Datasheet

126
Release Notes for Cisco IOS Release 12.1E on the Catalyst 6500 and Cisco 7600 Supervisor Engine and MSFC
OL-2310-11
Caveats
• Data Link Switching (DLSw) circuits might not connect using DLSw Ethernet redundancy. This
problem occurs when DLSw Ethernet redundancy is configured with the following commands where
the local-mac and remote-mac values are the same real MAC addresses of the remote host:
– dlsw transparent switch-support
– dlsw transparent map local-mac local-mac
– remote-mac remote-mac
If both DLSw routers are rebooted, clients can immediately establish a session with the remote host
through one of these DLSw routers using the real MAC address. This real circuit is outside of
Ethernet redundancy, and until the circuit is disconnected, Ethernet redundancy cannot be
established. This problem is resolved in Release 12.1(27b)E. (CSCeh18295)
• DLSw load balancing using the circuit-count configuration does not distribute circuits evenly. This
problem occurs when all the circuits attempt to connect at the same time. Configure the dlsw
load-balance round-robin command initially, start DLSw, and then configure using dlsw
load-balance circuit-count. This problem is resolved in Release 12.1(27b)E. (CSCeh18390)
• A system may lock for a long period of time while several instances of the following message are
displayed:
%CWAN_RP-4-SEMAHOG: Process 74 (MIP Mailbox)hogging CCB BLK Sema 10! calling proc 221
(IFCOM Msg Hdlr)
This problem is resolved in Release 12.1(27b)E. (CSCin79495)
• PPP packets are dropped while running software compression under heavy traffic. This problem is
resolved in Release 12.1(27b)E. (CSCsa47223)
• Malformed VTP packets can cause a reload when debugging output is enabled for VLAN Trunk
Protocol (VTP) events. This problem is resolved in Release 12.1(27b)E. (CSCsa82334)
• If a channel port receives a bridge protocol data unit (BPDU) message with an inconsistent VLAN
ID from a neighbor, the supervisor engine successfully puts the port in a Port VLAN ID (PVID)
inconsistent state but fails to block that port. The channel port could eventually move to the
forwarding state due to an update in the STP topology. This situation occurs when the channel port
is acting as a standby and is further blocked due to a PVID inconsistency. This situation could also
occur if the inconsistency is not cleared. This problem is resolved in Release 12.1(27b)E.
(CSCsb10031)
• When a Multiple Spanning Tree (MST) designated port receives a topology change (TC) from
another MST region, it does not become a boundary port and it flushes only the internal spanning
tree (IST) instance. This situation causes traffic loss in topologies that have VLANs mapped to
instances other than the IST. This problem is resolved in Release 12.1(27b)E. (CSCsb79590)
• A standby supervisor engine in SSO mode might reload. This problem occurs when SNMP fills a
data structure, and overwrites a byte of memory after the data structure. This problem is resolved in
Release 12.1(27b)E. (CSCsc07793)
• When the keep-exchanges argument in the frame-relay lmi-n391dte command has a value that is
lower than 3, Frame Relay autosensing does not function. This problem is resolved in
Release 12.1(27b)E. (CSCea30197)
• When you enter the eigrp log-event-type dual xmit transport command and enable attributes and
then enter the show ip eigrp events command to display the event log, a reload might occur . This
problem is resolved in Release 12.1(27b)E. (CSCdx67130)
• A software-forced reload might occur after you configure and unconfigure Automatic Protection
Switching (APS) revert. This problem is resolved in Release 12.1(27b)E. (CSCee91733)