is there way to pass in a parameter for retry_delay in a task? i have a task that imports google sheets, and want to set up some kind of exponential backoff to my attempts whenever i run into the API limit.

Hey @Harry Baker, unfortunately we don’t have exponential backoff. Are you on Cloud? You could use the task concurrency limiting and set it close to the API limit.

i'm probably gonna end up running everything from a local server, but might look into that. would it cause issues to have try/except logic within a task to create escalating wait times?

It wouldn’t…but how would you go about keeping the state for the next exponential retry?

yeah this is probably just gonna cause more issues. the amount of data i'm working with isn't that big so i can just throw in a couple time.sleep() calls to avoid hitting the API limit until i can figure out a more elegant solution

Yeah a random integer with

time.sleep(2**x)

would work.