultaneously on the router.
Workaround There is no workaround.
•
Symptoms The error message "%TRUNK-3-HBEAT: No longer receiving heartbeats from framer CPU" may be displayed.
Conditions This symptom is observed on a Cisco AS5800.
Workaround Use the firmware that is compiled using the version 96q1 compiler.
•
Symptoms A memory leak may occur on a router.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1(8). The symptom occurs at the "logger" process. There is also increased utilization in the "tty background" process. This is seen when the show process cpu EXEC command is issued.
Workaround Disable "logging synchronous" on the vty, aux, and console ports by entering the no logging synchronous line configuration command.
•
Symptoms A router may reload with a RedZone violation error.
Conditions This symptom is observed on a Cisco 2500 router that is running Cisco IOS Release 12.2(6.8)T when a large packet is received on the BRI interface. This symptom typically occurs when an ISDN call is terminated.
Workaround There is no workaround.
•
Symptoms A router configured with GRE tunnels may pause indefinitely and continuously scroll the following messages on the console:
%SYS-2-NOTQ: unqueue didnÌt find 0 in queue 62360144 -Process= "<interrupt level>", ipl= 1 -Traceback= 60538810 60536468 60536468 6015DB10 60431D64 60433D04 60433DC8 %SYS-2-BADSHARE: Bad refcount in retparticle, ptr=0, count=0 -Traceback= 60672220 60538818 60536468 60536468 6015DB10 60431D64 60433D04 60433 DC8Conditions The conditions under which these symptoms occur are not known at this time.
Workaround There is no workaround.
•
Symptoms Packets on a Versatile Interface Processor (VIP) are dropped.
Conditions This symptom is observed on a Cisco 7500 series router that is configured as a provider edge router in a Multiprotocol Label Switching (MPLS) virtual private network (VPN) switching environment when there are no distributed Cisco Express Forwarding (dCEF) entries for the remote virtual private network routing and forwarding (VRF) route.
This symptom occurs if a VRF is deleted after dCEF and the Border Gateway Protocol (BGP) are disabled from any previous MPLS VPN configuration.
Workaround Disable and reenable distributed dCEF by issuing the no ip cef distributed global configuration command followed by the ip cef distributed global configuration command on the Cisco 7500 series router. End-to-end connectivity is restored after this workaround is performed.
•
Symptoms A High-Speed Serial Interface (HSSI) logical DTE may not recover automatically from a HSSI cable fault. The transmission data light on the HSSI DTE may be unlit when this symptom occurs. The HSSI DTE may appear to be transmitting Local Management Interface (LMI) enquiries, but no LMI updates are received from the Frame Relay switch. Command output from the debug frame-relay lmi EXEC command may indicate that LMI inquiries are sent out from the router but the interface is not receiving any LMI updates from the Frame Relay switch.
Conditions This symptom is observed on a HSSI on a router if the HSSI cable between the CSU and the HSSI interface is unplugged and then plugged back in after the HSSI interface is declared to be in the down state.
Workaround Issue the clear interface hssi EXEC command on the logical HSSI DTE.
Alternate Workaround Issue the shutdown interface configuration command followed by the no shutdown interface configuration command on the logical HSSI DTE.
•
Symptoms A router may log the following error message:
%CALL_MGMT-1-CPM_Q_POOL: Cannot get memory for process watched queue entryCPU utilization is consumed by the call management process and may increase to 99 percent.
Conditions This symptom is observed on a Cisco 3620 router that is configured for analog modem calls.
Workaround There is no workaround.
•
Symptoms A universal access server may reload.
Conditions This symptom is observed on a Cisco AS5800 universal access server after trunk card configuration commands are issued.
Workaround There is no workaround.
•
Symptoms On a router the Versatile Interface Processor (VIP) cards may run out of processor memory very quickly.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1(12) and that has the ip inspect interface configuration command and distributed Cisco Express Forwarding (dCEF) configured. This symptom occurs because of a memory leak at the CEF interprocess communication (IPC) background process.
Workaround Disable the ip inspect interface configuration command.
•
Symptoms Active reverse Telnet sessions on a preconfigured port on a network module may stop transmitting data if the physical-layer async interface configuration command is configured on another port.
Conditions This symptom is observed on an 8-port asynchronous/synchronous network module (NM-8A/S) that has the physical-layer async interface configuration command configured on one port.
Workaround Do not configure the physical-layer async interface configuration command on a port if any of the ports on the network module has an active session. Clear the active lines before configuring the physical-layer async interface configuration command. Alternatively, the router can be reloaded after the command is configured.
•
Symptoms A Systems Network Architecture Switching (SNASw) router may reload in the routine ntl_avl_rotate_right. Traceback contains the following message:
0x6104B57C:nba_mmcpu_compare_proc_type(0x6104b578)+0x4 0x61143930:ntl_avl_find (0x611438f8)+0x38 0x6104B00C:nba_update_mm_stats(0x6104af6c)+0xa0 0x611472B8:nba_mm_free(0x61147294)+0x24 0x61144F98:nba_free_buffer(0x61144f00) +0x98 0x61147C2C:nba_send_ips(0x61147a78)+0x1b4 0x61140888:nbm_free_buffer (0x61140818)+0x70 0x61147C3C:nba_send_ips(0x61147a78)+0x1c4Conditions This symptom is observed when two downstream Low Entry Nodes (LEN) send Berkeley Internet Name Domains (BINDs) with the same Procedure Correlation Identifier (PCID) correlator at the same time. This situation should be a very rare occurrence, but some devices may use a random number when generating PCID correlators.
Workaround There is no workaround.
•
Symptoms Large packets cannot be passed on a router that has a High-Speed Serial Interface (HSSI) module.
Conditions This symptom is observed on an HSSI network module that is installed on a Cisco 3600 router that is running Cisco IOS Release 12.2(10) or Release 12.2(10.3)T. This symptom is observed only when large packets are sent.
Workaround Issue the mtu 1019 interface configuration command to set the maximum transmission unit (MTU) size of the HSSI interface to 1019 bytes. This workaround may not work in all cases.
•
Symptoms A Systems Network Architecture Switching Services (SNASw) router that is configured with a downstream port of conntype=len incorrectly advertises itself as nodetype=NN in the exchange identification (XID) exchange.
Conditions This symptom is observed on SNASW routers that are running Cisco IOS Release 12.0 T, 12.1 and 12.2.
Workaround There is no workaround.
•
Symptoms A Systems Network Architecture Switching Services (SNASw) dependent logical unit requester (DLUR) router cannot activate the pipe between the DLUR and the dependent logical unit server (DLUS). The following message may be displayed on the router:
**** 00001202 - EXCEPTION 512:492 (0) **** Locate search failed: search error Sense code = 0x08900060 Origin CP name = XXXXXXXX.XXXXXX Origin LU name = XXXXXXXX.XXXXXX Destination LU name = YYYYYYYY.YYYYYYConditions This symptom is observed on an SNASw router that is running Cisco IOS Release 12.2(6).
Workaround Restart the SNASw protocol by issuing the snasw stop privileged EXEC command followed by the snasw start privileged EXEC command on the router.
•
Symptoms A router may have to be reloaded because of memory allocation (MALLOC) failures.
Conditions This symptom is observed on a Cisco 7200 series router that has an 8-port fully channelized T1/E1 PRI, 256-channel port adapter.
The Cisco 7200 has to be reloaded because of a memory leak. This symptom occurs when heavy traffic is flowing through the interfaces and when the interfaces are forced to flap because a connected router reloads.
Workaround There is no workaround.
Novell IPX, XNS, and Apollo Domain
•
Symptoms Internetwork Packet Exchange (IPX) Service Advertisement Protocol (SAP) updates are not populated properly.
Conditions This symptom is observed when IPX Enhanced Interior Gateway Routing Protocol (EIGRP) is used.
Workaround Use Routing Information Protocol (RIP) SAP.
Wide-Area Networking
•
Symptoms A switch may not recover after an end-to-end ATM permanent virtual circuit (PVC) failure.
Conditions This symptom is observed on a Catalyst 2924XL switch that a Catalyst 2900 XL series ATM 155 single-mode (SM) medium-reach (MR) fiber module (WS-X2971-XL) when Operation, Administration, and Maintenance (OAM) management is used.
Workaround Avoid using OAM management. This precaution will ensure that the link remains up and that traffic will pass.
•
Symptoms UP and DOWN status messages may be displayed on the console.
Conditions This symptom is observed when a leased-line configuration is in the UP state, but the peer is not responding. This symptom occurs because PPP calls the interface reset vector regularly if the peer is not responding to the PPP attempts to communicate.
Workaround There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(15)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(15). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(15). This section describes severity 1 and 2 caveats and select severity 3 caveats.
The following information is provided for each caveat:
•
•
•
Basic System Services
•
Symptoms Some calls get rejected.
Conditions This symptom is observed on all voice platforms on which the call fallback mechanism is enabled.
Workaround There is no workaround.
•
Symptoms A router may fail to boot and display the following error message:
4B4 604DC4A0 %SYS-2-INTSCHED: 'idle' at level 4 -Process= "EnvMon", ipl= 4, pid= 8 -Traceback= 6050B024 604F5F98 604F8154 6092BBA8 607124FC 60542FD8 60543228 604DCConditions This symptom is observed on a Cisco router that is running Cisco IOS Release 12.1(11) and Simple Network Management Protocol (SNMP). The router has to be power-cycled to complete the boot process.
