Today, etcdctl endpoint health only performs a quorum GET. If server is corrupted or exceeded quota (which put the server into read only mode), this command incorrectly returns healthy.

Hello Philip!
I think there is an issue with this part of the code of rqlite (store/store.go).

func (s *Store) Database(leader bool) ([]byte, error) {
	if leader && s.raft.State() != raft.Leader {
		return nil, ErrNotLeader
	}
	// Ensure only one snapshot can take place at once, and block all queries.
	s.mu.Lock()
	defer s.mu.Unlock()

	f, err := ioutil.TempFile("", "rqlilte-snap-

Quorum doesn't return a message error clearly specifying the cause of an error (from a user standpoint) when the transaction manager fails to Quorum (e.g: 400 bad request, ...).

System information

Release: 2.2.3

Expected behaviour

Helpful error message

Actual behaviour

invalid argument 0: json: cannot unmarshal non-string into Go value of type common.Hash

The blockchain v2 reactor utilizes concurrency to saturate the bottleneck of writing blocks to disk. This concurrency is internal to the reactor where the reactor itself will launch and manage internal state machines running as go-routines. This configuration makes testing difficult as we don't know when messages processed by internal state machines will be processed and when we can assert that th

Description

Now it is in the changelog and release note.

Versions

develop

Additional Information

-----------Chinese-------
把各版本的升级提示移到单独的一个文件夹

Here is a rough idea of what this test should do:

• It should be quite similar to the other tests which are currently in place.
• It should isolate a follower node, write a data to the cluster which will cause that node to fall behind into lagging state, but not into snapshotting state.
• It should then spawn a task which will continue to write data to the cluster.
• Just after it is spawne

Description

Many of the tests in https://github.com/spacemeshos/go-spacemesh/blob/develop/cmd/node/node_test.go use goroutines to run and interact with a process concurrently. They have require statements inside these goroutines, which is a no-no because when a require statement fails inside a goroutine it does not stop the test, in some cases causing tests to hang indefinitely. Here's one s

Instead of a single validator (a.k.a. pivot node), the test net_dynamic_hb should pick a subset of validators at random for removal. This will better reflect the [interface](https://github.com/poanetwork/hbbft/blob/742ad7b83a3f613fd9a7b4ff88cfa82b10bd7dd3/src/dynamic_honey_badger/dyn