I set up an ECS agent that I can see running in prefect cloud, but when running a job I can see it is submitted but it will not execute. It is just stuck in the submitted state so not sure if I do not have enough storage of the agent can't get into one of the layers but there are not logs that say anything is wrong

Hi there, I have a CLI 2.0 question: is there currently a way to login to the Cloud API in a fully-scripted environment?

Hi! I’ve set up an ECS agent (Prefect Cloud 1.3.0) using this guide (great guide BTW). We’re using env variables to configure the used docker container. The problem is - newly spawned tasks in ECS do not get configured variables, and I do not see a way to explicitly forward them in the code. Is there a secure way to propagate env to flow runs other than specifying env in ECSRun (causing configuration duplication)?

Hello, is it just me or does anybody else also feel like Prefect 2 docs and UI are a downgrade compared to Prefect 1 docs and UI ? I feel Prefect 1 has much more polished and intuitive docs as well as the main UI. Is this because the main focus right now on Prefect 2 is to get the core framework right and then docs and UI will be upgraded and polished to bring it to Prefect 1 level ?

please help, prefect 2, when I try to run on a docker container

prefect cloud login --key...

it shows the following:

Hi all, what goes into "An Azure storage bucket path" when creating an Azure storage block? A couple of things I tried lead to "Container does not exist" errors from Azure. This is latest build of Prefect 2.0.2 with today's fix for .prefectignore included. (Yeah, I know... noob)

Hi all, trying to migrate my Prefect 1 GKE agent to Prefect 2 Orion and have it connect to Prefect Cloud. Followed the excellent demo on YouTube but when the pod starts up I get the following stack trace. I have tried 2.0.1 and 2.0.2 with different Python versions and I get the same error message back each time. Have also tried different ways to configure the kubectl secrets to see if that is the issue, wondering if others have run into the same issue?

Traceback (most recent call last):
  File "/usr/local/lib/python3.10/site-packages/prefect/cli/_utilities.py", line 41, in wrapper
    return fn(*args, **kwargs)
  File "/usr/local/lib/python3.10/site-packages/prefect/utilities/asyncutils.py", line 193, in wrapper
    return run_async_in_new_loop(async_fn, *args, **kwargs)
  File "/usr/local/lib/python3.10/site-packages/prefect/utilities/asyncutils.py", line 140, in run_async_in_new_loop
    return anyio.run(partial(__fn, *args, **kwargs))
  File "/usr/local/lib/python3.10/site-packages/anyio/_core/_eventloop.py", line 70, in run
    return asynclib.run(func, *args, **backend_options)
  File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 292, in run
    return native_run(wrapper(), debug=debug)
  File "/usr/local/lib/python3.10/asyncio/runners.py", line 44, in run
    return loop.run_until_complete(main)
  File "/usr/local/lib/python3.10/asyncio/base_events.py", line 646, in run_until_complete
    return future.result()
  File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 287, in wrapper
    return await func(*args)
  File "/usr/local/lib/python3.10/site-packages/prefect/cli/agent.py", line 93, in start
    async with OrionAgent(
  File "/usr/local/lib/python3.10/site-packages/prefect/agent.py", line 249, in __aenter__
    await self.start()
  File "/usr/local/lib/python3.10/site-packages/prefect/agent.py", line 237, in start
    await self.default_infrastructure._save(is_anonymous=True)
  File "/usr/local/lib/python3.10/site-packages/prefect/blocks/core.py", line 618, in _save
    await self.register_type_and_schema(client=client)
  File "/usr/local/lib/python3.10/site-packages/prefect/blocks/core.py", line 560, in register_type_and_schema
    block_type = await client.read_block_type_by_slug(
  File "/usr/local/lib/python3.10/site-packages/prefect/client.py", line 1093, in read_block_type_by_slug
    return BlockType.parse_obj(response.json())
  File "/usr/local/lib/python3.10/site-packages/httpx/_models.py", line 743, in json
    return jsonlib.loads(self.text, **kwargs)
  File "/usr/local/lib/python3.10/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/usr/local/lib/python3.10/json/decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/local/lib/python3.10/json/decoder.py", line 355, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

Hi, we're running a prefect server locally (prefect 1.0) and we're facing some issues (attaching

docker ps -a

output)

3f832d699b5e   prefecthq/server:core-0.15.11
 this one occupied nearly 40G
42G	3f832d699b5edd87e39af62f724bb143826f2c5ae929a0565d0ceb479d2befde-json.log

• These logs are taking so much space that our root gets full and our local prefect server goes down, so is there any way we can remove old logs, or divide these logs in dated files and remove old ones as per need, or can we set size limit for these logs file ? • and also is there any command to get prefect server status and agent status, so that we can send some alerts as per server status and agents' status.