Workaround Remove SNMP traps from the configuration.
EXEC and Configuration Parser
•
Symptoms A router may reload unexpectedly.
Conditions This symptom is observed if a configuration change is made by a user whose username has a percent sign (%) sign in it on a Cisco router that is running Cisco IOS Release 12.1.(12) and that has Network Time Protocol (NTP) configured.
Workaround There is no work around.
IBM Connectivity
•
Symptoms The following error messages appears on the router log:
May 22 15:14:04: %SYS-2-LINKED: Bad enqueue of 61EC814C in queue 6202479C -Process== "<interrupt level>", ipl== 4 -Traceback== 6041F50C 6041CC08 6123BAEC 60CA22D4 6023D1B4 6023C964 60232324 May 22 15:19:24: %BSC-3-BADLINESTATE: Line state Tx when receiving EOT on line Serial1/0When the router logs the messages, the router may or may not respond to the attached bisync/bsc devices.
Conditions This symptom is observed on different platforms and serial interface modules.
Workaround Try the Block Serial Tunneling (BSTUN) interface as half duplex or try a different serial module other than WIC-2T, such as NM-4A/S.
Interfaces and Bridging
•
Symptoms A router may reload because of an SegV exception.
Conditions This symptom is observed when FRF.9 hardware compression is used on a platform that supports hardware compression.
Workaround Disable hardware compression.
•
Symptoms A memory allocation failure (MALLOCFAIL) message is displayed when a cable is unplugged from a serial interface.
Conditions This symptom is observed on a Cisco 7206VXR router that is running Cisco IOS
Release 12.2(7a) and that has a Network Processing Engine (NPE-400) when a cable is unplugged from a serial interface and when the l3 bypass global configuration command is enabled.Workaround There is no workaround.
•
Symptoms An IP ping may not go through.
Conditions This symptom is observed in an integrated routing and bridging (IRB) environment on a bridging and bridging-to-routing path. The IP ping does go through on a routing and routing-to-bridging path.
Workaround There is no workaround.
•
Symptoms A router may reload if a circuit that has compression configured is removed while there is subsequent activity on a compression retry timer.
Conditions This symptom is observed on a Cisco router that is using software or hardware compression and that has FRF.9 Frame Relay compression configured. The activity on the compression retry timer occurs because of a transmission error and subsequent signaling of a compression restart sequence.
Workaround There is no workaround.
IP Routing Protocols
•
Symptoms The ip host-routing global configuration command is enabled and if routing is disabled, the networking device will stop all forms of routing and remove static routes that are defined in a configuration.
.Conditions This symptom is observed when the no ip host-routing global configuration command is issued in configuration terminal mode on a networking device.
Workaround To issue the no ip host-routing global configuration command, first enable the command using the ip host-routing global configuration command and then disable the command immediately using the no ip host-routing global configuration command. Do not exit the global configuration mode before issuing the no ip host-routing global configuration command.
•
Symptoms An external Autonomous System Boundary Router (ASBR) may choose a suboptimal path for an external type 2 route.
Conditions This symptom is observed on an ASBR router that is situated external to a network that has several Area Border Routers (ABRs).
Workaround There is no workaround.
•
Symptoms A router may reload.
Conditions This symptom is observed on a Cisco router when a large Network Address Translation (NAT) address pool is unconfigured.
Workaround There is no workaround.
•
Symptoms A router may reload when it is searching the Resource Reservation Protocol (RSVP) database.
Conditions This symptom is observed on a Cisco router that is running in the Route Processor Redundancy Plus (RPR+) or the Stateful SwitchOver (SSO) mode. The router reloads when a tunnel is up and when Multiprotocol Label Switching-traffic engineering (MPLS-TE),
Cisco Express Forwarding (CEF), and IP routing are unconfigured using the following sequence of commands:
no tag advertise-tags
no mpls ip
no mpls label protocol ldp
no ip routing
no ip cef
no mpls traffic-eng tunnelsWorkaround Issue the no mpls traffic-eng tunnels router configuration command to shut down all tunnels before issuing the no ip routing global configuration command.
•
Symptoms A router may reload with a bus error while the shortest path first (SPF) algorithm is computed.
Conditions This symptom is observed if multiple routers are advertising the same prefix in Type-5 or Type-7 link-state advertisements (LSAs).
Workaround There is no workaround.
•
Symptoms A software-forced reload may occur on a router.
Conditions This symptom is observed on a Cisco 12008 router that has a Multiprotocol Label Switching (MPLS) traffic engineering (TE) tunnel that is configured with an absolute metric when the tunnel is used with the Open Shortest Path First (OSPF) protocol. A watchdog timer event may be triggered, and the router may reload after the shutdown interface configuration command followed by the no shutdown interface configuration command is issued on the tunnel interface.
Workaround There is no workaround.
•
Symptoms A software-forced reload may occur on a router.
Conditions This symptom is observed on a Cisco router if an interface is shut down from another terminal while output from the show ip pim neighbor EXEC command is displayed.
Workaround There is no workaround.
•
Symptoms A router may reload.
Conditions This symptom is observed during withdrawal of two paths to the same destination at the same time.
Workaround There is no workaround.
•
Symptoms A router flushes link-state advertisements (LSAs) that have not been refreshed for more than 50 minutes.
Conditions This symptom is observed on a Cisco router that is running Cisco IOS
Release 12.0(19)ST1.Workaround There is no workaround.
Miscellaneous
•
Symptoms The following continuous stream of "%POT1E1-3-FWFATAL" error messages may occur on a router:
%POT1E1-3-FWFATAL: Bay 5: firmware needsresetdue to fw watchdog timeout % POT1E1-3-FWFATAL: Bay 4: firmware needsresetdue to fatal softwareerrorsConditions This symptom is observed on a Cisco 7206VXR router that is running Cisco IOS
Release 12.1(8.04) and using an eight-port multichannel T1 port adapter (PA-MC-8T1).Workaround There is no workaround.
•
Symptoms High memory utilization that results in malloc failures may occur on a router.
Conditions This symptom is observed if you use the ip route network-number network-mask ethernet [bay] slot/port EXEC command to configure static routes to a prefix on a Cisco router that is capable of distributed Multiprotocol Label Switching (MPLS) forwarding (such as a
Cisco 7500 series router or a Cisco 12000 series Internet router). If you enter the show cef linecard internal command, many interprocess communication (IPC) messages appear to be queued up under the category "rtc."Workaround When a route is configured using the ip route network-number network-mask ethernet [bay] slot/port EXEC command, make sure that you also provide a next hop IP address. Use the following command to configure a next hop IP address: ip route network-number network-mask ip-address ethernet [bay] slot/port (where IP address is the next hop IP.)
•
Symptoms The clear to send (CTS) signal on port 0 is inverted for a short period of time when binary synchronous communication (BISYNC) polling is started after the router is reloaded. This behavior causes an ATM machine to enter the error recovery mode and may prevent the ATM machine from recovering and reestablishing proper communication with the router.
Conditions These symptoms are observed on a Cisco 2600 router that has a 2-port serial low-speed asynchronous and synchronous WAN interface card (WIC 2 A/S) that is configured for BISYNC on port 0 and has the physical-layer async interface configuration command configured on port 1.
Workaround After the ports have been initialized through a reload or after the physical-layer async interface configuration command is removed from the configuration of a port, issue the shutdown interface configuration command followed by the no shutdown interface configuration command on the serial interface to reinitialize the hardware settings.
•
Symptoms On a router, the amount of available dynamic memory may decrease.
Conditions This symptom is observed as the Internet Key Exchange (IKE) negotiates IP Security (IPSec) tunnels if Rivest, Shamir, and Adleman (RSA) signature authentication is used. The loss of available memory may continue until the router is no longer able to operate because of a lack of dynamic memory.
Workaround Reload the router.
•
Symptoms Operation, Administration, and Maintenance (OAM) cells may be delayed in transmission, and OAM cells may be sent out of sequence.
Conditions This symptom is observed on an overloaded permanent virtual circuit (PVC). The PVC may go down when this symptom occurs.
Workaround There is no workaround.
•
Symptoms Systems Network Architecture Switching Services (SNASw) may not release memory as expected.
Conditions This symptom is observed on a Cisco 2600 router that is running SNASw and that is running Cisco IOS Release 12.2(6). The router may consume memory in a two-network node servers scenario in which one of the servers has failed and recovered. The SNASw directory services process is the process that consumes memory.
Workaround There is no workaround.
•
Symptoms IP connectivity may be disrupted after distributed Cisco Express Forwarding (dCEF) is configured on a router.
Conditions This symptom is observed on a Cisco 7500 series router that is functioning as a provider edge (PE) router and that is running tag switching or Multiprotocol Label Switching (MPLS). This symptom occurs if the router is running both cell-based and frame-based tag switching simultaneously.
Workaround There is no workaround.
•
Symptoms The input queue fills and the Systems Network Architecture Switching Services (SNASw) upstream link fails.
Conditions This symptom occurs if SNASw receives a protocol violation generating sense 8007 on an incoming bind frame from the virtual telecommunications access method (VTAM). This situation causes subsequent frames to be kept on the input hold queue. This behavior causes failures with sense 0805 on the VTAM, causing sessions to be stuck in the "pending session start" (PSEST) state. When this condition occurs, the input queue eventually fills and the SNASw upstream link fails.
Workaround Identify and terminate the affected Real-Time Transport Protocol (RTP) pipe from the VTAM.
