Docs Home
Viewing docs for
BYOCSelf-Managed

SQL Console

On this page

SQL Console provides a dedicated environment for running ad hoc SQL statements against a session cluster. Unlike SQL Drafts, which are compiled into persistent Deployments, SQL Console files are designed for one-time operations that do not need to run continuously.

Overview

SQL Console lets you write and execute SQL scripts interactively in the console. Each SQL Console file is a script you can run immediately against a session cluster — no deployment configuration required.

SQL Console vs. SQL Draft:

AspectSQL DraftSQL Console
OutputCreates a persistent DeploymentExecutes statements once in a session cluster
Use caseContinuous streaming or batch jobsAd hoc queries, DDL operations, data management
ExecutionDeployed as a jobRuns directly in the session cluster

Choose between SQL Editor and Scripts

The console lists two entries for writing SQL: SQL Editor and Scripts. SQL Console files, described on this page, are the files you create in Scripts. Both entries use the same editor layout, but they serve different purposes:

  • SQL Editor is for SQL jobs that run as a Deployment. You write a SQL draft, validate and debug it, then deploy it. Like JAR and Python Deployments, the result is a long-running application.
  • Scripts is for one-time operations, such as creating tables, inserting initial data, querying data, or running CALL commands. It works like the Flink SQL Client inside the console. A script runs once in a session cluster and can't become a Deployment.

A script keeps running when you close the browser tab, because it runs in the session cluster, not in your browser. It stops when its statements finish.

Access to Scripts follows access to session clusters. No separate permission restricts DDL or DML statements in Scripts. To limit who can run scripts, restrict access to the session clusters.

Key Use Cases

  • Run DDL statements to create, alter, or drop catalogs, databases, and tables.
  • Query data with DQL statements for ad hoc exploration and validation.
  • Insert or manipulate data with DML statements.
  • Execute CALL commands — for example, to compact or manage Apache Paimon tables.
  • Run EXPLAIN statements to inspect execution plans and optimize performance.

Prerequisites

  • A running session cluster in your workspace.
  • VERA 1.0.6 or later is required to use CALL commands.

Create a SQL Console File

  1. In the Console navigation pane, click Scripts.
  2. Click the Create draft icon.
  3. In the Create Script dialog, enter a name and select a location for the file.
  4. Click Save.

The new file opens in the editor. SQL Console files are saved automatically every minute.

Write and Edit SQL Scripts

SQL Console files support the following statement types:

  • DDL — CREATE, ALTER, DROP statements for catalogs, databases, and tables.
  • DML — INSERT INTO and similar data manipulation statements.
  • DQL — SELECT statements for querying data.
  • CALL — commands for managing external systems such as Apache Paimon.

You can write multiple statements in a single file. Statements are separated by semicolons.

Execute a Script

Scripts run in batch mode by default.

Run the Entire Script

Click the Run button in the top-right corner of the editor. All statements in the file are executed sequentially.

Run Selected Statements

Select the statements you want to run in the editor. A Run button appears at the top-left of the selection. Click it to execute only the selected statements.

Switch to Streaming Mode

To run statements in streaming mode, add the following SET statement at the top of your script before the statements you want to execute:

SQL
1SET 'execution.runtime-mode' = 'streaming';

View Results and Logs

After execution, results are displayed in the output panel below the editor:

  • CALL commands — execution logs are shown.
  • DDL statements — a success or failure message is shown.
  • DQL statements — the query result set is shown.

Delete a SQL Console File

To delete a SQL Console file, right-click the file in the Scripts tab and select Delete.

Limitations

  • SQL Console runs in a session cluster only. It cannot be deployed as a standalone Deployment.
  • No deployment debugging, version history, or code structure view is available.
  • CALL command support requires VERA 1.0.6 or later.
  • DDL and DML statements affect live metadata or data immediately upon execution.
Was this helpful?