Hello everybody Pricing is being checked as version 2 comes out. In the previous version 1, the price increased in proportion to the number of task runs. However, when I checked the price policy of version 2, the contents of the number of task runs disappeared. Did the number of runs and the price become irrelevant after upgrading to version 2?

Hi all, We have Prefect Server (2.0) deployed and I'm looking into adding users. Is this possible in a standard free and self-deployed Prefect Server? I dont require more teams to be setup, the default one is sufficient but would like to add users. I'm assuming this isn't possible, but I see it's free up to 3 users in Cloud though

Hello everyone, I have a Prefect 1.0 Question: In this example from @Anna Geller https://github.com/anna-geller/packaging-prefect-flows/blob/master/flows/azure_kubernetes_run_custom_azure_image.py Why do we need to configure a storage in the first place, if we’re going to package our prefect flow and codebase in an image anyway (

image="<http://prefectdemos.azurecr.io/community/flows|prefectdemos.azurecr.io/community/flows>"

)? I understand that the default storage is local and if we don’t set up a storage then our kubernetes agent can fail, but I just want to understand why we need storage in the first place when we have a

KubernetesRun

Hello Guys! I have a couple of questions about prefect's scheduler and agent 1. Is it possible to make Flow run Instances Dependent on each other. By that I mean two things a. Trigger another Deployment based on the final state of a current Deployment Flow run instance. For example, if I have an ETL flow run that loads data to a database, if successful, i want it to trigger another flow that would read the data from this database and proceed with other logic. b. Making the Run of a Scheduled Deployment dependent on previous runs, so that if i have a Deployment scheduled to run every 15 mins, the 11:00 AM run, and other proceeding runs, would halt if the 10:45 AM run failed. 2. I've been running prefect for a PoC for my use-case and have been facing difficulties with the agent. It seems that the agent would suddenly crash and print out a large error message in the middle of a flow run. This is part of the error message

Traceback (most recent call last):
  File "/usr/lib/python3.8/asyncio/selector_events.py", line 848, in _read_ready__data_received
    data = self._sock.recv(self.max_size)
ConnectionResetError: [Errno 104] Connection reset by peer

From my understanding, this means that there was some connection problems with a TCP/IP connection and it wasn't able to send an acknowledgment message, however, I dont understand why it happens and it seems to me that it happens randomly (probably not true). Does it have to do with the size of data being transmitted (I am extracting data from an S3 bucket for processing)? Thanks in Advance!

Hello everyone! Question about Prefect 2.0. We started working on versions early 2.0b8 and in this version was possibility to submit Flow code with REST API with

flow_data

key, so we deployed flows this way and in our case it was very useful. We need to deploy flow with REST API without using Prefect cli or prefect as a python package. Prefect runs on a totally separate server. Now in 2.0.2 I cannot find how to do this. I read about blocks but I cannot find that block used as default in Prefect to store flow data. So can anyone explain how implement same flow as with

flow_data

in version 2.0b8 and early? How can I just submit Flow code with rest api to deploy it on Server? Thanks in Advance!

In Prefect 2.0 CLI, is there a way when running

prefect deployment ls

to get only the deployment’s ids without the fancy drawing of the table? (it makes it difficult to parse the output to use with

xargs

). Then another thing, I’ve noticed that when doing

prefect deployment apply

if the deployment already exists in the cloud, it doesn’t get updated. Is there a way to force the re-deployment of an existing one?

when i build prefect deployment with gcs storage it tries to copy all my files including the virtaul enviroment to the gcs bucket and then fails (venv) haris@haris-Inspiron-3593:~/Desktop/snowflake$ prefect deployment build workflow/venue_flow.py:gen_venue_flow -n genvenueflow_deployment -t n1-standard-4 -i kubernetes-job --storage-block gcs/prefect2-snowflake /home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/flows.py176 UserWarning: A flow named 'gen_venue' and defined at 'workflow/venue_flow.py:6' conflicts with another flow. Consider specifying a unique

name

parameter in the flow definition:

