Arrangement and a method for handling failures in a network

Abstract

The present invention relates to an arrangement comprising an Ethernet carrier network, managed by a management system () and comprising a number of levels of management domains, each comprising a number of edge nodes () and a number of intermediate nodes (). Each edge node () comprises fault detection of connectivity check messages and fault locating means with holding means () adapted to hold path information for the paths to all other edge nodes. A fault detection means of an edge node having detected a fault is adapted to activate edge node fault locating means to locate the fault using the patch information in the edge node holding means to find a first target intermediate node substantially in the middle of the path containing the fault, and to, directly, send a first loop back message to said intermediate node. A response from said intermediate node indicates that the fault is located at the distant portion of the path, no response indicates that the fault is located at the proximate portion of the path. Further loop back messages are sent to consecutively identified intermediate nodes of the respective path portion identified to contain the fault by the respective preceding loop back message until the fault is localized.