mkuklik/raft

Raft implementation

Raft implementation

What is the better way to understand Raft Consensus Algorithm than implementing it from scratch.

TODO

• test ApplyCommittedEntries
• test AddCommand to 100% coverage
• testing follower AppendEntry w/ 100% coverage
• testing follower RequestVoteFollower w/ 100% coverage
• testing leader w/ 100% coverage
• refactor states directly to RaftNode
• Snapshots, using the Chandy-Lamport or another algorithm
• Membership Changes (§6 of the Raft paper and §4 of the dissertation)
• Log Compaction (§7 of the Raft paper and §5 of the dissertation)
• Web application to control and interface with a Raft node (using gRPC)
• compare to hashicorp & etcd implementation
• compare my implementationt to this blog post: https://eli.thegreenplace.net/2020/implementing-raft-part-0-introduction/

raft implementations in production

• https://github.com/hashicorp/raft
• https://github.com/etcd-io/etcd/blob/master/raft/README.md

Sources

• https://github.com/etcd-io/etcd/blob/master/raft/README.md

• https://ieftimov.com/post/understanding-bytes-golang-build-tcp-protocol/

• https://opensource.com/article/18/5/building-concurrent-tcp-server-go

• https://callistaenterprise.se/blogg/teknik/2019/10/05/go-worker-cancellation/

• https://rcrowley.org/articles/golang-graceful-stop.html , https://gist.github.com/rcrowley/5474430

• ? https://www.rodrigoaraujo.me/posts/golang-pattern-graceful-shutdown-of-concurrent-events/

• ? https://github.com/stephenafamo/orchestra