@flow(name='my_unique_name', ...)

warnings.warn( Found flow 'gen_venue' Manifest created at '/home/haris/Desktop/snowflake/gen_venue_flow-manifest.json'. Traceback (most recent call last): File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/cli/_utilities.py", line 41, in wrapper return fn(*args, **kwargs) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/utilities/asyncutils.py", line 193, in wrapper return run_async_in_new_loop(async_fn, *args, **kwargs) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/utilities/asyncutils.py", line 140, in run_async_in_new_loop return anyio.run(partial(__fn, *args, **kwargs)) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/anyio/_core/_eventloop.py", line 56, in run return asynclib.run(func, *args, **backend_options) # type: ignore File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 233, in run return native_run(wrapper(), debug=debug) File "/usr/lib/python3.8/asyncio/runners.py", line 44, in run return loop.run_until_complete(main) File "/usr/lib/python3.8/asyncio/base_events.py", line 616, in run_until_complete return future.result() File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 228, in wrapper return await func(*args) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/cli/deployment.py", line 502, in build file_count = await storage.put_directory() File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/filesystems.py", line 428, in put_directory return await self.filesystem.put_directory( File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/prefect/filesystems.py", line 250, in put_directory self.filesystem.put_file(f, fpath) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/fsspec/asyn.py", line 86, in wrapper return sync(self.loop, func, *args, **kwargs) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/fsspec/asyn.py", line 66, in sync raise return_result File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/fsspec/asyn.py", line 26, in _runner result[0] = await coro File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/core.py", line 1045, in _put_file await simple_upload( File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/core.py", line 1596, in simple_upload j = await fs._call( File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/core.py", line 392, in _call status, headers, info, contents = await self._request( File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/decorator.py", line 221, in fun return await caller(func, *(extras + args), **kw) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/retry.py", line 115, in retry_request return await func(*args, **kwargs) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/core.py", line 384, in _request validate_response(status, contents, path, args) File "/home/haris/Desktop/snowflake/venv/lib/python3.8/site-packages/gcsfs/retry.py", line 100, in validate_response raise ValueError("Bad Request: %s\n%s" % (path, msg)) ValueError: Bad Request: https://storage.googleapis.com/upload/storage/v1/b/prefect2-snowflake/o Payload size invalid. Expected size: 2. Actual size: 3. An exception occurred.

We are evaluating some solutions for our data workflows. For example, Google Cloud Composer (Airflow) and Prefect. If we decide to use Prefect, would be okay to starting using Prefect Cloud 2? It is still on Beta, when will be production ready?

Hey, Facing some issues with Orion 2.0.2 and not sure how to debug those. It seems to fail when calling

update_flow_run

in

engine.py

. I’m running a flow via the docker-container infrastructure type. It seems to start, but there’s no output on the UI, in fact the flow run status stays in “Pending” state, even after it fails on the work queue. I also see some TF log outputs which appear when importing TF, but the actual log output of my script is not showing up. Are there any hints on how to solve this?

We created a deployment:

<https://app.prefect.cloud/account/400***/workspace/8d8***/deployment/2f5***>

- it has the

staging

tag. We have a running agent:

$ /usr/bin/python3 /usr/local/bin/prefect agent start 360***
Starting agent connected to 
<https://api.prefect.cloud/api/accounts/400***/workspaces/8d8***>...
  ___ ___ ___ ___ ___ ___ _____     _   ___ ___ _  _ _____
 | _ \ _ \ __| __| __/ __|_   _|   /_\ / __| __| \| |_   _|
 |  _/   / _|| _|| _| (__  | |    / _ \ (_ | _|| .` | | |
 |_| |_|_\___|_| |___\___| |_|   /_/ \_\___|___|_|\_| |_|
Agent started! Looking for work from queue 
'360***'...

The work queue also has the tag

staging

. When I click on "Run" in the Cloud UI in the deployment, it is scheduled but the flow run has no tags assigned. And I guest this is the reason why it is never picked up by the agent. Or how can I debug this issue?

I'm trying to create a deployment with

prefect deployment apply deployment.yaml

but I keep getting an error from sqlalchemy:

sqlalchemy.exc.IntegrityError: (sqlite3.IntegrityError) FOREIGN KEY constraint failed

When I run

prefect deployment build file.py:flow_name --name some-deployment -sb azure/block-name

I get the error

TypeError: put_directory() got an unexpected keyword argument 'ignore_file'.

