This topic describes the major updates and bug fixes of the Realtime Compute for Apache Flink version released on May 6, 2024.
The version upgrade is incrementally rolled out across the network using a canary release strategy. Check the latest announcement on the right side of the Realtime Compute for Apache Flink management console for the upgrade schedule. The new features are available only after the upgrade is complete for your account. To request an earlier upgrade, submit a ticket.
Overview
This release improves development efficiency and O&M (operations and maintenance) capabilities.
More efficient development: Use code search to find SQL drafts by string — across all drafts at once. Use deployment-level variables to manage shared logic across batch deployments and streaming deployments without duplicating code.
Enhanced O&M: Configure alert rules in batch for multiple namespaces and deployments. Use alert levels and alert scheduling to handle different monitoring scenarios.
Features
| Feature | Description | References |
|---|---|---|
| Code search in SQL drafts | Search for any string across all SQL drafts at once. | N/A |
| Deployment-level variables | Define variables within a deployment and update their values during O&M. SQL users add variables in DDL (Data Definition Language); Java and Python users add variables in the arguments of the main function. | Manage keys |
| Application Real-Time Monitoring Service (ARMS) monitoring and alerting optimization | The name and ID of the corresponding Realtime Compute for Apache Flink workspace now appear on the Prometheus instance details page. When configuring alert rules in the ARMS console, you can now: configure rules for a group of namespaces and deployments that match specific filtering conditions; write custom PromQL to create alerts for metrics not supported by Realtime Compute for Apache Flink; and configure advanced options such as alert levels and alert scheduling. | Configure alert rules in the ARMS console. For Prometheus-specific setup, see Create an alert rule for a Prometheus instance. |
| Limit optimization | Two limits are updated: the per-namespace limit on UDF and JAR package uploads is removed, and the maximum number of files and subfolders within a folder is increased to 10,000. | Limits |
| Simple Log Service and StarRocks catalog creation via GUI | Create Simple Log Service and StarRocks catalogs directly from the console by using the GUI. | Manage Simple Log Service catalogs. For StarRocks catalogs, see Manage StarRocks catalogs. |