Analysis

The status updates, which are acquired from IP Availability Manager, from periodic SNMP monitoring, from periodic remote pings, and (optional) from Network Protocol Manager for BGP serve as input to the data model in the MPLS Manager repository. MPLS Manager applies the status updates to the appropriate attributes of the corresponding objects in the data model.

The data model involves a codebook that is created by VMware engineers as a result of their knowledge of IP and MPLS networks. The codebook identifies IP and MPLS problems and their symptoms, and presents a causality mapping between each problem and its set of symptoms.

MPLS Manager diagnoses root-cause availability problems by finding the problems in the codebook that can best explain the observed symptoms. It also uses the codebook to correlate the impact of a root-cause problem on the MPLS and VPN objects that are either inaccessible or impaired as a result of the root-cause problem. The root-cause problem is either a physical-transport problem that is diagnosed by IP Availability Manager, or an MPLS problem that is diagnosed by MPLS Manager.

MPLS Manager performs root-cause analysis on alarms that occur at different layers of the MPLS network domain in order to pinpoint the root cause. It also detects configuration errors that occur when deploying and maintaining MPLS, VPN, and BGP networks.