hi, how can I set

op_retry_policy

on

define_asset_job

?

tags={"dagster/max_retries": 2}

would probably work. But I'm also interested in controlling delay between retries..

Take a look at this, I think it answers your questions.... https://docs.dagster.io/concepts/ops-jobs-graphs/op-retries#retrypolicy

@op(
    retry_policy=RetryPolicy(
        max_retries=3,
        delay=0.2,  # 200ms
        backoff=Backoff.EXPONENTIAL,
        jitter=Jitter.PLUS_MINUS,
    )
)
def even_better():
    fails_sometimes()

That's not it, it's is for

op

. We have a job that materializes assets via

define_asset_job

. You can pass op retry policy (the one that you mentioned) when you define custom job via

job

decorator. But not for

define_asset_job

as far as I can tell.

If you want to define retry on the asset level you can use

retry_policy

via @asset (api link included). On the job level, I’m not sure if you can control the delay but there is a

retry_strategy

tag which I don’t use on asset jobs but probably still works. https://docs.dagster.io/_apidocs/assets

we're loading most of the assets from airbyte and dbt atm, with

load_assets_from_airbyte_instance

&

load_assets_from_dbt_project

, so no decorator. And I want to set retries based on the job and not on the asset, if possible.