We are excited to announce the second preview release of EventStoreDB 6.0.0.
This release addresses some of the feedback from the first preview.
This release is not intended to be used in production and is still rough around the edges, but we welcome your feedback as we prepare a release candidate.
If you encounter any issues, please don’t hesitate to open an issue on GitHub if there isn’t one already.
You can download the packages from the downloads page under the Pre-Release section.
If you are running on macOS, you need to run EventStoreDB in Docker, since currently servers using .NET Core’s gRPC implementation require the platform to support server ALPN, which macOS did not until Catalina. As soon as this restriction is lifted by the .NET Core platform, Event Store will release packages for macOS.
With the release of preview two of EventStoreDB, we are also shipping a new gRPC based operations client which as it currently stands has support for scavenging.
We have renamed the .NET gRPC Client to EventStore.Client.Grpc. In preview one, it was named EventStore.Grpc.Client.
The package is available on NuGet. The NuGet package from preview one will not receive updates and is deprecated.
To simplify the gRPC API, we have exposed the read direction in the gRPC Client as a parameter.
ReadAllForwardAsync and ReadAllBackwardAsync are now collapsed into ReadAllAsync with the first parameter being the direction of the read to be performed.
eventStoreClient.ReadAllAsync(
direction: Direction.Forwards,
position: Position.Start,
maxCount: 4096,
resolveLinkTos: true,
filter: null,
userCredentials: new EventStore.Client.UserCredentials("admin", "changeit"));
ReadStreamForwardAsync and ReadStreamBackwardAsync are now collapsed into ReadStreamAsync with the first parameter being the direction of the read to be performed.
eventStoreClient.ReadStreamAsync(
direction: Direction.Forwards,
streamName: "foo",
revision: StreamRevision.Start, count: 4096,
resolveLinkTos: true,
userCredentials: new EventStore.Client.UserCredentials("admin", "changeit"));
Note: This is a breaking change from preview one.
Improvements have been made to centralize replication logic for each type of operation. This improves performance and correctness.
We have introduced a health check which can be queried via the {server_address}/health/live endpoint. This endpoint will return a 204 status code once EventStoreDB is ready, and is now configured as the default health check in the Docker container.
We have included the retry count for persistent subscription’s HTTP based API.
We have addressed an issue where particular fields (notably $type) being present in the payload of a client operation would cause a re-serialization during response, dropping the field.
When experiencing commit timeouts, projections would attempt to write five times with a few seconds between each retry before they are marked as faulted. After this, manual intervention was required to re-enable the projections.
To alleviate the effects of commit timeouts during writes from projections, a system of exponential backoff with jitter has been added. This should prevent projection retries from overloading the server with write retries.
The retry count for projections has also been increased, as the previous number of five retries only corresponded with about 10 seconds.
In our ongoing effort to improve projections we have made a number of changes.
Some projection selectors and options were removed:
fromStreamCatalog selectorfromStreamsMatching selectordisableParallelism optioncatalogTransform optionThese selectors were only usable in queries or transient projections. If you were using them, you need to recreate any affected queries or transient projections.
As only fromStreamsMatching was officially documented, it was the only one of these officially supported.
Usage of fromStreamsMatching can be replaced with fromAll with an appropriate where modifier.
Following the release of Read-Only Replicas, Clones have been disabled by default.
The new default behaviour is for nodes in excess of the configured cluster size to be terminated upon joining.
If you need to make use of Clones nodes while transitioning to using read replicas, the old behaviour can be restored by setting the --unsafe-allow-surplus-nodes option on all nodes in the cluster.
Note: this does not affect users of EventStoreDB, only those implementing client SDKs. If you are developing a gRPC-based client, please update your proto contracts.
We have received some great feedback from the community around gRPC, more notably community members who have started building gRPC clients for a variety of platforms, and have adjusted the protocol buffers API definitions in response to that:
Windows-centric performance counters do not work with .NET Core, so many statistics are now collected using the Microsoft.Diagnostics.NETCore.Client package.
Note: Currently system-wide CPU and memory usage are not exposed via the statistics endpoint on Linux or macOS. This will be rectified in the release candidate.