•
Symptoms A Route Processor (RP) may experience a bus error or pause indefinitely when a crash test is performed.
Conditions This symptom occurs after the test crash command is issued on an active RP.
Workaround Reload or power-cycle the router.
•
Symptoms A router may reload.
Conditions This symptom is observed during the execution of the no e1 1 channel-group 0 command on the controller of a Multi-Channel E3 port adapter on a Cisco 7200 series router that is configured for IP routing.
Workaround Shut down the interface and then remove the channel group.
•
Symptoms In Y-Cable APS, with the manual-switch condition on the Protection channel, the Signal Degrade (SD) condition is applied and then removed from the Working channel. After the SD condition is removed, the Working channel may become the active channel again.
Conditions This symptom is observed when the SD condition on Working channel did not clear the lower-priority manual-switch condition on the Protection channel, which causes the switchback to Working channel when the SD condition is cleared.
Workaround There is no workaround.
•
Symptoms Active reverse Telnet sessions on a preconfigured port on a network module may stop transmitting data if the physical-layer async interface configuration command is configured on another port.
Conditions This symptom is observed on an 8-port asynchronous/synchronous network module (NM-8A/S) that has the physical-layer async interface configuration command configured on one port.
Workaround Do not configure the physical-layer async interface configuration command on a port if any of the ports on the network module has an active session. Clear the active lines before configuring the physical-layer async interface configuration command. Alternatively, the router can be reloaded after the command is configured.
•
Symptoms An Enterprise Extender (EE) link to a host may toggle between the up and the down states.
Conditions This symptom occurs when the physical unit (PU) link is not predefined on the host or when the PU is predefined as "DISNCT=xxx" rather than "DISNCT=NO." When this symptom occurs, the incorrect value for the ALIVE timer is passed between the two Real-Time Transport Protocol (RTP) endpoints in a connection setup.
Workaround Predefine the PU link on the host as "DISCNT=NO."
•
Symptoms A Systems Network Architecture Switching (SNASw) router may reload in the routine ntl_avl_rotate_right. Traceback contains the following message:
0x6104B57C:nba_mmcpu_compare_proc_type(0x6104b578)+0x4 0x61143930:ntl_avl_find(0x611438f8)+0x38 0x6104B00C:nba_update_mm_stats(0x6104af6c)+0xa0 0x611472B8:nba_mm_free(0x61147294)+0x24 0x61144F98:nba_free_buffer(0x61144f00)+0x98 0x61147C2C:nba_send_ips(0x61147a78)+0x1b4 0x61140888:nbm_free_buffer(0x61140818)+0x70 0x61147C3C:nba_send_ips(0x61147a78)+0x1c4Conditions This symptom is observed when two downstream Low Entry Nodes (LEN) send Berkeley Internet Name Domains (BINDs) with the same Procedure Correlation Identifier (PCID) correlator at the same time. This situation should be a very rare occurrence, but some devices may use a random number when generating PCID correlators.
Workaround There is no workaround.
•
Symptoms A router may reload.
Conditions This symptom is observed under rare circumstances when online insertion and removal (OIR) of an S card is executed.
Workaround Do not execute S card OIR. Wait for a maintenance window during which the router can be powered down and up.
•
Symptoms A router may reload.
Conditions This symptom is observed on a Cisco router with the tunnel configured and when both the tunnel source is removed and the tunnel has no "raw-packet-interface".
Workaround There is no workaround.
•
Symptoms A Systems Network Architecture Switching Services (SNASw) dependent logical unit requester (DLUR) router cannot activate the pipe between the DLUR and the dependent logical unit server (DLUS). The following message may be displayed on the router:
**** 00001202 - EXCEPTION 512:492 (0) ****
Locate search failed: search error
Sense code = 0x08900060
Origin CP name = XXXXXXXX.XXXXXX
Origin LU name = XXXXXXXX.XXXXXX
Destination LU name = YYYYYYYY.YYYYYYConditions This symptom is observed on an SNASw router that is running Cisco IOS
Release 12.2(6).Workaround Restart the SNASw protocol by issuing the snasw stop privileged EXEC command followed by the snasw start privileged EXEC command on the router.
•
Symptoms A router may experience a line protocol flap or go down in high traffic conditions.
Conditions This symptom is observed on a Cisco 7200 router that has a port adapter (PA-T3, PA-E3, or PA-H), a Network Processing Engine (NPE-400), or a Network Service Engine-1 (NSE-1).
Workaround Use dual interface versions of the port adapters mentioned above.
•
Symptoms There may be difficulty enabling Cisco Express Forwarding (CEF) using CEF commands, and packets may be process-switched.
Conditions This symptom is observed on a Cisco 7100 router.
Workaround There is no workaround.
TCP/IP Host-Mode Services
•
Symptoms On a router, the Karn's Round-Trip Time (KRTT) may not be bounded to RTTO*2**5.
Conditions This symptom is observed on a Cisco router that is running a Cisco IOS release that contains the workaround for DDTS CSCdu18397. When there are retransmissions that occur between the TCP endpoints, the KRTT value can get excessively large and the TCP connection drops. This situation frequently affects Data-Link Switching (DLSw). The DLSw peers sporadically drop.
Workaround There is no workaround.
Wide-Area Networking
•
Symptoms The Layer 2 Tunnel Protocol (L2TP) may get stuck.
Conditions This symptom is observed when the L2TP parses an invalid control message with a zero-length attribute-value (AV) pair.
Workaround There is no workaround.
•
Symptoms A Frame Relay over ISDN call may fail.
Conditions This symptom is observed if the data-link connection identifier (DLCI) on the subinterface is in an inactive state.
Workaround Configure the permanent virtual circuit (PVC) at the main interface rather than at the subinterface.
•
Symptoms A Cisco router may set up a tunnel.
Conditions This symptom is observed even if bidirectional Challenge Handshake Authentication Protocol (CHAP) authentication is not successful.
Workaround There is no workaround.
•
Symptoms A router may experience a memory leak.
Conditions This symptom is observed on a Cisco router that is configured for X.25 over the Link Access Procedure, Balanced (LAPB) links in a high error environment.
Workaround Correct the cause of the high error rate on the LAPB links.
•
Symptoms A Versatile Interface Processor (VIP), Gigabit Ethernet Interface Processor (GEIP), Gigabit Ethernet Interface Processor plus (GEIP+), or Packet OC-3 Interface Processor (POSIP) that is installed in a router may reload. The VIP may display the following error message when it reloads:
%DMA-1-DRQ_STALLED: DRQ stalled. Dumping DRQ.Conditions This symptom is observed on a Cisco 7500 router under heavy traffic conditions.
Workaround There is no workaround.
•
Symptoms A router may reload with a bus error displaying the following error message:
%ALIGN-1-FATAL: Illegal access to a low address addr=0x50, pc=0x6135D584, ra=0x60636AD0, sp=0x62338E68Conditions This symptom is observed on a Cisco 7200 router.
Workaround There is no workaround.
•
Symptoms On a router, IP connectivity may get lost and then be reestablished. The packets coming from the router will have TCP checksum failures, and the packets will be dropped.
Conditions This symptom is observed when a serial tunnel (STUN) connected from a Cisco 7200 series router that is running Cisco IOS Release 12.2(7.5) to another router over a Voice over Frame Relay (VoFR) link.
Workaround Reset the STUN connection.
•
Symptoms The values of the MIB objects ifHCInOctets and ifHCOutOctets may show an abnormally high rate of increase. These values may be much higher than the maximum possible rate or than the ifHighSpeed of that interface.
Conditions This symptom is observed on interfaces that represent a LAN emulation (LANE) client. This bandwidth utilization, as calculated from these objects, could be abnormally high and incorrect.
Workaround There is no workaround.
•
Symptoms A router may reload.
Conditions This symptom is observed on a Cisco router when dialer profiles are used to get links up to the destination with a Multilink PPP (MLP) configuration and if there is an idle timeout or the dialer profile interface is disconnected.
Workaround Use a rotary configuration instead of a dialer profile configuration.
•
Symptoms A packet is incorrectly dropped by the router.
Conditions This symptom is observed when the Frame Relay encapsulated packet is a Multiprotocol Label Switching (MPLS) packet on locally switched Frame Relay permanent virtual circuits (PVCs) that are configured with the connect or Frame Relay route interface configuration commands,. This condition affects only the Cisco 7500 series routers and only the Cisco IOS images that support MPLS switching, such as the rsp-pv-mz or rsp-jsv-mz image.
Workaround There is no workaround.
•
Symptoms A Frame Relay over ISDN call is not retried after it is dropped.
Conditions This symptom occurs when Frame Relay over ISDN is used. If a data-link connection identifier (DLCI) that is configured on a subinterface goes down (becomes inactive or is deleted), the corresponding subinterface is marked down. When the ISDN circuit is released, the subinterface remains in the down state. Subsequent attempts to bring up the ISDN circuit using the network address that is bound to that subinterface do not work.
Workaround Enter the shutdown interface configuration command followed by the no shutdown interface configuration command on the subinterface.
Resolved Caveats—Cisco IOS Release 12.1(14)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(14). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(14). This section describes severity 1 and 2 caveats and select severity 3 caveats.
Basic System Services
•
A Cisco router may reload when a probe is configured to operate at a frequency of 0 seconds and then is scheduled to run.
Workaround: The probe frequency of 0 seconds is illegal and must not be used.
•
A Cisco router that is running Cisco IOS Release 12.0(20.3)S1 may not be accessible through the Ethernet 0 interface.
