Caveat List for
Cisco IOS Release 12.2(33)CX
This chapter describes open and resolved severity 1 and 2 caveats and select severity 3 caveats:
-
The “Open Caveats” sections list open caveats that apply to the current release and may apply to previous releases. A caveat that is open for a prior release and is still unresolved applies to all future releases until it is resolved.
-
The “Resolved Caveats” sections list caveats resolved in a specific release, but open in previous releases.
The bug IDs are sorted alphanumerically.
 Note | The Caveats section includes the bug ID and a short description of the bug. For details on the symptoms, conditions, and workaround for a specific caveat you must use the Bug ToolKit at: https://tools.cisco.com/bugsearch. |
Caveat List for Cisco IOS Release 12.2(33)CX
This chapter describes open and resolved severity 1 and 2 caveats and select severity 3 caveats:
- The “Open Caveats” sections list open caveats that apply to the current release and may apply to previous releases. A caveat that is open for a prior release and is still unresolved applies to all future releases until it is resolved.
- The “Resolved Caveats” sections list caveats resolved in a specific release, but open in previous releases.
The bug IDs are sorted alphanumerically.
Note | The Caveats section includes the bug ID and a short description of the bug. For details on the symptoms, conditions, and workaround for a specific caveat you must use the Bug ToolKit at: https://tools.cisco.com/bugsearch |
Open Caveats—Cisco IOS Release 12.2(33)CX
|
Bug ID |
Description |
|---|---|
|
When the show licenses command is used, it displays incorrect count of licenses. |
|
|
When a wrong application-type (for example: cable application-type 7 include multicast 16711680) is configured, the Standby PRE crashes with MCL error. |
|
|
When the command line interface is used to access a mapped upstream interface, the command line interface stops responding. Shut down the mapped upstream interface or reset the line card, to recover from this issue. |
|
|
Route Processor (RP) CPU usage spikes to 99% as IP Input process causes a spike of about 65%, because of the high channel change rate per cable modem. |
|
|
IPC via Ironbus and datapath cannot go through some line cards and the active PRE. SLOT 8/1: Jan 28 22:29:59.080 SGT: %CR10K_CLNT-3-IPCTXERR: IPC send failure: client=5 entity=0 ses=8/1 msg=50 err=timeoutSLOT 8/1: Jan 28 22:30:05.692 SGT: %NICKEL10G-3-N10GERREVT: TBB - ERR EVENT OVERFLOW, port 0SLOT 8/1: Jan 28 22:30:05.692 SGT: %NICKEL10G-3-N10GERREVT: STT - ERR EVENT ERRPKT, port 0 |
|
|
SNMP entPhysicalModelName shows small form-factor pluggable (SFP) value as “N/” instead of “N/A” when the model name of the SFP module seated in the Cisco 5-Port Gigabit Ethernet SPA is not programmed for IDPROM. |
|
|
During Cable Vision testing on CRDC LS test bed, the modems take more than one hour to come online. |
|
|
DOCSIS 2.0 modems go offline. "SINGLE_BIT_SEU_INT" error is seen before the modems go offline. |
|
|
On the TenGigabitEthernet interfaces of the Cisco uBR10012 router configured with WAN load-balancing feature, the IPv6 traffic is dropped for the failed IPv6 RPFs. The IPv6 RPFs fail after switchover to the redundancy load-balancing interfaces. |
|
|
When many RSVP service flows are created by one Cable Modem (CM) and PRE switches over |
|
|
The show cable clock command displays the number of Timing, Communication and Control (TCC) cards in the Chassis as zero, as the DOCSIS Timing and Control Card (DTCC) continuously sends IPC timeout messages. |
|
|
CPUHOG occurs when a class-map with many filters are applied with a policy-map to a WAN interface and the following message is displayed: Jul 2 18:29:42.694 CST: %C10KEHSA-3-STANDBY_CPUHOG: CPUHOG on Remote PRE! Proc: "SuperACL", PID: 302, -Traceback= 0x10DCC700z 0x10DCC8D4z 0x10DCC970z 0x10DCCEB8z 0x10DD1504z 0x10BE3EECz 0x10BF0E24z 0x10BF87D4z 0x10BE9B44z 0x10DE |
|
|
A large number of BCM3142-3-TRUNCATED errors occur within short span of time. |
|
|
When packet filter groups are continuously created and deleted, and scripts are run with multi-user console, the PRE crashes at the cmts_create_pkt_filter_group_acl operation. |
|
|
The Cisco 10000 Series SPA Interface Processor-600 crashes after online insertion and removal of the Cisco 3 Gbps Wideband Shared Port Adapter. |
|
|
The Forwarding Information Base (FIB) table does not refresh after PRE5 online insertion and removal (OIR) and deleting the Multiprotocol Label Switching (MPLS) IP address on the interfaces. |
|
|
Malformed L2TP packets are found after PRE switchover. |
|
|
Unable to send IPC messages after PRE switchover. |
|
|
When the version string is longer than 256 characters, the line card DOS version string includes the string "set_chunk_alignment: arg %d not a power of two!” |
|
|
Working line card is powered off after the DOCSIS Timing Interface (DTI) clock reference mode is terminated and the standalone mode is restarted using the no cable clock dti command. |
|
|
Service Independent Intercept (SII) IPv4 tap does not work when citapStreamFlowId is configured. |
|
|
Inter-Process Communication (IPC) messages cannot be sent during the Route Processor (RP) switchover. |
|
|
Choppy voice is observed in the upstream direction during the calls due to packet loss on the Unsolicited Grant Service (UGS) flows on the Cisco CMTS with Cisco uBR-MC3GX60V line cards. |
