Datasheet

264
Release Notes for Cisco IOS Release 12.1E on the Catalyst 6500 and Cisco 7600 Supervisor Engine and MSFC
OL-2310-11
Caveats
• With Supervisor Engine 2, multicast packets that set the router alert option, like IGMP general
queries and membership reports, might not be handled properly, which might disrupt IGMP client
connectivity. This problem is resolved in Release 12.1(12c)E4. (CSCdy84078)
• With Supervisor Engine 2, packet loss might occur for a few seconds after routing protocol multicast
packets are received. This problem is resolved in Release 12.1(12c)E4. (CSCdy58383)
• If the shared tree and the shortest path tree (SPT) diverge because of a Reverse Path Forwarding
(RPF) change on the shared tree (normally triggered by a restoration of a failed link), the SPT is
pruned. A join to restore the traffic flow immediately follows the prune, which might interrupt traffic
briefly. This problem is resolved in Release 12.1(12c)E4. (CSCdu74664)
• If you configure an active member port of an EtherChannel as a SPAN source port, the port goes into
the suspended state and does not pass any traffic. In Release 12.1(12c)E4, you cannot configure an
active member port of an EtherChannel as a SPAN source port. (CSCdx81246)
Resolved General Caveats in Release 12.1(12c)E2
• Switchover to the redundant supervisor engine and MSFC incorrectly does not occur when excessive
interrupts on the supervisor engine cause the MSFC to reload. This problem is resolved in
Release 12.1(12c)E2. (CSCdx43539)
• A reload caused by excessive interrupts might fail and CPU-6-MONITOR-SP-NOT-HEARD
messages are displayed. This problem is resolved in Release 12.1(12c)E2. (CSCdy27356)
• With heavy traffic, a SYS-3-NULLIDB traceback message might be displayed and crypto tunnels
might go down. This problem is resolved in Release 12.1(12c)E2. (CSCdw01724)
• Spurious accesses occur when you clear one of the VPNv4 Multicast Border Gateway Protocol
(MBGP) members of a Multi-Protocol Border Gateway Protocol (MP-BGP) peer group. This
problem is resolved in Release 12.1(12c)E2. (CSCdw03988)
• Nonmaskable interrupts (NMIs) might cause a Supervisor Engine 1 to reload, and then the reload
might fail. This problem is resolved in Release 12.1(12c)E2. (CSCdy25902)
• After primary load-balancer (SLB or FWLB) failure and recovery, if no SLB connection objects
were bound to the sticky table entries on the backup load-balancer, the primary load-balancer sticky
table might be incomplete. This problem is resolved in Release 12.1(12c)E2. (CSCdy28514)
• Duplicate RMON alarms are sent when there is more than one snmp-server host command in the
configuration. This problem is resolved in Release 12.1(12c)E2. (CSCdx89905)
• If both IOS Server Load Balancing (SLB) and IOS Firewall Load Balancing (FWLB) are configured on
the same router, ICMP packets that need to be firewall load balanced are routed by SLB instead. This
problem is resolved in Release 12.1(12c)E2. (CSCdy18588)
• The show module command always displays the state of a redundant PFC as other. This problem
is resolved in Release 12.1(12c)E2. (CSCdy19682)
Resolved General Caveats in Release 12.1(12c)E1
• A multilink PPP bundle might forward traffic intermittently. This problem is resolved in
Release 12.1(12c)E1. (CSCin03257)
• To avoid a reload, do not shut down any interfaces while another user is displaying the output from
the show ip pim neighbor command. This problem is resolved in Release 12.1(12c)E1.
(CSCdx25551)
• A Gigabit Ethernet EtherChannel formed with DFC-supported ports and nonDFC-supported ports
and configured not to use PAgP may stop passing traffic if you reset or OIR one of the Gigabit
Ethernet switching modules. This problem is resolved in Release 12.1(12c)E1. (CSCdx00390)