Workaround: Load the software onto the router, copy the running configuration file to the startup configuration, and reload the router.
•
A Cisco router that is running Cisco IOS Release 12.1(12) may display the following traceback messages and reload after the clear cdp table privileged EXEC command is issued:
%ALIGN-3-TRACE: -Traceback= 604E42A0 604E39EC 604E37B0 604E32B0 6026BDE4 60277FCC 602C90F4 602C90E0
%ALIGN-3-TRACE: -Traceback= 604E42CC 604E39EC 604E37B0 604E32B0 6026BDE4 60277FCC 602C90F4 602C90E0
%ALIGN-3-TRACE: -Traceback= 604E42D0 604E39EC 604E37B0 604E32B0 6026BDE4 60277FCC 602C90F4 602C90E0There is no workaround.
IBM Connectivity
•
A Cisco 7200 router that is configured with data-link switching plus (DLSw+) Routing Information Field (RIF) passthrough peers may reload if:
–
–
–
Workaround: Perform the following steps:
a.
source-bridge max-hops3
source-bridge max-in-hops 3These commands limit the maximum number of hops from each end of the physical Token Ring interface to three hops. An additional fourth hop is used for the virtual ring. This configuration keeps the combined RIF length to within seven hops.
b.
dlsw timer sna-verify-interval 1200
dlsw timer netbios-verify-interval 1200These commands set the verify interval to 20 minutes (the default cache timeout is 16 minutes, or 960 seconds). This configuration prevents the router from entering the VERIFY state. The cache entry is deleted before the router can perform a VERIFY operation.
Interfaces and Bridging
•
ATM virtual circuit (VC) counters are not correctly incremented on a 1-port ATM OC3 multimode port adapter (PA-A1-OC3MM) that is installed on a Cisco 7500 router that is running Cisco IOS Release 12.1(7a)E2. There is no workaround.
•
A Versatile Interface Processor (VIP) may reload or record spurious access after class maps are configured for Frame Relay. There is no workaround.
•
On a Cisco 7200 router that is running Cisco IOS Release 12.1(12.4) and that has an ATM Lite port adapter, packet pings that are 500 bytes or larger in size may not go through the ATM Lite port adapter. Packet pings that are smaller than 500 bytes in size can be sent through the ATM Lite port adapter. There is no workaround.
IP Routing Protocols
•
A Cisco 7206VXR router that is running the Open Shortest Path First (OSPF) Protocol and acting as a designated router (DR) will generate router link states but may fail to generate network link states for a connected network. The OSPF neighbors will come up correctly on all routers in the network.
Workaround: Set the priority on the interface to 0 so that the router is not the DR for that link.
Miscellaneous
•
The byte counters on a serial interface on a Cisco 7500 series router that is running Cisco IOS Release 12.1(8) may not work properly. There is a difference between the number of bytes on one side of a serial link and the number of bytes on the other side of the link. The counter may also decrease. There is no workaround.
•
A Cisco AS5800 universal access server may experience a memory leak in the pool manager process. There is no workaround.
•
On a Cisco Route Switch Processor (RSP) that is running Cisco IOS Release 12.2 or 12.2 T with an Ethernet to Fast Ethernet channel combination topology, a fast cache entry is built when inbound NetFlow is configured on top of Cisco Express Forwarding (CEF) on an Ethernet interface. There is no workaround.
•
When an X.75 call is made on a E1 or PRI interface, a "SYS-3-INVMEMINT: Invalid memory action (malloc) at interrupt level" message may be displayed. This condition has no user impact. There is no workaround.
•
Significant throughput degradation may occur on a PPP multilink if the fragment delay on a 2-port multichannel E1 port adapter (PA-MC-2E1) that is configured for channel groups is set to a value that is lower than the default value. There is no workaround.
•
When a channelized port adapter (CT3) is used and if framing is reconfigured with the t1 1 framing esf privileged EXEC command, certain interfaces that are configured on that T1 line may stop passing traffic. This condition occurs only if the t1 1 framing esf privileged EXEC command is issued after channel groups are already configured on the T1 line and while the channel groups are passing traffic. The framing needs to be set only for the T1 line when the first channel group is configured and does not need to be reentered when a new channel group is added.
Workaround: Issue the t1 1 framing esf privileged EXEC command only when the first interface on a T1 line is configured.
•
When a channelized T3 port adapter (CT3) that is configured with multiple channel groups is used, all interfaces may stop passing traffic if T1 frames are received on one of the groups that has voice signaling enabled.
Workaround: To clear this condition, shut down the interface that corresponds to the channel group that is receiving the invalid frame. If any of the other interfaces continues to flap after the interface that is receiving the invalid frame is shut down, the interface has to be reconfigured.
•
A Cisco 7204VXR router that is running the c7200-a3js-mz image of Cisco IOS Release 12.2(5a) may experience a software-forced reload. There is no workaround.
•
A Route Switch Module (RSM) that is running Cisco IOS Release 12.0(20) may experience alignment error corrections at the rsp_ipfib_feature_switch process.
Workaround: To clear this condition, disable IP Cisco Express Forwarding (CEF) and use fast switching instead.
•
If a Cisco router receives a gratuitous Address Resolution Protocol (ARP) for a static ARP entry with an alias keyword, the router may reload because of a bus error.
Workaround: Do not use an alias keyword on a static ARP entry. An alias keyword allows the router to answer ARP requests for the static ARP entry, but in many cases an alias keyword is not required because the router needs the static ARP entry only for its own use.
•
A Cisco Router Route Processor (GRP, RSP or NPE) may reload when a traffic engineering (TE) tunnel interface is disabled and reenabled immediately using the shutdown interface configuration command followed by the no shutdown interface configuration command. The router may also exhibit this behavior either when tag switching is enabled and disabled using the no tag-switching ip interface configuration command followed in quick succession by the tag-switching ip interface configuration command or when a loopback interface is disabled and reenabled using the shutdown interface configuration command followed immediately by the no shutdown interface configuration command. This behavior may also occur when a file is copied to the running configuration to change the state of a tunnel.
Workaround: Wait for at least a minute after the shutdown interface configuration command is issued before entering the no shutdown interface configuration command on a tunnel interface or its associated loopback interface. Wait for at least a minute after the no tag-switching ip interface configuration command is issued before entering the tag-switching ip interface configuration command. Shut down all tunnel interfaces before copying a file to the running configuration.
•
When an online insertion and removal (OIR) is performed on a Versatile Interface Processor (VIP) that is installed on a Cisco 7500 router, or if the single line card reload (SLCR) feature is used, traffic forwarding issues may occur after the VIP reloads. If the router uses either a static or a connected route to send traffic out of an interface that is on the failed VIP, forwarding may stop after an OIR or an SLCR procedure is performed.
Workaround: Issue the clear cef linecard adjacency EXEC command to recover from this condition.
•
A Cisco 7500 series router that is configured with the main interface in backup and subinterfaces with the same IP address has different behavior at boot-up between Cisco IOS Release 11.3(11)WA4(14) and Cisco IOS Release 12.0. During operation of the router, the duplicate IP addresses can be configured in both versions of Cisco IOS software. However, at boot-up the duplicate IP addresses are removed from the configuration in Cisco IOS Release 12.0.
Workaround: Manually configure the affected interfaces again after the reboot.
•
A Versatile Interface Processor (VIP) may reload if distributed Multilink PPP (dMLP) is configured on a channelized T3 or E3 interface after the router is reloaded and booting up. There is no workaround.
•
A Cisco router may experience a bus error and reload when the clear ip bgp * command is entered at the console and a large number of routes have been imported. There is no workaround.
•
The performance of the encryption path degrades when Cisco IOS Release 12.2(6.8) is used with a hardware encryption card. The loss in performance occurs because encrypted packets are process-switched instead of being fast-switched. This condition occurs when IP Security (IPSec) is applied to the interfaces while the hardware encryption card is used. There is no workaround.
•
On a Cisco AS5800 universal access server that is running Cisco IOS Release 12.1(11) and that has Resource Pool Management (RPM) or Resource Pool Manager Server (RPMS) enabled, the access server will show that modems are in use even when there are no active users connected. There is no workaround.
•
Systems Network Architecture Switching Services (SNASw) intermediate session routing (ISR) sessions that have an enterprise extender (EE) upstream may have messages that are unnecessarily segmented. Messages may be segmented because the CAPACITY on the physical unit (PU) definition on the host is too low or because the wrong primary send or receive basic transmission unit (BTU) size is being used by the SNASw router.
After the segmenting of messages starts to occur (for either reason), the End Of Message (EOM) segment may be corrupted, causing the other end of the Rapid Transport Protocol (RTP) connection to identify a gap and request that the message to be rebroadcasted. For a customer, this behavior caused significant delays as it is taking nearly two minutes before the gap detected status was returned to the SNASw router.
Workaround: To prevent the segmenting from occurring, add CAPACITY=4M to the PU definition on the host.
•
A Cisco 7206VXR router that has a Network Processing Engine (NPE-300) and that is running Cisco IOS Release 12.1(8.4) may reload because of a memory corruption if a service policy output with a policy map is configured using the policy-map policy-map-name global configuration command and if a 24-character long class-map-name is configured using the class-map class-map-name global configuration command.
Workaround: Use a map name that is less than or equal to 23 characters or a map name in which the character length is not a multiple of four.
