Datasheet

244
Release Notes for Cisco IOS Release 12.1E on the Catalyst 6500 and Cisco 7600 Supervisor Engine and MSFC
OL-2310-11
Caveats
• Some NAT translations do not expire. This problem is resolved in Release 12.1(13)E4.
(CSCdz44155)
• When you remove a GBIC from one port on a module, you receive SNMP traps for all ports on the
module. This problem is resolved in Release 12.1(13)E4. (CSCdz37642)
• Cisco IOS server load balancing (Cisco IOS SLB) connectivity might fail following a
“%ICC-SP-5-WATERMARK” message. This problem is resolved in Release 12.1(13)E4.
(CSCdy37563)
• The SNMP cefcFRUInserted and cefcFRURemoved notifications are not generated when a GBIC is
inserted or removed. This problem is resolved in Release 12.1(13)E4. (CSCdy03042)
• Changes in the Unicast routing table can cause an inconsistency between software and hardware
programming of the RPF interfaces of specific multicast groups. This inconsistency causes
group-specific multicast traffic to be lost. This problem is resolved in Release 12.1(13)E4.
(CSCdz44110)
• Some traffic sent through a Layer 2 EtherChannel that includes interfaces on different
DFC-equipped switching modules is lost. This problem is resolved in Release 12.1(13)E4.
(CSCdz29883)
• Configuring extended range VLANs causes a memory leak. This problem is resolved in
Release 12.1(13)E4. (CSCdy46743)
• A Supervisor Engine 1 with an MSFC might display the following message and reload:
%RPC-2-FAILED: Failed to send RPC request mapping_sp:get_current_mappings
This problem is resolved in Release 12.1(13)E4. (CSCdz50453)
• With multicast routing disabled, multicast traffic is routed through interfaces where PIM is
configured. This problem is resolved in Release 12.1(13)E4. (CSCdz42307)
• With ACLs configured, traffic is not routed following a switchover to the redundant supervisor
engine. This problem is resolved in Release 12.1(13)E4. (CSCdz55647)
• With BGP routing configured to use parallel links, CPU usage might be 100% for an extended period
if the link with the lowest IP address fails. This problem is resolved in Release 12.1(13)E4.
(CSCdz79139)
• IGMP snooping fast leave is not honored during the time interval between receipt of a general query
and the value of the query response interval (see RFC2236, Section 8.3) contained in the received
general query. You can configure the query response interval with the “ip igmp
query-max-response-time” interface command. For example, if the query response interval is 10
seconds in the received general query, then the interface command “ip igmp snooping fast-leave”
configured on a VLAN interface corresponding to the VLAN on which the query was received has
no effect for 10 seconds after this general query is received in this VLAN. In environments with very
high IGMP leave rates, IGMP snooping might not process some leave messages. This problem is
resolved in Release 12.1(13)E4. (CSCdz65685)
Resolved General Caveats in Release 12.1(13)E3
• Some physical ports in EtherChannels might fail to carry traffic, resulting in unexpected traffic loss.
This problem is resolved in Release 12.1(13)E3. (CSCdy48112)
• In a network that is configured for multicast redundancy, which has a high volume of multicast
traffic where the last-hop Catalyst 6500 series switches or Cisco 7600 Series Routers are the
Anycast RP pair, and SPT threshold infinity is configured, the non-DR router might experience high
CPU. This problem is resolved in Release 12.1(13)E3. (CSCdz48825)