Apache Cassandra Changelog #10

The latest release of Apache Cassandra is 4.0.1 (pgp, sha256, and sha512). This is a rapid release to fix a critical bug in Gossip on large clusters. Please read the release notes and let us know if you encounter any problems.

Note: As the docs are not yet updated, the bintray location for Debian users is replaced with the ASF’s JFrog Artifactory location.

See the download section for the latest stable and older supported versions of source and binary distributions.

To stay up-to-date, we recommend joining the Cassandra mailing list.

The Project Management Committee (PMC) is pleased to announce that Aleksei Zotov has been invited to become a committer, and he has accepted! Thank you for all your contributions over the years, Aleksei, and congratulations! 👏

After the release of Apache Cassandra 4.0, we have many proposed features (Cassandra Enhancement Proposals or CEPs) under discussion, such as CEP-15: General Purpose Transactions and CEP-17: Storage Attached Index. The current list of CEPs and the process for proposing one are here. If you’re a developer, join the conversation today and help shape the future of Apache Cassandra!

CEP-11 has passed. This proposal aims to enable users to select a memtable implementation suitable for their workload and hardware, including long-lived and persistent memtables. More details are on the wiki.

The vote for CEP-13 has also passed. This is to add a DenyListing partitions feature, enabling operators to disallow the reading of specific partition keys that could cause undesirable effects on the entire node or within the cluster. You can read more about this in the CEP wiki entry.

We have a formal proposal from Brian Houser for CEP-16: Auth Plugin Support for CQLSH. This will ensure that CQLSH has the same support for different authentication methods that we have for other Cassandra drivers. For more details, read the CEP wiki entry.

As always, there are lots of discussions going on. Ekaterina Dimitrova started a good conversation on standardizing configuration and JVM parameters and drove a discussion on JDK 17 support, which is now being developed. We are also defining which code changes target which types of release, deciding when a minor becomes a major, for example, and when feature flags should be used. Josh McKenzie is putting together a wiki on this, and we will run a blog piece on the outcomes from the discussion in the hope it helps other open source projects facing similar questions.

The latest release of Apache Drill 1.19, an open source tool that enables end-users to query multiple data sources using SQL, now includes a Cassandra plugin.

The Kubernetes SIG has decided to move away from bi-weekly meetings as the original reasons for running them have been met. The SIG will be moving to a meetup-style gathering where users will share information and exchange ideas. If you want to share something at the next gathering, please contact Patrick McFadin on the cassandra-kubernetes Slack channel.

“We have modeled customer rewards data in Capital One using Apache Cassandra to serve real-time microservice-based workloads (customers accessing their rewards online) and batch Apache Spark workloads (customer statements) at the same time.” - Gokul Prabagaren

“At Glassdoor, we use Cassandra to store various datasets needed by search, machine learning models, and other systems. Since most of these systems are built on Java, we use Java drivers for Cassandra, for read, write and delete operations.” - Ravi Varsha

Do you have a Cassandra case study to share? Email cassandra@constantia.io.