Node Protection Although traffic protection techniques may be used to route around
failed links or nodes in the core of the network, they are generally unable to protect
against failure of the ingress or egress LSRs, since customers attach directly to these.
It is also, of course, desirable that core LSRs be as resilient as possible in order to minimise
the protection switching required to bypass failed nodes.
Modern routers, and LSRs, are designed to have no single point of failure??”power
supplies, switching fabrics, and control cards are duplicated and network interfaces
are distributed across multiple independent line cards. However, in order to take
advantage of this hardware redundancy, additional software support is required. With
MPLS 431
respect to the routing and signalling protocols used to establish LSPs, there are two
basic techniques that may be used:
?– Nonstop routing In order for routing and signalling protocols to continue to
operate when a control card switchover occurs (due to failure of the active control
card), it is necessary to replicate all state changes resulting from the operation of
those protocols from the active to the backup control card.
Pages:
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971