•
A Cisco router that is running Cisco IOS Release 12.1(12.6) may reload when the debug crypto ipsec command is enabled. An invalid IP Security (IPSec) message may be displayed when this condition occurs. The following error message may be displayed before the router reloads:
CRYPTO-4-RECVD_PKT_INV_SPI: decaps: rec'd IPSEC packet has invalid spi for destaddr=201.201.30.1, prot=51, spi=0x1C0D0620(470615584) SYS-3-INVMEMINT: Invalid memory action (malloc) at interrupt levelWorkaround: Do not enable the debug crypto ipsec command.
•
After a Cisco AS5800 universal access server has been running for more than 10 hours under stress conditions, modem ISDN Channel Aggregation (MICA) technologies modems may be marked as bad. There is no workaround.
•
Spurious memory accesses are observed at the dsx3_controller_t1_framing process when extended super frame (ESF) framing is configured on a Channelized T3 Interface Processor (CT3IP) controller. Spurious memory accesses may also occur if the router is rebooted while framing is set to ESF. There is no workaround.
•
An invalid cache adjacency exists on a line card but not on the Route Processor (RP).
Workaround: Issue the clear cef linecard slot-number adjacency EXEC command on the line card.
•
A line card may be stuck in an off-for-download state on a Cisco 7500 router or a Cisco 12000 router. This condition is indicated by the output of the show cef linecard EXEC command. This condition is caused by an interprocess communication (IPC) error with another line card during the Forwarding Information Base (FIB) table download process. There is no workaround.
Wide-Area Networking
•
A Cisco router that has a Route Switch Processor (RSP) may pause indefinitely at the rsp_fs_free_memd_pak process and display the following message on the console:
%RSP-3-BADBUFHDR: freeing MEMD pak, address 0 -Traceback= 60367710 60373AA4 602D30B8 600DA700 602413E0 602416D8 60FCE4D8 60FCC CD0 602223C8 602224F4 6036A260 60378E70 60331290There is no workaround.
•
An ISDN trap is not generated after a call is connected. The following ISDN objects are affected:
–
–
–
–
–
–
–
There is no workaround.
•
When an IP digital subscriber line (DSL) switch is used as a Layer 2 Tunneling Protocol (L2TP) network server (LNS), the IP DSL switch may pause indefinitely or reboot if the ping sweep (from 64 to 1500) is generated from the customer premises equipment (CPE). There is no workaround.
•
A Cisco 3600 router that is running Cisco IOS Release 12.1(5)T5 may experience a memory leak at the ISDN process. There is no workaround.
•
A versatile interface processor (VIP) may reload because of memory corruption when Inter-Switch Link (ISL) encapsulation is disabled and reenabled on an interconnected Gigabit Ethernet interface on a provider edge (PE) router. This problem occurs as traffic is flowing between two customer edge (CE) routers that are connected through a virtual private network (VPN). There is no workaround.
•
The virtual-access counters and the RADIUS accounting data exceed the real value. This condition was observed on a Cisco 7200 PA-A3 port adapter and a Cisco 6400 NRP2-SV when a Layer 2 Tunneling Protocol (L2TP) Network Server (LNS) used an ATM permanent virtual connection (PVC) as an ingress interface for L2TP tunnels.
Workaround: Use xEthernet as the ingress interface.
•
A Cisco 7500 router that has the Route Switch Processor (RSP 4) may reload when it is running an experimental image that is based on Cisco IOS Release 12.2(6.6)T. The router reloads with a bus error because of the following corrupted program counter (PC) in the dialer code:
%ALIGN-1-FATAL: Corrupted program counter pc=0x10009, ra=0x10009, sp=0x62DB13D8There is no workaround.
•
A Cisco router may display a traceback message at the propagate_hunt_rprofile_changes() function when a configuration is changed on the dialer interface. There is no workaround.
•
Downstream Systems Network Architecture switching services (SNASw) users may experience session disconnects if the users are connecting to the SNA switch port that is an ATM LAN Emulation (LANE) subinterface. This behavior occurs when a router is running Portable SNA (PSNA) Advanced Peer-to-Peer Networking (APPN) software. This behavior may also occur when a router is running just PSNA. There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(13a)
Cisco IOS Release 12.1(13a) is a rebuild of Cisco IOS Release 12.1(13). Caveats listed in this section are resolved in Cisco IOS Release 12.1(13a) but may be open in previous Cisco IOS releases.
•
Cisco routers and switches running Cisco IOS software and configured to process Internet Protocol version 4 (IPv4) packets are vulnerable to a Denial of Service (DoS) attack. A rare sequence of crafted IPv4 packets sent directly to the device may cause the input interface to stop processing traffic once the input queue is full. No authentication is required to process the inbound packet. Processing of IPv4 packets is enabled by default. Devices running only IP version 6 (IPv6) are not affected. A workaround is available.
Cisco has made software available, free of charge, to correct the problem.
This advisory is available at
http://www.cisco.com/warp/public/707/cisco-sa-20030717-blocked.shtml
Resolved Caveats—Cisco IOS Release 12.1(13)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(13). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(13). This section describes severity 1 and 2 caveats and select severity 3 caveats.
Interfaces and Bridging
•
On a Cisco 7200 router that is running Cisco IOS Release 12.1(12.4) and that has an ATM Lite port adapter, packet pings that are 500 bytes or larger in size may not go through the ATM Lite port adapter. Packet pings that are smaller than 500 bytes in size can be sent through the ATM Lite port adapter. There is no workaround.
Miscellaneous
•
The rtr reaction-configuration command can be used to trigger another probe when a timeout, a threshold violation, or a verify error occurs. However, the probe to be triggered gets activated only once during the first occurrence of the reaction condition. If the reaction condition occurs again (for example, when the timeout condition clears) and the triggered probe is in the pending state the probe should get activated again.
Workaround: This condition is resolved in Cisco IOS Release 12.0(21)S.
•
An error can occur with management protocol processing. Please use the following URL for further information:
http://www.cisco.com/cgi-bin/bugtool/onebug.pl?bugid=CSCdw65903
Resolved Caveats—Cisco IOS Release 12.1(12c)
Cisco IOS Release 12.1(12c) is a rebuild release for Cisco IOS Release 12.1(12). The caveats in this section are resolved in Cisco IOS Release 12.1(12c) but may be open in previous Cisco IOS releases.
•
When an online insertion and removal (OIR) is performed on a Versatile Interface Processor (VIP) that is installed on a Cisco 7500 router, or if the single line card reload (SLCR) feature is used after the VIP reloads, traffic that is using static routes that are defined to point out interfaces on the failed VIP may fail. The static routes that are affected are routes that are defined within a Multiprotocol Label Switching (MPLS) Virtual Private Network routing and forwarding (VRF) instance.
Workaround: Issue the clear cef linecard adjacency EXEC command to recover from this condition.
Resolved Caveats—Cisco IOS Release 12.1(12b)
Cisco IOS Release 12.1(12b) is a rebuild of Cisco IOS Release 12.1(12). The caveats in this section are resolved in Cisco IOS Release 12.1(12b) but may be open in previous Cisco IOS Releases.
•
An error can occur with management protocol processing. Please use the following URL for further information:
http://www.cisco.com/cgi-bin/bugtool/onebug.pl?bugid=CSCdw65903
Resolved Caveats—Cisco IOS Release 12.1(12a)
Cisco IOS Release 12.1(12a) is a rebuild release for Cisco IOS Release 12.1(12). The caveats in this section are resolved in Cisco IOS Release 12.1(12a) but may be open in previous Cisco IOS releases.
•
A Cisco 7200VXR router that is using any unchannelized serial port adapter (PA) and any processor other than the Network Processing Engine (NPE-300) may experience line flaps at high traffic rates and display the following message:
MUESLIX-1-HALT: Mx serial: Serial6/0 TPU halted: cause 0x3 status 0x00371A00Carrier transitions and wedged output queues may also occur. This condition affects the following port adapters:
–
–
–
–
Multichannel port adapters such as the PA-MC-T3 or the PA-MC-2T3+ are not affected. This condition affects only the Cisco 7200VXR router. There is no workaround.
Resolved Caveats—Cisco IOS Release 12.1(12)
This section describes possibly unexpected behavior by Cisco IOS Release 12.1(12). All the caveats listed in this section are resolved in Cisco IOS Release 12.1(12). This section describes severity 1 and 2 caveats and select severity 3 caveats.
Access Server
•
A Cisco AS5800 series universal access server that is running Cisco IOS Release 12.1(7.2) may experience the following error and traceback messages:
%TTY-3-AUTOCONFIG: TTY1/9/29: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/33: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/34: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/57: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/62: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/56: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/75: Modem auto-configuration failed
%TTY-3-AUTOCONFIG: TTY1/9/64: Modem auto-configuration failed Queued messages:
%SYS-3-LOGGER_FLUSHING: System pausing to ensure console debugging output. %SYS-3-LOGGER_FLUSHED: System was paused for 00:00:02 to ensure console debugg ing output.
%SYS-2-MALLOCFAIL: Memory allocation of 1048 bytes failed from 0x6053A7F0, poo l Processor, alignment 0 -Process= "Crash writer", ipl= 0, pid= 120 -Traceback= 604412B0 60442F08 6053A7F8 6046BD4C 604743BC 604006DC 604008AC 60437D2C 60437D18
%SYS-2-CFORKMEM: Process creation of Async tty Reset failed (no memory). -Process= "Serial Background", ipl= 0, pid= 7 -Traceback= 60463CD0 6045605C 6065EB10 6065ED80 609C4F30 6000FF3C 60437D2C 60437D18
%MODEM-1-DL_FAIL: Firmware download failed for slot 3341 module_mask 609A8D88 %SYS-3-LOGGER_FLUSHED: System was paused for 00:00:00 to ensure console debugg ing output.The access server then reloads with the following bus error:
System returned to ROM by bus error at PC 0x6097DD50, address 0xDEADBEFB at 10:43:38Workaround: Use a Cisco IOS release that contains the fix for this caveat.
