@Marvin, how do i fail a flow if a task fails?

Hey guys, I am getting this error after helm install prefect/prefect-worker

I am getting this error while executing a flow via Cloud Prefect 2.0 "RuntimeError: This portal is not running". Any clue what is going wrong. The same flow was working fine before I use Cloud Prefect 2.0. Actually it finishes all my flows and tasks successfully and at the end it crashes and it happens to all my existing flows. line 1512, in propose_state raise prefect.exceptions.Abort(response.details.reason) prefect.exceptions.Abort: This run has already terminated.

I have scheduled flow run using runner but saw below warning, does going forward this can be any issue

<frozen runpy>:128: RuntimeWarning: 'prefect.engine' found in sys.modules after import of package 'prefect', but prior to execution of 'prefect.engine'; this may result in unpredictable behaviour

Hi guys, can I use S3 block to download files from S3 bucket? or am I just using the block as secret to the AWS S3Download task? any examples?

Hi all! I was wondering how we can change the UI port from 8080 to something else. I have changed the

PREFECT__SERVER__UI__PORT

and

PREFECT__SERVER__UI__HOST_PORT

to try and follow the config.toml, but the UI service is still launching and saying “👾👾👾 UI running at localhost:8080 👾👾👾” and I dont know what else to change in the UI service 😞 As a follow up, what is the difference between the UI port and the UI host port? EDIT: I see echo

UI running at localhost:8080

is hardcoded in ./start_server.sh in the UI service, is this meant to be the case?

Hi, I'm new to Prefect and installed a local server. Last week everything seemed to work and I started to write my first flows. They were executed as expected, also the UI worked great. However today morning, I tried to continue working on my flows, but for some reason the UI was frozen and I just restarted the containers through docker-compose (down and up again). After that I experienced a curious / weird output in the Postgres logfile (I will explain below). Even the UI was working fine again, the logfile of Postgres was quite confusing. So I thought, that something has messed up with my agent or other communication issues between Prefect server and the database. I decided to make a new clean setup in a new clean virtual environment. I also setup a path to a new database, and deleted the .prefect folder in the home directory. But even with the clean setup the weird behavior of Postgres started right away: ==> More details in thread

Hi all, I'm overhauling our current Prefect 2.19.1 to Prefect 3.1.0. Because our codebase dynamically generates a lot of deployments we use Prefect Python and not Prefect YAML to manage everything. Surprisingly the Python code has really stayed mostly the same between versions, which is great. Unsurprisingly some of our side scripts we use to manage the Cloud are now broken. One in particular we use to clear backlogs of Late runs is giving me fits. This code used to work

import asyncio

from prefect.states import State
from prefect.client import get_client

MAX_RUNS_TO_DELETE = 5000
PREFECT_OFFSET_MAX = 200


async def remove_all_flows():
    client = get_client()
    for i in range(0, MAX_RUNS_TO_DELETE, PREFECT_OFFSET_MAX):
        flow_runs = await client.read_flow_runs(offset=i)
        for flow_run in flow_runs:
            if flow_run.state_name == "Late":
                flow_id = flow_run.id
                print("deleting", flow_id)
                await client.set_flow_run_state(flow_run_id=flow_id, state=State(type="CANCELLED"))


asyncio.run(remove_all_flows())

But now returns an error

pydantic.errors.PydanticUserError: `StateCreate` is not fully defined; you should define all referenced types, then call `StateCreate.model_rebuild()`.

No amount of defining State parameters (or passing directly a StateCreate object) fixes the issue. I'm increasingly convinced the old code is now a dead end. Could the Prefect team advise on what the idiomatic way to cancel a Late flow run would be under Prefect 3.1+?

@Marvin Why am I getting this error:

ValueError: Unable to create hash - objects could not be serialized.
  JSON error: Unable to serialize unknown type: <class 'pandas.core.frame.DataFrame'>
  Pickle error: cannot pickle 'weakref.ReferenceType' object

This often occurs when task inputs contain objects that cannot be cached like locks, file handles, or other system resources.

To resolve this, you can:
  1. Exclude these arguments by defining a custom `cache_key_fn`
  2. Disable caching by passing `cache_policy=NONE`

@Marvin How do I configure the prefect dashboard to run on http://0.0.0.0:4200? and it to use "http://0.0.0.0:4200/api" as well