Thank you, I've applied the suggested changes and increased the exponential backoff factor for our retries, plus I set the job status check to every two minutes instead of the 30 seconds that we were doing before. Problem is, I'm still seeing the 429 errors occasionally. We do have a large number of ACOs that we pull data for regularly (up to 58 I believe now), and to save a ton of time, we do some of that work in parallel, so I can only assume that's causing us to hit the limit.
My follow-up question is: Does every API call count towards this total? (Job status check, token retrieval, file download, etc) My concern is that since there are potentially hundreds of files to download per ACO, and we have many ACOs, we hit that limit pretty regularly, especially on a fast network connection, and the only thing we can do is to space out our ACOs more, which will slow us down significantly.