This is an instructional implementation of the Raft consensus algorithm.

It is not intended for production use.

A grpc server is used for inter-node communication. We use a health-server with an observer pattern to monitor the health of the nodes and advertise the leader. By using a serviceConfig, we can use the health server to find the leader and send requests to it. We are hence complying with the Raft specification that clients should only send requests to the leader (linearizable reads and writes).

We provide a client api that can be used to send requests to the leader. See the raft-test.go file for an example of how to use the client. We provide our own grpc resolver to find the leader in suite_test.go.

Run the tests:

$ make tests

Update protobuf definitions:

$ make proto-raft

Update protobuf raft client definitions:

$ make proto-raft-client