Hey all 🙂 How would I go about getting exponential backoff when raising retry requested exception? is there a way to access the retry count from the context? Thanks!

https://docs.dagster.io/concepts/ops-jobs-graphs/op-retries#retrypolicy Backoff, including Jitter, max retries are available

thanks, not under

RetryRequested

though unless I'm reading wrong https://docs.dagster.io/concepts/ops-jobs-graphs/op-retries#retryrequested

Sorry sunday night, missed the requested aspect. Only looks like you can set the seconds to way when manually doing it

ah there is a

context.retry_number

🙂

If that works please let me know

yup works 🙂 exponential backoff implemented like so

RetryRequested(max_retries=5, seconds_to_wait=2**context.retry_number)