Silly question, perhaps, but is the new prefect v3 usable with the current prefect cloud?

Not a silly question! Prefect Cloud supports Prefect v 2.x and 3.x

Thx! Will give it a try!

Awesome! Let us know what you think 😄

@Dylan are we sure about 3.x, we updated ut everything is broken

Hey Rupin, I'm sorry to hear you're encountering some friction while upgrading. First, gentle reminder that tagging other users is against our code of conduct. If you need dedicated or immediate support you can reach out to us at sales@prefect.io or you can book time with an engineer. I haven't seen that issue before. To be clear, did you upgrade the Prefect version on your worker before attempting to run a flow/deployment on 3.0.0rc9? If not, I would try that first since the version of the worker must be greater than the version of the flow/deployment. If your worker is what's erroring, I encourage you to open an issue with a minimum reproducible example and we will attempt to reproduce your issue.

ah did not see the code of conduct, what do u mean by did you upgrade the Prefect version on your worker? The worker is pulling ECR upgraded with 3.0.0rc9

So there are a few components here that have prefect versions: • The Prefect Worker, version usually determined by docker image as you described above • The Flow Run or deployment/flow instance, determined by a variety of factors depending on how its run

if that is the case, does not work we did flow.deploy() fresh that still the problem remains. I was think you are talking about prefect work-pool create --type ecs:push need redo

You shouldn't need to recreate the work pool from the CLI

To be clear - you will need to redeploy the worker (the deployed infrastructure, running on ECS) - u mean construct_etl.deploy(XXX) need redo

Can you share a little bit more about how you have your worker & deployment set up?

Try recreating the worker ecs service following the steps here: https://docs-3.prefect.io/integrations/prefect-aws/ecs_guide#step-3-creating-an-ecs-worker-service

I might misunderstood something, I though everything is automatically provisioned, we literally just do prefect work-pool create --type ecs:push --provision-infra prefect-ecs-fargate-work-pool-example, and provision everything, and then deploy via flow.deploy(), we never need touch any worker config

Can you share your ecs service config?

Could u provide some context on the config u talk about?

Ahh okay that makes sense. You can try reprovisioning: https://docs-3.prefect.io/3.0rc/deploy/dynamic-infra/push-runs-serverless#provision-infrastructure-for-an-existing-push-work-pool If that doesn't work I think you may need to write up a GitHub issue with a minimum reproducible example and we can triage and address from there

Dp u know if I reprovision the existing deployment ID will change? this is very important for us, as we use in many external API calls, if that is the case I will just revert

No this shouldn't impact your deployment at all

wonderful, thank u so much for detailed explain

Of course!