•
After a Cisco AS5800 access server is reloaded, the modem modules on the Cisco AS5800 universal access server double-density modem card (DS58-144DM-CC) may not be downloaded with the firmware as specified by the Service Processing Element (SPE) commands in the configuration. Modem modules are loaded with the default Cisco IOS software bundled firmware and are indicated by the "!-Upgrade request is pending message." This condition occurs regardless of the slot, Modem ISDN Channel Aggregation (MICA) technologies card, or modem module that is used.
Workaround: Manually reconfigure a SPE with the specified firmware location for the modem module range that is affected. This configuration will cause the router to download the firmware into each of the MICA technologies cards separately.
If you are running four MICA cards in slots 4 through 7, the following global configuration commands need to be configured:
spe 1/4/0 1/4/11 firmware location flash0:mica-modem-pw_2_7_2_1.bin !
spe 1/5/0 1/5/5 firmware location slot0:mica-modem-pw_2_7_2_1.bin !
spe 1/5/6 1/5/11 firmware location flash0:mica-modem-pw_2_7_2_1.bin !
spe 1/6/0 1/6/11 firmware location slot0:mica-modem-pw_2_7_2_1.bin !
spe 1/7/0 1/7/11 firmware location flash0:mica-modem-pw_2_7_2_1.binThese configuration commands will cause the router to download the firmware into the MICA separately and confine the affected modem modules to the last modules on one slot (for example 5 to 11 or 10 to 11).
To workaround this problem, alter the modem recovery time to 15 minutes later than the reload time using the following global configuration commands (assuming that the reload time is set at 4 am):
modem recovery maintenance time 4:15
modem recovery maintenance stop-time xx:yy
modem recovery maintenance window xx:yyNote: The maintenance time must be configured to start after the reload time.
Basic System Services
•
If you perform a write erase that erases the configuration in NVRAM, the boot variables are not changed.
Workaround: Change the boot configuration and configuration register, then save the configuration before erasing NVRAM.
•
The Flash MIB implementation for a High End System (HES) in Cisco IOS software does not provide correct information for the following objects:
–
–
–
There is no workaround.
•
A Cisco router that is using a compression service adapter (SA-Comp/4) may appear to be using software compression instead of hardware compression when the show compress EXEC command is entered. There is no workaround.
•
When a very large number of managed chunks is allocated, the chunk manager may cause a router to unexpectedly reload when those chunks are freed. There is no workaround.
•
A router stops responding to the command-line interface (CLI), pauses indefinitely, and subsequently reloads if the power-on self test (POST) data has the following attributes:
–
–
–
There is no workaround.
•
A router may reload if the show running-config privileged EXEC command is issued while the community string in the snmp-server community string global configuration command is set to a string that has more than 128 characters.
Workaround: Avoid using community names that have more than 128 characters.
•
A Cisco router that is configured with Data Compression over Frame Relay Implementation Agreement (FRF.9) compression may experience a loss of synchronization between the compression peers and may not complete FRF.9 negotiation.
Workaround: Enter the shutdown followed by the no shutdown interface configuration commands simultaneously on both sides of the serial interface. When the Cisco IOS software is upgraded to correct this condition, both compression peers have to be upgraded. If only one compression peer is upgraded, negotiation may continue to fail.
•
A Cisco Versatile Interface Processor (VIP) may reload at the vip_fib_fs process when the VIP is sending traffic. There is no workaround.
•
If the target address is longer than 92 bytes for Response Time Reporter (RTR) Domain Name System (DNS) probes, the Cisco IOS software will reload.
Workaround: Avoid looking up host names that are longer than 92 bytes.
IBM Connectivity
•
A router that is running Cisco IOS Release 12.1(10.1) or a later release with data-link switching (DLSw) and Ethernet Redundancy (ER) may reload. There is no workaround.
•
Cisco IOS Release 12.1(10) software will fail to correctly determine the Fiber Distributed Data Interface (FDDI) port adapter revision level. FDDI port adapters with revision levels 1.03 to 1.10 and 1.14 and later revisions can process the routing information field (RIF) of source-route bridging (SRB) frames in hardware. Cisco IOS Release 12.1(10) software may incorrectly read the revision level as 1.1 and begin processing SRB frames in software. This condition impacts performance and incorrectly reports drops when a frame is dropped after the frame has transversed the source-bridge ring-group number. There is no workaround.
•
A Cisco 1720 router that is running Cisco IOS Release 12.2(2)T1 or Release 12.2(5) with the data-link switching plus (DLSw+) Ethernet Redundancy feature enabled may reload because of a segmentation violation (SegV) exception.
Workaround: Remove the DLSw+ Ethernet Redundancy feature and configure data-link switching (DLSW) with transparent bridging on the Ethernet segment of the router.
•
In a network of Cisco 7200 routers that use the network processing engine (NPE-225), source-route bridging (SRB) frames are dropped under a light load when data-link switching (DLSw) or TCP is used to connect SRB to the Fiber Distributed Data Interface (FDDI) on the mainframe. DLSw circuits are disconnected if there is a high load.
Workaround: Use the NPE-200 instead of the NPE-225 network processing engine.
Interfaces and Bridging
•
A Versatile Interface Processor (VIP) may reload if the following command sequence is applied to the controller on a Channelized T3 Interface Processor (CT3IP):
–
–
–
All commands that are entered after the no t1 3 timeslots 1-24 command may cause the VIP to reload.
Workaround: To prevent the controller from reloading, enter the commands in the following order:
–
–
–
or
–
–
–
Be certain to enter the no t1 3 timeslots 1-24 command after the no t1 3 clock source line command.
•
A Cisco 7100 router that is running the c7100-jk2o3s-mz.121-5a.E image of Cisco IOS Release 12.1(5a)E may experience spurious memory access when an Ethernet interface is enabled. There is no workaround.
•
A Cisco router that is running Cisco IOS Release 12.1(8a)E1 with a channelized E1 ISDN PRI port adapter (PA-2CE1) or a channelized T1 ISDN PRI port adapter (PA-2CT1) and that has class-based fair weighted queueing (CBWFQ) configured may experience a reload. There is no workaround.
•
There is a continuous leak of memory when transparent bridging is configured and when an access control list (ACL) is applied to an interface. There is no workaround.
•
A Cisco Catalyst 6000 switch or a Cisco 7500 series router with a T1/E1 or T3/E3 port adapter may fail to perform TFTP after a certain number of online insertion and removals (OIRs) are performed on the Versatile Interface Processor (VIP) or Flex WAN cards that have the T1/R1 or T3/E3 port adapters installed. There is no workaround.
•
A high CPU utilization condition may occur when a Cisco 7206VXR router that is running Cisco IOS Release 12.1(9) is installed with a PA-A2-4T1C-OC3SM or PA-A2-4T1C-T3ATM port adapter. In the command output from the show process cpu EXEC command, the "uSecs" of "Net Background" and "ATM periodic" indicate increasing values.
Workaround: Enter the no shutdown command on the ATM interface.
•
A Cisco 7500 series router that is running Cisco IOS Release 12.0(16)S01 with a channelized T3 (CT3) single wide port adapter on a Versatile Interface Processor (VIP) may experience flapping interfaces that are created on the CT3 controller.
Workaround: Reload the VIP as a temporary workaround.
•
The ip mtu bytes interface configuration command may be missing from a Token Ring interface after the interface is reloaded. There is no workaround.
•
When a dedicated Token Ring port adapter (PA-4R-DTR) is used with a Versatile Interface Processor (VIP-2) on a Cisco 7500 router, the Token Ring interface that is configured for full duplex and 16 MB does not recover if the Cisco Catalyst 3920 switch that is configured with a full duplex port (FDX-port) (that the PA-4R-DTR interface is connected to) is power cycled.
Workaround: If the Cisco Catalyst 3920 switch port is configured for AUTO and not FDX-port, the PA-4R-DTR interface will recover in the half-duplex more, The PA-4R-DTR interface will recover in the full-duplex mode after the shutdown interface configuration command followed by the no shutdown interface configuration command is entered on the interface.
IP Routing Protocols
•
A Cisco 7500 router that is running a Cisco IOS release later than Release 12.10(5) with distributed Cisco Express Forwarding (dCEF), Network Address Translation (NAT), and access lists configured may fail to transmit pings that are sent between two routers through the Cisco 7500 router. There is no workaround.
•
A Cisco router that is running Cisco IOS Release 12.2(3) may reload unexpectedly with a software-forced reload when running Enhanced Interior Gateway Routing Protocol (EIGRP). This condition affects routers that are running EIGRP-IP, EIGRP-IPX, and EIGRP-APPLETALK.There is no workaround.
•
When a static route is configured through an interface and the interface is subsequently shut down, the static route is still shown as invalid. This situation will occur only if all of the following conditions are met:
–
–
–
–
Workaround: Clear the static route using the clear ip route destination gateway privileged EXEC command.
•
A router may reload if NetFlow-data is exported to a multicast address.