|
|
When SNMP polling docsSubmgt3CPeIpTable, docsSubMgtCPeIpTable is done with about 7MD and 700 cable modems but no CPE is added to the CMTS, the following error message is observed periodically on Route Processor, and the SNMP poll fails: "SNMP-3-INPUT_QFULL_ERR: Packet dropped due to input" |
|
|
Wideband cable modem is online using a Wideband-Cable interface that has less channels than it is capable of bonding even when a higher channel Wideband-Cable interface is available for the modem to use. This occurs when a modem capable of 16 or 24 channels initially acquires on a Modular-Cable controller that only has a 4 channel Receive Channel Configuration (RCC) and is then steered to a different Modular-Cable controller with 4, 8, and 12 channel RCC available. When the cable modem moves to the controller with 12 channels available, load balancing configures a 4 channel RCC for the modem. |
|
|
An SSH user is shown to be logged in on VTY port 1 although there is no user logged in as SSH user on VTY port 1. |
|
|
Online Insertion and Removal (OIR) operation on a Cisco UBR-MC20X20V line card causes access to wrong memory addresses, resulting in “Spurious memory access” error and tracebacks. |
|
|
When oversubscribed IPv6 traffic is sent, a Toaster stall error occurs and the PXF crashes. |
|
|
The following messages and tracebacks occur on a CMTS running Cisco IOS Release 12.2(33)SCG3. There were no occurrences of PRE or line card switch-overs. This occurs in and when modems are generating excessive TEK requests. |
|
|
When you enable TLV35/63 feature (limiting CPE IP/IPv6 addresses) in the configuration file of the cable modem, the DHCP server continues to assign IP addresses in response to the restricted CPE's requests which are made through the CMTS. The restricted CPEs do not get the assigned addresses. |
|
|
The association between the cable modem and the access group is lost during Dynamic Channel Change (DCC) or PRE switch over, when an association is created using the cable modem CM-MAC access-group or cable device CM-MAC access-group commands |
|
|
When the DEPI EQAM Statistics feature is enabled on a chassis fully loaded with Cisco 6G SPA cards, CPU utilization increases by 20%. |
|
|
Traceback occurs at the dhcpv6_relay_process operation: REMOVE OVERLAP ENTRY:CPE Found MAC:7cb2.1b0b.f316, from:Cable8/0/14, MAC:7cb2.1b0b.f316, IP:0.0.0.0, IPv6:2001:100:103:7:7EB2:1BFF:FE0B:F5D2, SID:45, found sid:45, ptr_sid:45 |
|
|
The Cisco UBR-MC20X20V line card crashes with the following message in the crash information file: Unexpected exception to CPU :vector 300, PC = 0x2DCD7F0 , LR = 0x2DCD598 |
|
|
When changes are made to upstream or downstream RF Channels or frequencies, channel bonding stops working. DOCSIS 3.0 and Wideband cable modems register as DOCSIS1.0 or Narrowband modems. |
|
|
When the parser config cache interface command is configured and the linecard is switched over and reverted back, modems fail to come online. the modems are stuck in reject(m) state and the shared-secret configuration is lost. |
|
|
When PRE high Availability is triggered repeatedly by a script and line card switch over is also triggered by a crash, the following memory leak occurs on the protect line card: BC1EE44 4 2818D4E0 (GUARDIAN_DATA) BC1EE5C 4 2818D4E0 (GUARDIAN_DATA) BC1EE74 4 2818D4E0 (GUARDIAN_DATA) |
|
|
When Online Insertion and Removal (OIR) is performed on line card, the PRE reloads unexpectedly. |
|
|
When the clear cable modem all delete command is used, mass modem re-registration may result in modem(s) dropped into the CMTS Filter Group 255 due to missed Inter-Process Communication (IPC) between Route processor and line cards. |
|
|
After enabling IPv4 leasequery, when downstream traffic is sent to CPE and the line card is switched over an reverts back, the downstream traffic is blocked. |
|
|
When snmp-server enable traps docsis-resil resil-events command is configured, all the options specified for resil-events. This is expected behavior. |
|
|
When the codeword counters (using CLI configuration) for an upstream interface reach the 32-bit limit, they get reset. This provides wrong counter information. However, the counter information provided by SNMP is accurate. |
|
|
When a high number of messages are generated, the logger process fails due to a full message queue. The interface slot fails, causing failure of the line card or even the RP. |
|
|
When cable modem max-cpe X configuration is set in the global configuration with a value smaller than the total number of CPEs behind a cable modem, the following error is logged frequently: %UBR10000-5-MAXHOST Interface CableX/Y/Z, New host with IP address 0.0.0.0 and MAC xxxx.xxxx.xxxx on SID xxx (CM xxxx.xxxx.xxxx) is ignored in error handling path. |
|
|
When a modem device class has been excluded from load balancing group, the output of the show cable load-balance docsis-group modem-list command does not show “X” for those modems that are in the excluded list, under the MUPFXLR field. |
|
|
IPDR stops sending new records to the collector and the queuedUnacknowledged count is stuck at a non-zero value. |
|
|
On a Cisco uBR7200VXR router, the SNMP counter for ifHCOutOctets rolls back to zero when the 32-bit value for Integrated-Cable downstream interfaces is reached. This counter is expected to hold up to the maximum 64-bit value. |
Feedback