OAM mapping is a mechanism that enables a way of deploying OAM end-to-end in a network where different OAM tools are used in different segments. For instance, an Epipe service could span across the network using Ethernet access (CFM used for OAM), pseudowire (T-LDP status signaling used for OAM), and Ethernet access (E-LMI used for OAM). Another example allows an Ipipe service, where one end is Ethernet and the other end is PPP, MLPPP, or HDLC.
In the SRĀ OS implementation, the Service Manager (SMGR) is used as the central point of OAM mapping. It receives and processes the events from different OAM components, then decides the actions to take, including triggering OAM events to remote peers.
Fault propagation for CFM is by default disabled at the MEP level to maintain backward compatibility. When required, it can be explicitly enabled by configuration.
Fault propagation for a MEP can only be enabled when the MA comprises no more than two MEPs (point-to-point).
Fault propagation cannot be enabled for eth-tun control MEPs (MEPs configured under the eth-tun primary and protection paths). However, failure of the eth-tun (meaning both paths fail) is propagated by SMGR because all the SAPs on the eth-tun go down.