Datasheet

246
Release Notes for Cisco IOS Release 12.1E on the Catalyst 6500 and Cisco 7600 Supervisor Engine and MSFC
OL-2310-11
Caveats
• In a configuration that contains many route maps, if you ping from a router to a VLAN interface
messages like the following may be displayed and the TCAM screening status may be displayed as
INACTIVE:
00:03:41: :FM-ODM: Maximum Number of entries exceeded, ODM gives up!
odm_intra_feature_merge:odm_merge fail with code=1
00:03:41: %FM-2-TCAM_MEMORY: Interface Vlan711 processor memory low programming
ingress ACLs
This problem is resolved in Release 12.1(13)E3. (CSCdz16693)
• A unique engine ID needs to be added to NDE version 5 export for traffic from the PFC2 to
distinguish it from traffic from the MSFC2. This problem is resolved in Release 12.1(13)E3.
(CSCdz27636)
• A system with a WS-X6K-SUP2-2GE and WS-F6K-MSFC2 that is running Release 12.1(13)E may
reset when connecting a server to a WS-X6548-RJ-45. This problem is resolved in
Release 12.1(13)E3. (CSCdz29085)
• If you use a script to remotely login to a Catalyst 6500 series switch or a Cisco 7600 series router,
the system may experience a memory leak. This problem is resolved in Release 12.1(13)E3.
(CSCdz30206)
• A system with a Supervisor Engine 1 and an MSFC2 does not allow the flow mask to be set to full
flow. This problem is resolved in Release 12.1(13)E3. (CSCdz30894)
• When you configure the IOS server load balancing feature called RADIUS load balancing on
multiple virtual servers with the sticky radius framed-ip command, some RADIUS framed-ip
sticky database entries may be deleted prematurely, and stateful backup may not operate correctly.
The problem occurs for framed-ip sticky database entries that are created as a result of processing
RADIUS packets associated with any virtual server that is not the first virtual server brought into
service with framed-ip sticky configured. This problem is resolved in Release 12.1(13)E3.
(CSCdz34575)
• Entering a client virtual server Cisco IOS SLB configuration command hangs the system. This
problem is resolved in Release 12.1(13)E3. (CSCdz37802)
• A Cisco 7603 running Release 1.1(11b)E1 reloads when you configure Cisco IOS SLB. This
problem is resolved in Release 12.1(13)E3. (CSCdz51826)
• When adding 1000 VLANs, VTP calculates the download buffer size and if the size is over 64k, a
memory corruption might occur causing the system to reset. This problem is resolved in
Release 12.1(13)E3. (CSCdz54132)
• When a ping for a multicast group reaches a Supervisor Engine 1 system that has the ip igmp
join-group command configured on the interface, a (S,G) entry is created and a shortcut is
downloaded. For subsequent pings, duplicate replies are sent because the Supervisor Engine 1
forwards two packets to the MSFC, which then sends out two Unicast replies to the source. This
problem is resolved in Release 12.1(13)E3. (CSCin24629)
• If you access the ifEntry MIB through SNMP and at the same time delete VLANs through the CLI,
the system may reset. This problem has been fixed in and after Release 12.1(13)E3. (CSCdx89085)
• The following message may display:
%LINK-SP-2-NOSOURCE: Source idb not set-Process= "<interrupt level>", ipl= 1,
pid= 78 -Traceback=XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX
This problem is resolved in Release 12.1(13)E3. (CSCdy18548)
• When a multicast host sends a leave message, the multicast stream is not immediately terminated if
a general query has just taken place. This problem is resolved in Release 12.1(13)E3. (CSCdy56062)