Workaround: Do not configure a multicast address; use a unicast address instead.
•
After several hours of operation, a Cisco 1750 router that has Network Address Translation (NAT) and Port Address Translation (PAT) enabled may fail to establish new PAT sessions. New PAT sessions cannot be established from a single add-pool with overload. This condition does not occur when a nonoverload configuration is used. There is no workaround.
•
Some type 3 link-state advertisements (LSAs) may be generated with MAX-AGE for a virtual private network (VPN) Open Shortest Path First (OSPF) network after the corresponding Border Gateway Protocol (BGP) updates are received for a site that has a backdoor link. The clear ip route destination gateway privileged EXEC command can be used to force LSA generation. There is no workaround.
•
On a Cisco router that is running Cisco IOS Release 12.0(19)ST, the Border Gateway Protocol (BGP) does not mark the autonomous system (AS) paths that exceed the limit configured by the bgp maxas-limit configuration command as invalid. There is no workaround.
•
Multicast traffic stops after the shutdown interface configuration command is entered followed by the no shutdown command interface configuration command on a Packet-over-SONET (PoS) interface. This condition may occur when any link-up or link-down event takes place.
Workaround: To clear this condition, enter the clear ip mroute * EXEC command on the PoS interface.
ISO CLNS
•
After a DECnet neighbor is removed from a network, the neighbor may still be advertised in Intermediate System-to-Intermediate System (IS-IS), link-state packet (LSP), until the LSP times out. The new LSP that is generated contains the correct information. This condition is observed in Cisco IOS Release 12.0(5)T.
Workaround: Enter the clear isis * EXEC command.
•
DECnet Phase IV and Phase V translation may not work after an upgrade from Cisco IOS Release 12.0(18) to Release 12.1(9) or 12.2(1b). Packets that arrive at the router are not translated or sent to the end node but are instead looped into the network until the lifetime expires. There is no workaround.
Miscellaneous
•
A Cisco 6400 series router that is running Cisco IOS Release 12.1(1)DC1 or 12.1(3)DC1 may experience a memory leak in the PPP authentication process. The memory leak is most noticeable when the router is terminating a large number of PPP sessions and there is a high level of PPP authentication processing. There is no workaround.
•
When a member link is shut down in a distributed Multilink PPP (dMLP) bundle, the Multilink interface may flap. The following is a sample syslog output is displayed when this condition occurs:
%SYS-5-CONFIG_I: Configured from console by console
%LINK-5-CHANGED: Interface Serial0/0/0:0, changed state to administratively down
%LINK-3-UPDOWN: Interface Multilink1, changed state to down
%LINEPROTO-5-UPDOWN: Line protocol on Interface Serial0/0/0:0, changed state to down
%LINEPROTO-5-UPDOWN: Line protocol on Interface Serial0/0/1:0, changed state to down
%LINEPROTO-5-UPDOWN: Line protocol on Interface Multilink1, changed state to down
%LINK-3-UPDOWN: Interface Multilink1, changed state to up
%LINEPROTO-5-UPDOWN: Line protocol on Interface Serial0/0/1:0, changed state to up
%LINEPROTO-5-UPDOWN: Line protocol on Interface Multilink1, changed state to upThere is no workaround.
•
When you perform an online insertion and removal (OIR) on a trunk card, a corresponding D channel configuration is lost.
Workaround: Manually reconfigure the interface after performing the OIR.
•
On a Cisco 7200 series I/O controller card with a Fast Ethernet (FE) interface (RJ-45) that is configured with Protocol Independent Multicast (PIM) and full duplex, the line protocol flaps when the cable is removed or fails instead of just going to line protocol down.
Workaround: Remove either PIM or full duplex from the interface.
•
A Cisco AS5800 universal access server router shelf may reload intermittently when the show modem EXEC command is issued. There is no workaround.
•
On a Cisco 7200 series router that is running Cisco IOS Release 12.1(5a), a PA-8T synchronous serial port adapter that is configured with High-Level Data Link Control (HDLC) over leased lines may have ports 4 to 7 or ports 0 to 4 going in the up or down state without any visible cause.
Workaround: Reload the router.
•
The E1 feature board (FB) does not boot with Cisco IOS Release 12.2(0.11) or Release 12.2(0.12). There is no workaround.
•
A Cisco AS5300 series access server may reload under heavy load and display the following stack:
Router#show stack
Minimum process stacks
Free/Size Name 5704/6000 Reset ipc queue 2572/3000 fstp init 2388/3000 allegro libretto init 7388/12000 Init 7436/9000 DHCP Client 5288/6000 RADIUS INITCONFIGInterrupt level stacks:
Level Called Unused/Size Name 2 5484628 7988/9000 Low IRQ Int Handler 3 19 8448/9000 High IRQ Int Handler 4 36100877 8600/9000 Console Uart 6 0 9000/9000 Parity interrupt 7 61183144 8604/9000 NMI Interrupt Handler
System was restarted by bus error at PC 0x60C10D9C, address 0x20 Stack trace from system failure:
FP: 0x6205E0D0, RA: 0x60C10D9C FP: 0x6205E128, RA: 0x60ADB01C FP: 0x6205E150, RA: 0x60C17D08 FP: 0x6205E170, RA: 0x60C17E2C FP: 0x6205E288, RA: 0x60C2ACC8 FP: 0x6205E2A8, RA: 0x60C2B430 FP: 0x6205E2F0, RA: 0x60C2576C FP: 0x6205E320, RA: 0x60C25AF0There is no workaround.
•
One side of a circuit emulation service (CES) E1 Port of a Cisco LightStream LS1010 switch connection may remain consistently in the YELLOW alarm or ALARM INTEGRATION state. There is no workaround.
•
In a Cisco AS5800 universal access server that is running Cisco IOS Release 12.2(1), PPP may fail to start for modem calls. This occurs when a UPC324 card is reconfigured from Router-shelf1 to another Router-shelf2 in a split shelf mode. There is no workaround.
•
When a Hot Standby Router Protocol (HSRP) router with a higher priority is added to a network, the HSRP state changes continuously from "Active" to "Speaking," from "Speaking" to "Standby," and from "Standby" to "Active." When the HSRP state changes from "Active" to "Speaking," the router resets the interface in order to remove the HSRP MAC address from the interface MAC address filter. The switch detects this link state change on the interface, and a Spanning Tree Protocol transition takes place. The spanning tree takes 30 seconds (twice the default Forward Delay time of 15 seconds) to transition the port into the Forwarding state.
Workaround:
–
–
–
–
–
•
The Tag Distribution Protocol (TDP) uses TCP port 711 for communication between TDP peers. This is enabled on a router by default in Cisco IOS Release 12.1. The port becomes active once tag-switching is enabled on a single interface. The TCP port will remain open even after tag-switching is disabled on the router, both at the interface level and the global level.
Workaround: Reload the router to close the TCP port.
•
Creating new channel-groups or subinterfaces on a PA-A3 port adapter may cause the corruption of an existing subinterface on the same controller. When this situation occurs, the corrupted subinterface changes its status to down/down and triggers an "output frozen" condition that results in "cbus complex" restarts on all Versatile Interface Processors (VIPs). Sometimes it is possible to bring the corrupted interface back into service by undoing the channel-group change and clearing the E3 controller. There is no workaround.
•
A router shelf may reload at WATCHDOG TIMER RESET when the pri-group timeslots 1-24 command is removed from the controller configuration. There is no workaround.
•
The logging synchronous line configuration command may cause logging to stop.
Workaround: Remove this command.
•
A Cisco AS5400 universal access server that is running Cisco IOS Release 12.2 for incoming E1/R2 calls sends an abnormal clear-forward (AB=10) signal in the seize-acknowledge state. Instead, the Cisco AS5400 should send a congestion signal followed by a clear-backward (AB=11) signal and wait for the remote site to send a clear-forward signal. Switches of certain types take exception to the abnormal signal and block the circuit. There is no workaround.
•
A Cisco router may experience a memory leak when both Generic Routing Encapsulation (GRE) and IP Security (IPSec) are configured on a network. There is no workaround.
•
The Permanent Virtual Connection (PVC) to Virtual LAN(VLAN) bindings are not intact after reloading the ATM Module (WS-X516X) with a large number of PVC to VLAN bindings.
Workaround: Copy the startup configuration to running configuration using the following command on ATM Line card. copy startup-config running-config
•
A Cisco 7200 series router that uses Cisco Encryption Technology (CET) may lose Digital Signature Standard (DSS) keys when the router is upgrading to newer versions or reverting to older versions of Cisco IOS Release 12.1 software.
Workaround: Generate a new pair of DSS keys and exchange the new DSS keys with the peer routers.
•
IP Multicast does not work over a GRE tunnel when IPSec is configured. Routing protocols may work.
Workaround: Remove IPSec or send IP Multicast over a different unencrypted tunnel
•
A phantom entry may appear for a dial shelf controller on a Cisco AS5800 universal access server in the ENTITY-MIB.physicalTable.
Workaround: Reload the chassis.
•
When the no shutdown command is entered after the shutdown command is entered on a circuit emulation service (CES) card, a burst of bit errors is noticed on the circuit on some other ports on the same peripheral adapter module (PAM). This condition is observed on the Cisco 8540 multiservice switch router (MSR), the Catalyst LS1010 switch, and the Cisco 7200 series router and can affect both E1 or T1 interfaces. There is no workaround.
