
363
Cross-Platform Release Notes for Cisco IOS Release 12.0S
OL-1617-14 Rev. Q0
Caveats
Resolved Caveats—Cisco IOS Release 12.0(33)S10
• CSCtu19667
Symptoms: A router reports an lbl configuration sync failure similar to:
%REDUNDANCY-3-CONFIG_SYNC: Active and Standby lbl configuration out of sync
Conditions: This symptom is seen when “ip host” commands are modified.
Workaround: The errors are cosmetic and do not accurately reflect the state of the configuration
sync.
• CSCtw53776
Symptoms: A Cisco 12000 linecard crashes when NetFlow version 1 is configured on the router.
Conditions: The crash occurs if the “ip flow export x.x.x.x <port>” command is used instead of the
“ip flow export destination x.x.x.x <port>” command.
Workaround: Configure NetFlow version 5 or version 9.
• CSCtw61050
Symptoms: A ping with a large packet size fails.
Conditions: When there are more than seven buffer pools.
Workaround: Configure a carve-level default or reduce the set of MTUs so that none of the pools is
reduced.
• CSCtx19461
Symptoms: If the BGP IPv6 table has a ::/X prefix present that is learned from another 6PE router,
we see the following issues.
1. The 6PE router stops advertising the ::/0 prefix. Instead, it advertises the ::/X prefix that is present
in the BGP table.
2. The 6PE router does not withdraw the ::/0 prefix.
3. Output from the “show /jointfilesconvert/415844/bgp ipv6 unicast” command does not show prefix ::/0. Instead, it shows
prefix ::/X.
Conditions: If the BGP IPv6 table has a ::/X prefix present that is learned from another 6PE router
and you are advertising ::/0.
Workaround: There is no workaround.
• CSCtx36490
Symptoms: There is a stuck stale nexthop 0.0.0.0 for routes that are redistributed into EIGRP from
BGP. This stale nexthop gets promoted to the top of the EIGRP topology table upon a route/router
flap. This will result in a redistribution loop in the RIB.
Conditions: The redistribution loop in the RIB causes the route to be continuously installed in the
RIB from BGP and EIGRP, resulting in a high number of CEF updates. This high number of CEF
updates results in the RP running out of memory due to CSCtq97113.
Workaround: Clear out the stale nexthop 0.0.0.0 from the EIGRP topology table by issuing the “clear
ip eigrp [vrf <vrf-name>] <AS-number> topology <prefix> <mask>” command.
In a PE-CE environment, if there is a peering between two PEs that are learning the prefix from the
same CE site, the stale nexthop 0.0.0.0 could get promoted in the EIGRP topology table upon a route
flap. This could, in turn, trigger the looping. To avoid this, you may remove any such redundant
peering.
A detailed explanation of this workaround is provided in the enclosure “explanation of
workaround”.
Kommentare zu diesen Handbüchern