[c-nsp] 12.2(33)SXI revisited

[A.L.M.Buxey at lboro.ac.uk]

hi,

okay - after fruitful time on the test/devel router and then
a prolonged period on 2 select routers, we're now SXI across
the board.... which has finally thrown up an error (of sorts!)
hurrah! :-)

having exhausted my usual supply of cisco and google (and c-nsp ;-) )
searches I thought I'd throw this one to the list before
visiting my TAC friends for the 3rd time this week.

note: so far we only obsrve this error on systems with 2 Sup720's in
SSO redundancy mode

this is whats being logged:

Feb 10 12:26:38: SP: Critical error processing oif ltl index change event (0100.5b00.073c, XXX, 0xB8B)
Feb 10 12:30:32: SP: Critical error processing rpf ltl index change event (0100.5b1f.fefd, XXX, 0xBAB)
Feb 10 12:30:32: SP: Critical error processing oif ltl index change event (0100.5b1f.fefd, XXX, 0xBAB)
Feb 10 12:30:48: SP: Critical error processing rpf ltl index change event (0100.5b1f.fefa, XXX, 0xBB3)
Feb 10 12:30:48: SP: Critical error processing oif ltl index change event (0100.5b1f.fefa, XXX, 0xBB3)
Feb 10 12:31:54: SP: Critical error processing rpf ltl index change event (0100.5b07.0076, XXX, 0xBBF)

hmm. very nice . XXX represents a VLAN number. if done simple obfuscation
of the MAC and this info just because of policy - however, those MACs *do* 
start with 0100 which is multicast L2 ....so currently I've got a feeling
that this is a problem with TTL=1 (and being dropped - eg ghost or rubbish
multicast app) and its hitting the RPF MLS  - what say you all?

alan