Out of the box, there appear to be several issues with running the Airflow DBT examples using Docker.
The 1.0.x release of dbt should be installed using pip install dbt-core or pip install dbt-<connector>, e.g., pip install dbt-postgres. This applies both to the Dockerfile and to requirements.txt
Dependency resolution complains about a few packages, ultimately resulting in Airflow failing to start up. I've been able to resolve this by pinning dbt-postgres<1.0.0, wtforms==2.3.3 , and werkzeug<1.0.0 in requirements.txt, which is required by Airflow 1.10.9.
The dbt_project.yml is missing the config-version: 2 setting, which prevents DAGs from executing
Out of the box, there appear to be several issues with running the Airflow DBT examples using Docker.
pip install dbt-core
orpip install dbt-<connector>
, e.g.,pip install dbt-postgres
. This applies both to the Dockerfile and torequirements.txt
dbt-postgres<1.0.0
,wtforms==2.3.3
, andwerkzeug<1.0.0
inrequirements.txt
, which is required by Airflow 1.10.9.dbt_project.yml
is missing theconfig-version: 2
setting, which prevents DAGs from executingairflow/ge_tutorials_dag_with_great_expectations.py
, though it looks like https://github.com/superconductive/ge_tutorials/pull/16 addresses this