On September 15, 2026, all Cloud Composer 1 versions and versions 2.0.x of Cloud Composer 2 will reach their planned end of life. You will not be able to use environments with these versions. We recommend planning migration to Cloud Composer 3. Cloud Composer 2 versions 2.1.x and later are still supported and are not impacted by this change.
This page explains how to connect to a Cloud SQL instance that runs
the Airflow database of your Cloud Composer
environment and run SQL queries.
For example, you might want to run queries directly on the Airflow database,
make database backups, gather statistics based on the database content, or
retrieve any other custom information from the database.
Before you begin
Run a SQL query on the Airflow database
To connect to the Airflow database:
Create a DAG with one or more SQLExecuteQueryOperator operators. To get
started, you can use the example DAG.
In the sql parameter of the operator, specify your SQL query.
Trigger the DAG, for example, you can do it
manually or wait until it runs on a schedule.
Example DAG:
importdatetimeimportosimportairflowfromairflow.providers.common.sql.operators.sqlimportSQLExecuteQueryOperatorSQL_DATABASE=os.environ["SQL_DATABASE"]withairflow.DAG("airflow_db_connection_example",start_date=datetime.datetime(2025,1,1),schedule_interval=None,catchup=False)asdag:SQLExecuteQueryOperator(task_id="run_airflow_db_query",dag=dag,conn_id="airflow_db",database=SQL_DATABASE,sql="SELECT * FROM dag LIMIT 10;",)
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-25 UTC."],[[["\u003cp\u003eThis page explains how to connect to and run SQL queries on the Cloud SQL instance that hosts the Airflow database for your Cloud Composer environment.\u003c/p\u003e\n"],["\u003cp\u003eWhile direct access to the Airflow database is possible, it's advised to use alternative methods like the Airflow REST API or CLI commands whenever feasible.\u003c/p\u003e\n"],["\u003cp\u003eTo connect, create a DAG with \u003ccode\u003ePostgresOperator\u003c/code\u003e operators, specifying the SQL query in the \u003ccode\u003esql\u003c/code\u003e parameter, and uploading/triggering it.\u003c/p\u003e\n"],["\u003cp\u003eIt is not recommended to add custom tables or change the schema of the airflow database.\u003c/p\u003e\n"],["\u003cp\u003eBacking up the environment's data, including the Airflow database, should be done using snapshots instead of the deprecated database dumping method.\u003c/p\u003e\n"]]],[],null,["# Access the Airflow database\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\n**Cloud Composer 3** \\| [Cloud Composer 2](/composer/docs/composer-2/access-airflow-database \"View this page for Cloud Composer 2\") \\| [Cloud Composer 1](/composer/docs/composer-1/access-airflow-database \"View this page for Cloud Composer 1\")\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\nThis page explains how to connect to a Cloud SQL instance that runs\nthe [Airflow database](/composer/docs/composer-3/environment-architecture#airflow-database) of your Cloud Composer\nenvironment and run SQL queries.\n\nFor example, you might want to run queries directly on the Airflow database,\nmake database backups, gather statistics based on the database content, or\nretrieve any other custom information from the database.\n| **Important:** We recommend to avoid directly accessing the Airflow database, if it is possible to use other approaches such as [Airflow REST API](/composer/docs/composer-3/access-airflow-api) or [Airflow CLI commands](/composer/docs/composer-3/access-airflow-cli) instead.\n\nBefore you begin\n----------------\n\n| **Warning:** Don't add your own custom tables to the Airflow database and don't change the schema of the Airflow database. Don't add users or databases to the Cloud SQL instance that hosts the Airflow database.\n\nRun a SQL query on the Airflow database\n---------------------------------------\n\nTo connect to the Airflow database:\n\n1. Create a DAG with one or more SQLExecuteQueryOperator operators. To get\n started, you can use the example DAG.\n\n | **Caution:** Your **SQL query might run more than once** because of the DAG schedule and catchup. If you want to run the SQL query only once, set `schedule_interval` to `None`, `catchup` to `False`, and then [trigger the DAG manually](/composer/docs/composer-3/schedule-and-trigger-dags#manually).\n2. In the `sql` parameter of the operator, specify your SQL query.\n\n3. [Upload](/composer/docs/composer-3/manage-dags#add) this DAG to your environment.\n\n4. Trigger the DAG, for example, you can do it\n [manually](/composer/docs/composer-3/schedule-and-trigger-dags#manually) or wait until it runs on a schedule.\n\nExample DAG: \n\n import datetime\n import os\n\n import airflow\n from airflow.providers.common.sql.operators.sql import SQLExecuteQueryOperator\n\n SQL_DATABASE = os.environ[\"SQL_DATABASE\"]\n\n with airflow.DAG(\n \"airflow_db_connection_example\",\n start_date=datetime.datetime(2025, 1, 1),\n schedule_interval=None,\n catchup=False) as dag:\n\n SQLExecuteQueryOperator(\n task_id=\"run_airflow_db_query\",\n dag=dag,\n conn_id=\"airflow_db\",\n database=SQL_DATABASE,\n sql=\"SELECT * FROM dag LIMIT 10;\",\n )\n\nFor more information about using the SQLExecuteQueryOperator, see the\n[How-to Guide for Postgres using SQLExecuteQueryOperator](https://airflow.apache.org/docs/apache-airflow-providers-postgres/stable/operators.html)\nin the Airflow documentation.\n\nDump database contents and transfer them to a bucket\n----------------------------------------------------\n\n| **Deprecated:** This approach is deprecated. Instead, use [snapshots](/composer/docs/composer-3/save-load-snapshots) to back up the environment's data, including the Airflow database contents.\n\nWhat's next\n-----------\n\n- [Access Airflow REST API](/composer/docs/composer-3/access-airflow-api)\n- [Run Airflow CLI commands](/composer/docs/composer-3/access-airflow-cli)"]]