NSP periodically initiates a set of reachability checks on managed NEs based on the reachability policies.

The reachability status of a managed device indicates the results of the last set of reachability checks.

• If the NE responds to all reachability checks, it is reachable from the NSP system.

• If some checks fail and others pass, the NE is partially reachable.

  The reachability policy raises an alarm when its check fails, for example, PingConnectionProblem.

• If all checks fail, the NE is unreachable.

  If the NE becomes unreachable, a ReachabilityProblem alarm is raised.

For model-driven NEs, the checks performed are based on the reachability policy types:

• A ping reachability check is an ICMP ping to the NE.

• For other protocols, such as NETCONF or GRPC, a socket ping is performed. NSP confirms that the NE is reachable on the target port for the protocol.

  Note: The socket ping confirms availability of the port. It does not verify that mediation is configured. If the NE is reachable but a connection cannot be established, check the mediation policies and/or certificates, as applicable.

See the artifact guide for the NE family for the list of protocols to use for reachability checks, that is, for the list of reachability policies you need to create.

For classic NEs reachability is determined by the ping result of the active management IP address. This can be in the in-band or out-of-band IP address; NSP uses the associated reachability policy to perform reachability checks.