Andrew J. Stone

I'm not actively working on this project at the moment, but it is in production use. If you give me a patch with a test I'd be happy to test...

@evanmcc Yes, reads are operations that don't mutate state. I agree with you that the client shouldn't have to know about the properties of a given operation. From re-reading what...

I see what's going on. We are talking about 2 different state machines. The one I was referring to that should only rely on opaque operations is the [FSM implementing...

This is really only a problem if all nodes go down. Otherwise a single node in the quiescent cluster state will continuously try to reconnect to all nodes it knows...

This should probably use the root consensus group as a config store.

This almost certainly needs to be implemented before real clients attempt to use the system. Otherwise client behavior will have to change with regard to idempotent retries (i.e. they won't...

I think this should probably just be a [Jepsen](https://jepsen.io/) test.

The only downside would be allocation overhead. However, for these particular messages, they are used infrequently enough that boxing should allow a net win overall. In the future, they may...

Since we are going to likely use a root namespace to allow consistent deletion of namespaces we may as well add this to the diskless 1.0 milestone as well. It's...

Currently the [architecture doc](https://github.com/andrewjstone/rabble/blob/master/doc/architecture.md) for rabble is pretty good. There is also a doc [describing the VR code](https://github.com/vmware/haret/blob/master/haret/src/vr/README.md). There still needs to be an overall picture though that describes the...