This topic describes the release notes for fully managed Flink and provides links to relevant references. The release notes provide the major updates and bug fixes in fully managed Flink in the version that was released on February 20, 2023.
Overview
A new official version of fully managed Flink is released on February 20, 2023. This version is developed based on Ververica Runtime (VVR) 6.0.5 of Apache Flink 1.15.3. All defects in Apache Flink 1.15.3 are fixed in this version of fully managed Flink. In this version, the StarRocks connector is officially launched to facilitate data reading from source tables and data writing to result tables in StarRocks databases.
Features
Feature | Description | References |
Data type mappings in real-time data ingestion into data lakes and data warehouses by using the Flink Change Data Capture (CDC) connector | When you use the CREATE DATABASE AS or CREATE TABLE AS statement to ingest data into a data lake or a data warehouse, the data types of columns in the tables of upstream and downstream data stores can be different if the data types are compatible. This helps prevent unnecessary errors. In this scenario, the downstream data store can only be Hologres. | |
Multiple matching of a single element in a group pattern or the NO SKIP pattern supported by Flink SQL complex event processing (CEP) | Flink SQL CEP supports group patterns that are similar to A(B C)+ and the NO SKIP pattern that is similar to A B{4,10}?. | |
Configuration of buckets and log consumption from the checkpoints that are stored in a specified consumer group by using the Log Service connector | The Log Service connector allows you to specify the number of buckets for Logstores that contain more than 64 shards. The connector can also consume logs from the checkpoints that are stored in a specified consumer group. | |
StarRocks connector | The StarRocks connector can be used for result tables and source tables. | N/A |
Addition of metadata, such as op_ts, to the synchronization process during the execution of the CREATE TABLE AS statement | When the CREATE TABLE AS statement is executed, the metadata, such as op_ts, can be synchronized. |
Fixed issues
The following issue is fixed: After you modify the parameters in the WITH clause of the Message Queue for Apache Kafka connector in a deployment, the deployment may fail to start.
The following issue is fixed: When Flink CDC uses a Java Database Connectivity (JDBC) driver to read data from a database in an asynchronous thread, the cause for an out of memory (OOM) error cannot be displayed.
The issue that the error message
Table program cannot be compiled.appears due to an error in Codegen is fixed.