This is an automated email from the ASF dual-hosted git repository.
shahar1 pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/airflow.git
The following commit(s) were added to refs/heads/main by this push:
new a1cd080b7db Stop recommending psycopg3 connection strings on older
Airflow (#70619)
a1cd080b7db is described below
commit a1cd080b7db25252d92ccbaa7cdf7f2c19378f4b
Author: Shahar Epstein <[email protected]>
AuthorDate: Fri Sep 11 21:52:20 2026 +0300
Stop recommending psycopg3 connection strings on older Airflow (#70619)
---
providers/amazon/docs/executors/batch-executor.rst | 6 +++---
providers/amazon/docs/executors/ecs-executor.rst | 6 +++---
providers/amazon/docs/executors/general.rst | 15 +++++++++++++++
providers/amazon/docs/executors/lambda-executor.rst | 6 +++---
providers/celery/provider.yaml | 5 ++++-
.../src/airflow/providers/celery/get_provider_info.py | 4 ++--
6 files changed, 30 insertions(+), 12 deletions(-)
diff --git a/providers/amazon/docs/executors/batch-executor.rst
b/providers/amazon/docs/executors/batch-executor.rst
index 8f71ba7d8c5..51ffe2b433f 100644
--- a/providers/amazon/docs/executors/batch-executor.rst
+++ b/providers/amazon/docs/executors/batch-executor.rst
@@ -296,9 +296,9 @@ Create a Job Definition
- ``AIRFLOW__DATABASE__SQL_ALCHEMY_CONN``, with the value being the PostgreSQL
connection string in the following format using the values set during the
`Database section <#create-the-rds-db-instance>`_ above:
-.. code-block:: bash
-
- postgresql+psycopg://<username>:<password>@<endpoint>/<database_name>
+.. include:: general.rst
+ :start-after: .. BEGIN SQL_ALCHEMY_CONN
+ :end-before: .. END SQL_ALCHEMY_CONN
7. Add other configuration as necessary for Airflow generally (see `here
<https://airflow.apache.org/docs/apache-airflow/stable/configurations-ref.html>`__),
the Batch executor (see :ref:`here <config-options>`) or for remote logging
(see :ref:`here <logging>`). Note that any configuration changes should be made
across the entire Airflow environment to keep configuration consistent.
diff --git a/providers/amazon/docs/executors/ecs-executor.rst
b/providers/amazon/docs/executors/ecs-executor.rst
index fd534cb27de..54fdc927db2 100644
--- a/providers/amazon/docs/executors/ecs-executor.rst
+++ b/providers/amazon/docs/executors/ecs-executor.rst
@@ -330,9 +330,9 @@ Create Task Definition
- ``AIRFLOW__DATABASE__SQL_ALCHEMY_CONN``, with the value being the PostgreSQL
connection string in the following format using the values set during the
`Database section <#create-the-rds-db-instance>`_ above:
-.. code-block:: bash
-
- postgresql+psycopg://<username>:<password>@<endpoint>/<database_name>
+.. include:: general.rst
+ :start-after: .. BEGIN SQL_ALCHEMY_CONN
+ :end-before: .. END SQL_ALCHEMY_CONN
- ``AIRFLOW__ECS_EXECUTOR__SECURITY_GROUPS``, with the value being a comma
separated list of security group IDs associated with the VPC used for the RDS
instance.
diff --git a/providers/amazon/docs/executors/general.rst
b/providers/amazon/docs/executors/general.rst
index a337fffcbf8..8de62ba3f6d 100644
--- a/providers/amazon/docs/executors/general.rst
+++ b/providers/amazon/docs/executors/general.rst
@@ -312,6 +312,21 @@ You will be prompted to enter the password if the
connection is successful.
.. END DATABASE_CONNECTION
+.. BEGIN SQL_ALCHEMY_CONN
+
+.. code-block:: bash
+
+ postgresql+psycopg2://<username>:<password>@<endpoint>/<database_name>
+
+.. note::
+ ``psycopg2`` works on every Airflow version these executors support. From
Airflow 3.2.0 you may use
+ ``postgresql+psycopg://`` instead, provided ``psycopg`` (v3) is installed.
