EventStoreDB 5.0.8

Version 5.0.8 has been released which contains a number changes and bug fixes.

EventStoreDB 5.0.8 is out!

This release contains some important bug fixes related to TCP connections for both the server and the .NET client. The fixes bring an overall improvement to the stability of subscriptions.

It is available for the following operating systems:

• Windows
• Ubuntu 18.04
• Ubuntu 16.04
• CentOS 7 (Commercial version)
• Amazon Linux AMI (Commercial version)
• Amazon Linux 2 AMI (Commercial version)
• Oracle Linux 7 (Commercial version)

Highlights

Core

• When a long running node in a cluster has observed a large number of elections, the SendViewChangeProof that is scheduled could cause cluster instability.

Projections

• Always load up partition state which could previously have caused stale state’s to be observed.

Client

• Don’t auto resubscribe TCP catchup subscriptions which could potentially cause duplicate events to be observed. It is now expected that users will resubscribe their subscription in SubscriptionDropped to recover from TCP connection losses.
• Introduce the ability to clients to gossip over TLS when using gossip seeds.
• Provide the ability for the Event Store 5.x TCP Client to handle the old and new cluster terminology. The terminology changed from Slave to Follower and from Master to Leader.

Please consult the Changelog for more details.

Where can I get the packages?

Downloads are available on our [website]/downloads/).

The packages can be installed using the following instructions.

Ubuntu 16.04/18.04 (via packagecloud)

curl -s https://packagecloud.io/install/repositories/EventStore/EventStore-OSS/script.deb.sh | sudo bash
sudo apt-get install eventstore-oss=5.0.8-1

Windows (via Chocolatey)

choco install eventstore-oss -version 5.0.8

Client Packages
EventStore Client

dotnet add package EventStore.Client --version 5.0.8

EventStore Embedded Client

dotnet add package EventStore.Client.Embedded --version 5.0.8

Upgrade Procedure

To upgrade a cluster, a usual rolling upgrade can be done:

• Pick a node (Start with slave nodes first, then choose the master last)
• Stop the node, upgrade it and start it

Event Store 5.0.8 Changelog

Bug Fixes

• #2367 - (Client) Don’t auto resubscribe TCP catchup subscriptions.
• #1959 - (Client) Introduce the ability to clients to gossip over TLS.
• #2376 - (Client) Handle old and new cluster terminology for node NodePreference.
• #2366 - (Core) Only send view change proof on system init.
• #2345 - (Projections) Always load up partition state.

How do I provide feedback?

We appreciate any feedback via either GitHub Issues or forum.