•
A Protocol Independent Multicast-sparse mode (PIM-SM) shared tree is not pruned in full at the Protocol Independent Multicast-Rendezvous Point (PIM-RP) during a shortest path tree (SPT) switchover. There is no workaround.
•
Phantom objects may be created in the ENTITY-MIB.physicalTable with the type DS1 and the name DS-T1 on a Cisco AS5800 universal access server.
Workaround: This condition will clear after the access server is reset.
•
A Cisco 7500 router with an Enhanced ATM port adapter (PA-A3) may issue "invalid memory allocation (MALLOC) at interrupt level" error messages when the show atm vc [vcd | interface interface-number] privileged EXEC command is entered. There is no workaround.
•
A Cisco router running Cisco IOS Release 12.2 as an Asynchronous Transfer Mode (ATM) Label Edge Router (LER) that is configured as a generic routing encapsulation (GRE) Tunnel source may find that it is transmitting GRE packets over the tag control VC (0/32) of an ATM tag sub-interface instead of the Tag Virtual Circuit (TVC) bound to the tunnel destination. If the ATM Label-Switching Routers (LSRs) do not have a route to the tunnel destination, the GRE packets will be dropped. There is no workaround.
•
The tx-ring-limit ring-limit ATM VC configuration command does not work when it is used to configure a multichannel E3 port adapter (PA-MC-E3) for a Cisco 7200 router. There is no workaround.
•
A Cisco router may lose memory resources when lines are flapping and tag-switching ip is enabled. This condition is most severe when there are multiple adjacencies to the same peer. There is no workaround.
•
If a Traffic Engineering (TE) tunnel is configured between two Provider Edge (PE) routers and Label Distribution Protocol (LDP)/Tag Distribution Protocol (TDP) is configured in the tunnel, Virtual Private Network (VPN) connectivity through the two PE routers will not function. There is no workaround.
•
A Cisco AS5800 universal access gateway router shelf may reload if a DuoDensity Modem Module (DMM) feature board is rebooted using the hw-module {slot number | subslot subslot/subcard} reset EXEC command. There is no workaround.
•
When the Hot Standby Router Protocol (HSRP) is running on an interface, a router that is actively responding to the shared HSRP address does not put the address into the Simple Network Management Protocol (SNMP) ipAddrTable. There is no workaround.
•
Modem ISDN Channel Aggregation (MICA) technologies firmware that is bundled with Cisco IOS software on a Cisco 3600 router has to be upgraded from version 2.7.2.0 to version 2.7.3.0.
Refer to the release notes for MICA firmware 2.7.3.0 at the following URL:
http://www.cisco.com/univercd/cc/td/doc/product/access/acs_serv/5300/sw_conf/sw _ports/fwpwrn/mica/mca2730.htm
Workaround: Download the mica-modem-pw.2.7.3.0.bin portware from the MICA modem portware homepage at the following URL: http://www.cisco.com/pcgi-bin/tablebuild.pl/mica
Transfer the bin file into one of the Flash cards on the router. The Cisco IOS software will check the version of the bin file against the bundled portware and automatically choose the latest version of the portware.
•
If a Traffic Engineering (TE) tunnel is configured between two Provider Edge (PE) routers, Virtual Private Network (VPN) connectivity through the two PE routers will not function. All PE to PE traffic should be normally routed through the tunnel.
Workaround: Run Cisco IOS 12.0(17)ST3 or earlier on the PE router or remove the auto route announce statement of the tunnel interface and configure a static route through the tunnel.
•
A dial connection from a Cisco 800 router to a Cisco 3640 router may fail after a few days. A reload of the Cisco 3640 router may allow the connection to be restored temporarily, but the connection may fail again. There is no workaround.
•
When a Simple Network Management Protocol (SNMP) set shutdown command or no shutdown command is sent to the T1 or T3 controller on a Cisco AS5400 universal access server, a blank description command is added to the configuration for the controller that received the set command. The configuration for the controller does not change if the controller already has a description defined. There is no workaround.
•
When using a PA-FE-FX port adaptor on a VIP for a 7500 platform, an OIR will change the configuration of the Fast Ethernet interface(s) to "duplex auto" and "spped auto". There is no workaround
•
The current implementation of Cisco IOS software is not fully compliant with RFC 2547bis. RFC 2547bis describes the procedures that must be implemented to specific extended communities when route attributes are passed from a customer edge (CE) router to a provider edge (PE) router. This DDTS enforces those procedures. The CE router may suggest a particular route target for each route from the route targets that the PE router is authorized to attach to the route. The PE router would then attach only the suggested route target rather than the full set. This situation gives the CE administrator some dynamic control of the distribution of routes from the CE.
With the current Cisco IOS software, the PE router allows the CE router to attach route targets in an update without verifying that they are a subset of route targets to which the virtual routing and forwarding instance (VRF) attaches. This condition causes the routes to end up in a VRF instance when they are not supposed to.
Workaround: Configure the VRF route map on the PE router to overwrite the extended community attribute to avoid the leakage of routes to other VRFs.
•
A Gigabit Ethernet interface on a Cisco 7500 router may experience Inter-Processor Communication (IPC) memd buffer problems after the output becomes stuck and may display the following message:
%RSP-3-RESTART: interface GigabitEthernet0/0/0, not transmitting Output Stuck on GigabitEthernet0/0/0Workaround: Avoid using auto-negotiation.
•
Distributed Multilink PPP (DMLPPP) and Versatile Interface Processor (VIP) local switching may not work properly on a 2-port channelized T3 port adapter. There is no workaround.
•
The Multilink Point-to-Point Protocol (ML-PPP) may fail when the fragments are less than 100 bytes in size. There is no workaround.
•
A Cisco Catalyst 6000 SP (Supervisor) switch that is running Cisco IOS Release 12.1(7)E may reload with a bus error. After the SP reloads with a bus error, the Route Processor (RP) is automatically reloaded by the SP. There is no workaround.
•
A Cisco router may reload when the no tag-switching ip interface configuration command is configured on an interface that has the tag-switching ip interface configuration command enabled on its subinterfaces. There is no workaround.
•
A Cisco Catalyst 6500 switch that is running Cisco IOS Release 6.3(1) with two Supervisor Engine 2 (WS-X6K-SUP2-2GE) with a Multilayer Switch Feature Card 2 (WS-F6K-MSFC2) and that has a FlexWAN module (WS-X6182-2PA) with an enhanced ATM OC3 multimode port adapter that is running Cisco IOS Release 12.1(7a)E1 may experience a buffer leak on the MSFC2 and display the following message:
Small buffers, 104 bytes (total 36773, permanent 50, peak 88803 @ 1w5d):
25 in free list (20 min, 150 max allowed)
95159588 hits, 286991 misses, 316049 trims, 352772 created
17027 failures (0 no memory)The sscop_sendSdPdu process is not releasing the memory buffers and causes memory allocation (malloc) errors, which cause interprocess communication (IPC) issues that may subsequently cause the FlexWAN module to be disabled and the following error message to be displayed:
%SYS-2-MALLOCFAIL: Memory allocation of 276 bytes failed from 0x40210A74, pool I/O, alignment 32This condition occurs only if the Cisco Catalyst 6509 is booted up without any configuration or if the Cisco Catalyst 6509 has just been configured with a new configuration. Memory leaks do not occur if the Cisco Catalyst 6509 is booted up after the configuration is saved to NVRAM and if any subsequent configuration changes are made.
•
Systems Network Architecture (SNA) Switching Services (SNASw) physical units (PU) may be stuck in the pending Activate Physical Unit (ACTPU) state. There is no workaround.
•
A Cisco 2600 router that is running Cisco IOS Release 12.1(10) may experience problems with the Address Resolution Protocol (ARP) when Cisco Express Forwarding (CEF) is enabled.
Workaround: Disable CEF.
•
A Cisco 7200 router that is configured with a voice port adapter and with T3 or High-Speed Serial Interface (HSSI) backhaul may experience packet drops when a low latency queueing (LLQ) service policy is attached to the T3 or HSSI interface. There is no workaround.
•
When a Cisco AS5800 universal access server is configured to download non-default firmware onto Cisco Modem ISDN channel aggregation (MICA) modems after a reload, the configured firmware is not downloaded. The firmware for the MICA modem is not downloaded even though debug messages from the debug modem firmware EXEC command indicate that the firmware has been downloaded. The status of firmware download can be further verified using the show spe version EXEC command.
Workaround: To clear this condition, configure the Cisco AS5800 after it is reloaded to download the default version of the MICA modem firmware and reconfigure the access server to download the nondefault version of the MICA modem firmware.
•
A Cisco AS5800 universal access server router shelf may reload unexpectedly when system processing engines (SPEs) that are already marked as BAD are cleared using the clear spe [slot | slot/spe] EXEC command. This condition may impact the service of the router shelf. There is no workaround.
•
The Multiprotocol Label Switching (MPLS) tags that are associated with a virtual private network (VPN) (vrf x.y.z.w) and a default route (0.0.0.0) are inconsistent between the main Cisco Express Forwarding (CEF) table and the distributed CEF (dCEF) table on the outbound Versatile Interface Processor (VIP) card.
Workaround: To recover from this condition, enter the clear ip route vrf vrf-name EXEC command.
•
The command-line interface (CLI) does not prompt for the erase keyword when the copy [/erase] source-url destination-url EXEC command is entered. This behavior does not allow a file system to be erased using the copy [/erase] source-url destination-url EXEC command.
Workaround: Enter the erase filesystem: command before entering the copy EXEC command.
•
Guest
