GridGain 9.0.17 Release Notes

GridGain 9.0.17 simplifies cluster initialization and adds a multicast node finder.

Overview

GridGain 9.0.17 is a stability release that is dedicated to improving user experience and fixing usability issues.

Major Changes

Cluster Initialization Changes

This release features major changes in the cluster init command.

• The new --license parameter should now be used to provide the license file.

• You no longer need to specify cluster metastorage or management group nodes. If not specified, nodes will be selected automatically according to the following logic:

  • If the cluster size is 3 or fewer, all nodes are used for the metastorage group (CMG and metastorage).

  • If the cluster size is 4, three nodes are used to maintain an odd number, improving consensus efficiency.

  • If the cluster size is 5 or more, five nodes are used to balance fault tolerance and overhead.

New Features

Multicast Node Finder

This release provides a way to enable optional multicast node finder. When enabled, GridGain uses multicast to discover other nodes. By default, static node finder is used.

To start using multicast node finder:

• Modify the multicast configuration. The example below shows default multicast configuration:

  ignite {
    network {
      nodeFinder {
        multicast {
          group="239.192.0.0"
          port=47401
          resultWaitTime=500
          ttl=-1
        }
      }
    }
  }

  You can modify it with the node config update command, for example:

  node config update ignite.network.nodeFinder.multicast.group=239.5.0.0

  Multicast address must be valid. You can learn more in IANA guidelines for multicast.

• Enable multicast finder:

  node config update ignite.network.nodeFinder.type=MULTICAST

Now, this node will be using multicast to find other nodes in the specified multicast group.

Improvements and Fixed Issues

Issue ID	Category	Description
IGN-27155	General	Fixed a rare NPE that could happen when querying the LOCKS system view.
IGN-27154	General	Improved output formatting for metrics exported to log.
IGN-27146	General	The oomBufferSizeBites configuration was renamed to oomBufferSizeBytes.
IGN-27136	General	Fixed an issue that could cause repeated warnings on long-running clusters.
IGN-27133	General	Leases updated message is now correctly printed every 10 updates.
IGN-27111	SQL	STATE column in LOCAL_PARTITION_STATES system view was renamed to PARTITION_STATE.
IGN-27103	General	Fixed a possible exception during data recovery with no data on node.
IGN-27023	Platforms and Clients	Python DB API: Added support for macOS.
IGN-27004	General	Fixed an exception that could happen when primary data replica left the cluster during transaction.
IGN-26983	Platforms and Clients	.NET: Fixed error code string representation.
IGN-26813	General	Added Tuple support to data streamer with receiver as payload, argument, result.
IGN-26773	SQL	BINARY data type is no longer supported.
IGN-26742	Cluster Discovery	Added support for multicast node finder.
IGN-26631	SQL	Fixed an issue that cause incorrect index column sorting when using catalog API.
IGN-26374	General	Metastorage group and cluster name are no longer required to initialize the cluster.
IGN-26273	General	Node names in the cluster must now be unique.
GG-42591	General	Improved Docker image size.
GG-42493	Cluster SQL Engine	TIMESTAMP data type is no longer supported for data expiry. Use TIMESTAMP WITH LOCAL TIME ZONE instead.
GG-42372	Platforms & Clients	.NET: Added GridGain.Extensions.Caching.Ignite package.
GG-42269	Licenses	New --license flag was added to cluster init command.
GG-41951	Cluster Metrics & Monitoring	Fixed NotCompliantMBeanException that could happen after enabling JMX exporter.
GG-41467	GridGain Integrations	Kafka source: added schema evolution support.

Known Limitations

Data Restoration After Data Rebalance

Currently, data rebalance may cause partition distribution to change and cause issues with snapshots and data recovery. In particular:

• It is currently not possible to restore a LOCAL snapshot if data rebalance happened after snapshot creation. This will be addressed in one of the upcoming releases.

• It is currently not possible to perform point-in-time recovery if data rebalance happened after table creation. This will be addressed in one of the upcoming releases.

Data Center Replication with Multiple Data Centers

Complex Data Center Replication topologies (for example, the ones involving cycles) of 3 or more data centers are not supported. This will be addressed in an upcoming releases.

GridGain 8 Features

The following features of GridGain 8 are not available in this version, and will be added in upcoming versions:

• Rack-Awareness

• Tracing

• Service Grid

SQL Performance in Complex Scenarios

There are known issues with the performance of SQL read-write transactions in complex read-write scenarios. These issues will be addressed in an upcoming releases.

Installation and Upgrade Information

System View Changes in GridGain 9.0.16

In GridGain 9.0.17, we continue to improve system view naming for more clarity and consistency across all system views. In this release, the column in the LOCAL_PARTITION_STATES view has been renamed from STATE to PARTITION_STATE.

The legacy column remains available for backward compatibility but is deprecated and will be removed in a future release.

System View	Old Name	New Name
LOCAL_PARTITION_STATES	STATE	PARTITION_STATE

We Value Your Feedback

Your comments and suggestions are always welcome. You can reach us here: http://support.gridgain.com/.