Cisco Cisco IOS Software Release 12.2(35)SE

Page de 34
 
20
Release Notes for the Catalyst 2970 and Catalyst 3750 Switches, Cisco IOS Release 12.2(18)SE
OL-5023-04
Open Caveats
In the Front Panel view or Topology view, CMS does not display error messages in read-only mode 
for these switches:
Catalyst 2900 XL or Catalyst 3500 XL member switches running Cisco IOS 
Release 12.0(5)WC2 or earlier
Catalyst 2950 member switches running Cisco IOS Release 12.0(5)WC2 or earlier
Catalyst 3550 member switches running Cisco IOS Release 12.1(6)EA1 or earlier
In the Front Panel view, if the switch is running one of the software releases listed previously, the 
device LEDs do not appear. In Topology view, if the member is an LRE switch, the CPE devices that 
are connected to the switch do not appear. The Bandwidth and Link graphs also do not appear in 
these views.
Open Caveats
These sections describe the open caveats with possible unexpected activity in this software release:
Open Cisco IOS Caveats
These are the severity 3 Cisco IOS configuration caveats on the Catalyst 2970 and Catalyst 3750 
switches:
CSCeb67510 
When both the sharing and shaping weights are enabled, the receiving rates might not follow the 
shared bandwidth weight if the priority queue is enabled on the egress queue.
The workaround is to use lower values of the shaped and shared weights for queues other than the 
first queue when the egress priority queue is enabled and if shaping in other queues is required.
CSCed04063
When the kerberos clients mandatory global configuration command is entered on a switch and 
the switch is connected to a host that does not support Kerberos through a Telnet session, the switch 
might halt and then fail when the you press the Enter key.
The workaround is to not use the kerberos clients mandatory global configuration command.
CSCed18488
When (*,G) and (S,G) entries are created in a multicast routing table on a remote port by 
Protocol-Independent Multicast-Sparse Mode (PIM-SM) registering, the RPF leak flag is not set for 
hardware entry for the group. This behavior causes high CPU utilization when the CPU receives 
non-RPF traffic in some topologies.
The workaround is to configure the access list for the group to drop non-RPF traffic in hardware.