Handling failure of network function instances in a network function set

Abstract

The present disclosure relates to systems, methods, and computer-readable media for recovering from a failure of a network function instance from a network function set within a telecommunications network. The systems described herein involve a peer network function that obtains profile information including a hashing algorithm associated with handling a failure condition. The peer network function additionally maintains a session identifier corresponding to a session between the peer network function and the network function instance. In response to the network function instance failing, the peer network function can implement the hashing algorithm to determine which of a plurality of additional network function instances from the network function set should be used to re-establish the session. The hashing algorithm includes attributes that ensure consistent selection between peer network functions as well as balancing network traffic between network function instances of the network function set.