I am trying to upgrade it to tokio 1. The current work is https://github.com/ritedb/riteraft/tree/tokio-1 .

It seems that it has passed the compilation, but it cannot work with actix-web for the time being. Even if use actix-web 4 that has not been officially released.

I saw the snapshot interface is very simple now, only copy the mem data to the followers' node.
Anyone knows how to implement a snapshot for the persistent database which is more suitable for the real-world use case.

Hi all, I use this library in my project, and find it will crash after 255th commit. I also test the example code, it also crashes after subunit around 255 puts.
Here is one log from my project, the crash occurs when the mailbox sends a message and receives an error result.

log
node2_1  | Jul 07 15:16:44.508 INFO [commit: 255] ignored snapshot [index: 245, term: 2], snapshot_term: 2, snapshot_index: 245, commit: 255, raft_id: 3, version: 1.0.0
node3_1  | Jul 07 15:16:44.508 INFO [commit: 255] ignored snapshot [index: 245, term: 2], snapshot_term: 2, snapshot_index: 245, commit: 255, raft_id: 2, version: 1.0.0

node1_1  | thread 'tokio-runtime-worker' panicked at 'called `Result::unwrap()` on an `Err` value: Unknown', src/main.rs:217:55

Crash code line82 on example
let result = data.0.send(message).await.unwrap();

https://github.com/async-raft/async-raft is An implementation of the Raft distributed consensus protocol using the Tokio framework.

Need to wait for the upstream tikv/raft-rs to release 0.6.0 .

Related: tikv/raft-rs#429

I'm using the memstore example and starting 3 nodes like so:

./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5001 --web-server=0.0.0.0:8001
./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5002 --web-server=0.0.0.0:8002 --peer-addr=0.0.0.0:5001
./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5003 --web-server=0.0.0.0:8003 --peer-addr=0.0.0.0:5001

When I kill the first process (5001), the last two don't seem to be figuring out a new leader. Elections seem to be triggering over and over, neither (5002, 5003) wins the election, but they do seem to be requesting votes from each other.

Am I passing the wrong arguments ?

Thanks for the help

Hi,

Thank you for building such a wonderful crate. It is very interesting.
However, I'm trying to run the example code, but the third node doesn't seem to join the cluster correctly, is there a problem with my command execution?

$ ./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5001 --web-server=0.0.0.0:8001
$ ./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5002 --web-server=0.0.0.0:8002 --peer-addr=0.0.0.0:5001
$ ./target/debug/riteraft-memstore --raft-addr=0.0.0.0:5003 --web-server=0.0.0.0:8003 --peer-addr=0.0.0.0:5001

$ curl -XGET http://localhost:8001/put/1/AAA

$ curl -XGET http://localhost:8001/get/1
Some("AAA")

$ curl -XGET http://localhost:8002/get/1
Some("AAA")

$ curl -XGET http://localhost:8003/get/1
None

How can I solve this problem?

Thanks,