All Products
Search

This Product

    Realtime Compute for Apache Flink:December 26, 2022

    Document Center

    Realtime Compute for Apache Flink:December 26, 2022

    Last Updated:Jan 31, 2023

    This topic describes the release notes for fully managed Flink and provides links to the relevant references. The release notes provide the major updates and bug fixes in fully managed Flink in the version that was released on December 26, 2022.

    Overview

    A new version of fully managed Flink is officially released on December 26, 2022. This version is developed based on Ververica Runtime (VVR) 6.0.4 of Apache Flink 1.15.2. In this version, multiple features of the Hologres connector and MySQL Change Data Capture (CDC) connector are supported, and the capabilities and usability of the dynamic Flink complex event processing (CEP) feature are optimized.

    Note

    If your deployment uses VVR 6.X, we recommend that you upgrade the VVR version to VVR 6.0.4 at the earliest opportunity to ensure the stability of your deployment.

    Features

    • The MySQL CDC connector and the Postgres SQL CDC connector support all features provided by Flink CDC V2.3.

    • The MySQL CDC connector allows you to specify the consumer offset of binary log data.

    • The Hologres connector that is in binlog mode supports the TIMESTAMP data type.

    • Hologres catalogs support the ALTER TABLE statement. This statement is used to modify the properties of a Hologres table.

    • The connector for a Hologres source table supports filter pushdown. This way, the connector does not read full data from a Hologres source table. This optimizes the read performance in scenarios where a large amount of data is read.

    • Dynamic Flink CEP allows you to use the WITHIN clause to specify the time interval between events.

    • Dynamic Flink CEP allows you to specify a condition in the JSON-formatted rule description to pass parameters.

    • Specific features of dynamic Flink CEP are optimized. The error message that appears when an unsupported condition is specified is optimized. If a pattern that is parsed based on an aviator condition in a deployment is invalid, the error is ignored and the deployment does not fail.

    • The MATCH_RECOGNIZE clause of the static CEP is enhanced. Multiple clauses, such as SHOW TIMEOUT MATCHES, Not Next, Consecutive, Until, Followed By, Followed By Any, and Not Followed By, are supported.

    • The connector for a Message Queue for Apache Kafka source table can be used to check whether the consumer group specified in a deployment is the same as an existing consumer group.

    • The Log Service DataStream connector is supported. For more information, see SLS DataStream Connector.

    Fixed issues

    • The error message that appears when the MySQL CDC connector fails to consume binary log data is optimized. The error message that appears when duplicate server IDs are used is optimized.

    • The CVE-2022-25857 vulnerability in SnakeYAML V1.27 is fixed.

    • The description for the real error information about specific issues, such as NullPointerException (NPE) and array-index out of bounds, is optimized for clarity.