DBT asset causes `All assets in a job must share the same BackfillPolicy` error about dagster HOT 7 CLOSED

I encountered this also. #21259 (comment)

from dagster.

I think the best thing to do is to revert to 1.7.5 and wait for a proper fix.

from dagster.

@aaaaahaaaaa @the4thamigo-uk Yeah I kind of resolve that issue for my dbt assets just by being explicit and setting the same backfill policy - in your case BackfillPolicy(max_partitions_per_run=None) for all dbt assets. In my case the partitioned ones ware causing some headaches. I might make sense to build a factory that will return you dbt assets with explicitly defined BackfillPolicys

from dagster.

This is related to the feature request I raised here

• #22356

from dagster.

@mgierada That would be fine except for the fact that I have a job that executes unpartitioned dbt assets and partitioned ones. My dbt project is split into two separate @dbt_asset multi assets. One contains the unpartitioned dbt models, and the other contains the partitioned dbt models. Unfortunately, dagster-dbt forces me to have to specify BackfillPolicy::single_run() for the unpartitioned @dbt_asset.

If I set everything to single_run() then I am forced to have to change my implementation of my non-dbt partitioned assets to support processing of multiple partitions, which is a pain.

from dagster.

@the4thamigo-uk I feel you. I am facing the same issue in my project. I raised a feature request #22356 that would be a workaround or a permanent fix, depending on the context.

from dagster.

New release seems to fix the issue: https://github.com/dagster-io/dagster/releases/tag/1.7.10

from dagster.