Yea we do @ classdojo; AMA
# random
d
Yea we do @ classdojo; AMA
d
Do you use mwaa and datahub with the lineage backend?
d
yes
with datahub lib version 0.8.31
d
Can you share how you set it up?
d
sure, there’s not much to it; we use the python library as opposed to adding an entrypoint to call the CLI. We use the pipeline class to write recipes inline as JSON
@dazzling-judge-80093 is there anything specific about the setup?
that you’d be interested in hearing more about
oh, sorry, you specifically asked about the lineage backend
we use the environment variables provided in the datahub documentation:
Copy code
AIRFLOW__LINEAGE__BACKEND=datahub_provider.lineage.datahub.DatahubLineageBackend
AIRFLOW__LINEAGE__DATAHUB_KWARGS={"datahub_conn_id":"datahub"}
and then we use AUTO for the airflow inlets, and specify our outlet dataset
just a FYI on this - MWAA doesn’t seem to support the lineage config, at least not on 2.0.2; I’m trying what someone else suggested (use env vars in your DAG), but i’m not convinced it’s going to work in production based on local testing with the mwaa runner