Hi, I saw that there were some issues earlier today with Prefect Cloud. We’re currently still seeing a lot of our ECSRun tasks fail to begin execution correctly- are we potentially seeing delayed after effects from the earlier service disruption?

it could be. ECS is a bit hard to debug since it's serverless. In the worst case, you may reregister your flows. This will delete any currently scheduled runs and create new runs due to the new flow version

I think this should be stable now though. Do you have any errors?

No errors that I can see- just a message that the task was submitted for execution and then another message ~15 minutes later that it was rescheduled by a Lazarus process. Followed by another task was submitted message

We have been seeing stuff like this on ECS where jobs have been failing to start. I think this is on the ECS side where it can’t get the resources

Gotcha, I’ll investigate on the ECS side. Thank you!