Kotlin creator JetBrains has updated its Ktor framework for building Kotlin clients and servers. Version 1.6 sees the tool ending some long-standing issues by adding the option to monitor client progress and support Bearer authentication. The Ktor team also came up with a plugin to ignore trailing slashes during route matching and taught the routing function to offer the overload available in post for put and patch as well.

Icinga for Windows lands with rewritten check handling

The team behind Icinga for Windows has pushed out version 1.5 of the infrastructure monitoring tool. According to the announcement it used the last couple of weeks to rework the way checks are executed in order to improve performance, get rid of some bugs, and make the code easier to maintain and extend.

Other than that the tool is supposed to convert Byte and Time units to be better readable and includes new plugins Invoke-IcingaCheckHTTPStatus and Invoke-IcingaCheckMPIO. While the former lets users check web services directly from Windows machines and add some thresholds and checks, Invoke-IcingaCheckMPIO allows checking for assigned MPIO paths and printing alerts.

Cloudera to turn private before end of year

Big Data company Cloudera has entered a definitive agreement to be acquired by affiliates of investment firms Clayton, Dubilier & Rice and KKR for approximately $5.3 billion. The move goes hand in hand with becoming a private company, which Cloudera president Mick Hollison hopes will make them “even more agile” and help in pursuing “exciting new markets that offer tremendous growth opportunities”.

Speaking of growth, the company also announced the acquisition of data transformation and integration provider Datacoral and cloud data lake firm Cazena. According to Hollison, these will help Cloudera usher in a new era of “self-service” by automating complex operations like migrating to Cloudera’s data platform or connecting to data sources.

Alibaba Cloud makes PolarDB sources publicly available

Cloud provider Alibaba has open sourced its PostgreSQL-based database system PolarDB under the Apache License 2.0. The project is said to extend PostgreSQL into an easy to scale, share-nothing database, offering high availability, data consistency, and ACID compliance amongst other things. Each node of the system stores data and executes queries, while coordination is realised via message passing. More open source projects are planned to follow in the near future.

Databricks introduces protocol to simplify data sharing

Since sharing data between different parties can be quite tricky, data experts Databricks have come up with a REST protocol called Delta Sharing to share access to parts of a dataset stored in the cloud. The approach makes use of the company’s Delta Lake and the Apache Parquet format to let data providers share what they want of a table.

Access is then managed through a sharing server running between data source and recipient which knows the protocol. The company already released connectors for pandas, Apache Spark, Rust and Python so that data users can connect directly to the shared data through familiar tools instead of having to go through export procedures first. More connectors (for open source as well as commercial clients) are promised to be in the works.