Hey there, We have a Flyte-Binary deployment in AWS, for which I provision an RDS database and S3 buckets for the Flyte backend. Recently we have considered moving to Flyte-Core instead. Would we be able to reuse said RDS and S3 buckets so we don't lose run history?

Taking a backup before reinstall is recommended. The DB schemas don't change with the Helm chart you use, so you should be fine. Just curious, what are the motivations to switch to flyte-core?

• Adding more replicas so reduce downtime, since Flyte-Binary pod crashes often and has memory leak • Can't find propeller metrics in Flyte-Binary, whereas Flyte-Core should have them available

Thanks Nuno. It's the second time I hear about sudden crashes on flyte-binary. Feel free to provide only the information you have handy, but it would be helpful to see if there's a pattern. • Do you have logs from those events? • What integrations have you enabled? • What version of flyte are your running?