Airflow before 3.2.0 does not
+ guarantee SQLAlchemy 2.0 (2.11 and 3.0.x pin ``<2.0``), and SQLAlchemy 1.4
has no
+ ``postgresql+psycopg`` dialect — Airflow fails to start with
``sqlalchemy.exc.NoSuchModuleError``.
+
+.. END SQL_ALCHEMY_CONN
+
+
.. BEGIN ECR_STEPS
Create an ECR Repository
diff --git a/providers/amazon/docs/executors/lambda-executor.rst
b/providers/amazon/docs/executors/lambda-executor.rst
index 4ceb5c3bea9..7e5ae77bc86 100644
--- a/providers/amazon/docs/executors/lambda-executor.rst
+++ b/providers/amazon/docs/executors/lambda-executor.rst
@@ -332,9 +332,9 @@ Finally create the function:
- ``AIRFLOW__DATABASE__SQL_ALCHEMY_CONN``, with the value being the PostgreSQL
connection string in the following format using the values set during the
`Database section <#create-the-rds-db-instance>`_ above:
-.. code-block:: bash
-
- postgresql+psycopg://<username>:<password>@<endpoint>/<database_name>
+.. include:: general.rst
+ :start-after: .. BEGIN SQL_ALCHEMY_CONN
+ :end-before: .. END SQL_ALCHEMY_CONN
- ``AIRFLOW__LAMBDA_EXECUTOR__QUEUE_URL``, with the value being the URL of the
SQS queue created above.
diff --git a/providers/celery/provider.yaml b/providers/celery/provider.yaml
index 2f3122c1f36..ad6d6e0278f 100644
--- a/providers/celery/provider.yaml
+++ b/providers/celery/provider.yaml
@@ -248,10 +248,13 @@ config:
The use of a database is highly recommended
When not specified, sql_alchemy_conn with a db+ scheme prefix will
be used
https://docs.celeryq.dev/en/latest/userguide/configuration.html#task-result-backend-settings
+ The psycopg2 driver in the example below works on every supported
Airflow version.
+ Set db+postgresql+psycopg:// only on Airflow 3.2.0 and later -
earlier releases do not
+ guarantee SQLAlchemy 2.0, which is where the psycopg (v3) dialect
lives
version_added: ~
type: string
sensitive: true
- example: "db+postgresql+psycopg://postgres:airflow@postgres/airflow"
+ example: "db+postgresql+psycopg2://postgres:airflow@postgres/airflow"
default: ~
result_backend_sqlalchemy_engine_options:
description: |
diff --git a/providers/celery/src/airflow/providers/celery/get_provider_info.py
b/providers/celery/src/airflow/providers/celery/get_provider_info.py
index 02c71ab36ac..7fd3ba29bde 100644
--- a/providers/celery/src/airflow/providers/celery/get_provider_info.py
+++ b/providers/celery/src/airflow/providers/celery/get_provider_info.py
@@ -126,11 +126,11 @@ def get_provider_info():
"default": "redis://redis:6379/0",
},
"result_backend": {
- "description": "The Celery result_backend. When a job
finishes, it needs to update the\nmetadata of the job. Therefore it will post a
message on a message bus,\nor insert it into a database (depending of the
backend)\nThis status is used by the scheduler to update the state of the
task\nThe use of a database is highly recommended\nWhen not specified,
sql_alchemy_conn with a db+ scheme prefix will be
used\nhttps://docs.celeryq.dev/en/latest/userguide/configuration. [...]
+ "description": "The Celery result_backend. When a job
finishes, it needs to update the\nmetadata of the job. Therefore it will post a
message on a message bus,\nor insert it into a database (depending of the
backend)\nThis status is used by the scheduler to update the state of the
task\nThe use of a database is highly recommended\nWhen not specified,
sql_alchemy_conn with a db+ scheme prefix will be
used\nhttps://docs.celeryq.dev/en/latest/userguide/configuration. [...]
"version_added": None,
"type": "string",
"sensitive": True,
- "example":
"db+postgresql+psycopg://postgres:airflow@postgres/airflow",
+ "example":
"db+postgresql+psycopg2://postgres:airflow@postgres/airflow",
"default": None,
},
"result_backend_sqlalchemy_engine_options": {
