High Availability
244 Basic System Configuration Guide
Nonstop forwarding is used to notify peer routers to continue forwarding and receiving
packets, even if the route processor (control plane) is not working or is in a switch-over state.
Nonstop forwarding requires clean control plane and data plane separation and usually the
forwarding information is distributed to the line cards. This method of availability has both
advantages and disadvantages. Nonstop forwarding continues to forward packets using the
existing stale forwarding information during a failure. This may cause routing loops and
black holes, and also requires that surrounding routers adhere to separate extension standards
for each protocol. Every router vendor must support protocol extensions for interoperability.
Nonstop Routing (NSR)
With NSR on the SR-series router devices, routing neighbors are unaware of a routing
process fault. If a fault occurs, a reliable and deterministic activity switch to the inactive
control complex occurs such that routing topology and reachability are not affected, even in
the presence of routing updates. NSR achieves high availability through parallelization by
maintaining up to date routing state information, at all times, on the standby route processor.
This capability is achieved independently of protocols or protocol extensions, providing a
more robust solution than graceful restart protocols between network routers.
The NSR implementation on the SR-series routers supports all routing protocols. NSR makes
it possible to keep the existing sessions (BGP, LDP, OSPF, etc.) during a CPM switchover,
including support for MPLS signaling protocols. Peers will not see any change.
Protocol extensions are not required. There are no interoperability issues and there is no need
to define protocol extensions for every protocol. Unlike nonstop forwarding and graceful
restart, the forwarding information in NSR is always up to date, which eliminates possible
blackholes or forwarding loops.
Traditionally, addressing high availability issues have been patched through non-stop
forwarding solutions. With the implementation of NSR, these limitations are overcome by
delivering an intelligent hitless failover solution. This enables a carrier-class foundation for
transparent networks, required to support business IP services backed by stringent SLAs. This
level of high availability poses a major issue for conventional routers whose architectural
design limits or prevents them from implementing NSR.
CPM Switchover
During a switchover, system control and routing protocol execution are transferred from the
active to the standby CPM.
An automatic switchover may occur under the following conditions:
To Next Page
Loading...
