Does anyone know whether it's possible to disable

dbt_columns

? Updating it simply seems like too heavy an operation. Following a deploy, the job has stalled with the following as the last lines in the debug log and is still running - 60 minutes later:

2023-10-31 13:51:55.454714 (MainThread): 13:51:55  Elementary: [dbt_columns] Flattening the artifacts.
2023-10-31 13:55:05.528915 (MainThread): 13:55:05  Elementary: [dbt_columns] Flattened 112703 artifacts.
2023-10-31 13:55:05.531158 (MainThread): 13:55:05  Elementary: [dbt_columns] Comparing the artifacts state.
2023-10-31 13:55:05.925794 (MainThread): 13:55:05  Elementary: [dbt_columns] Artifacts changed.

I did try disabling it in

dbt_project.yml

, but that just gives me a compilation error due to `dbt_artifacts_hashes`:

Compilation Error
  Model 'model.elementary.dbt_artifacts_hashes' (models\edr\dbt_artifacts\dbt_artifacts_hashes.sql) depends on a node named 'dbt_columns' which is disabled

Hi @Sindre Grindheim, I'm happy to update that I just pushed a var you can use to disable the columns collection. please update the package from revision since we haven't released a version yet:

packages:
  - git: <https://github.com/elementary-data/dbt-data-reliability.git>
    revision: 44bb408608775571e6128997458a27d67d63279c

then you can add the following to your dbt_project.yml:

vars:
    disable_dbt_columns_autoupload: true

this will not affect the report or alerts in any way, just stop collecting the column information. Please let me know if all works well

Hello, I updated to the latest version of elementary CLI and package 0.14.1 and now I have this error. I put the var, and it's not working.

Hey @Emanuel Junior, the car should still be supported 🤔 Can you share with me how you configured it, and your dbt.log?

We haven't been 100% happy with the functionality of the flag either, so we exclude the model specifically for our deploys:

--exclude dbt_columns

Hey @Sindre Grindheim, thank you for the feedback! Can I ask why you weren't happy with the flag functionality? It will help us improve 🙏

Because the model still runs during deploy runs if the package version has been updated. E.g

dbt run --select state:modified+ --full-refresh