I then went into prefect/cli/deployment.py and removed the ignore_file arg in line 470 and ran the command again and only

file.py

is added to my azure container and everything seems to be stuck in an endless loop. Has anyone had this problem/know what could have gone wrong? Cheers 🙂 TLDR:

prefect deployment build

stuck in endless loop and not accepting ignore_file arg.

is there any writeup why prefect is build around the idea of storing code in a block storage instead of a docker image? And related, @Anna Geller I saw you are planning to write some deployment recipes. Would it be possible to also write one that packs everything into a docker image instead of splitting dependencies and flow code?

Why can't I create users with the GraphQL API? Even though my account is an admin one ... 🤔 What kind of permissions do I need?

Hi, how can I import SFTP task class?

prefect.tasks.sftp.sftp

I’ve tried 1.20, 2.0 version but all don’t have this module available.

And do we have a list of what each of those roles mean ? For example what is "audit trail" ? Should my normal "users" have read access to it ? 🤷

Hello, I'm evaluating using prefect 2.0 with a local server. I think I'm missing a very basic concept about deployment. I want to set up a flow that is called many times a day, each time given a different data file as the parameter. I can do this quite easily locally by calling my flow's method with the file path as the parameter. But if I want to use the Orion server to manage these flow runs, I don't see how to launch a flow from an external application sending a different parameter value as the parameter value is tied to the deployment. It seems strange to have to create a deployment each time with this parameter (lots of extra work, and will quickly have thousands of deployments each used exactly once.)

Another livestream today with @Kalise Richmond starting in 1 hour, if you want to talk to someone live about Prefect 2.0! Register here to join!

Hi, Community. Thank you a lot for your help in advance. I have a question related to an issue we have seen in the past. I will proceed to describe the situation: 1. Suppose the there is a parent flow P with version 1 and a child flow C with version 1. They run daily. 2. The parent flow invokes the child flow through StartFlowRun. 3. Suppose P1 is running and it is about to execute C1. 4. However, all flow are re-registered sequentially. The end result is that P2 and C2 become available. 5. The StartfFlow task for C1 remained in a pending state and then there was a TriggerFailed error. 6. C did not register any new Flow Run. 7. Agents did not receive any information on C. So, I was wondering if concurrency problems are possible if flows of flows are running and suddenly new versions are registered for all of them.

Hello 🙂 I have a question about Work Queues (image on thread). I have two work queues with the same tag (staging), if I triggered multiple runs (lets say five), this runs will be evenly distributed between the available agents? I’m asking this because I dont see a way to check on which queue the task was executed, and from the output of the CLI looks like only one agent is picking up everthing. I tested triggering the jobs with the agents off, and the list of flows appeared on both Work Queues as pending.

Hey guys! I am having an issue with running flows as KubernetesJob together with prefect UI v2.0.2 in my GCP Kubernetes. let me describe all the steps. I guess this may be useful for other guys in community, as prefect docs/ho-tos are becoming quickly outdated due to rapid development: • save the output of

prefect kubernetes manifest orion

to

k8s_deployment.yaml

(please add the ability to set non-default k8s namespace) • change PREFECT_API_URL from http://orion:4200/api to http://0.0.0.0:4200/api •

kubectl apply -f k8s_deployment.yaml

(it already has command: [“prefect”, “orion”, “start”, “--host”, “0.0.0.0”, “--log-level”, “WARNING”] for API container) • check the GCP admin. The agent/ui/api seems to be healthy. •

kubectl port-forward deployment/orion 4200:4200

• check the UI is healthy via http://127.0.0.1:4200/. •

prefect config set PREFECT_API_URL=<http://0.0.0.0:4200/api>

• register GCS block via admin panel. name=prefect. path: gs://{my-path}. (prefect block register -f gcs_block.py didn’t work for me for some reason) •

prefect deployment build ../flows/backtest.py:backtest_flow --name dev --tag dev --infra kubernetes-job --storage-block gcs/prefect

(should i run it from root of the project?) • checked my GCS bucket. The files did appear in the local folder and GCS. •

prefect deployment apply backtest_flow-deployment.yaml

• check UI http://127.0.0.1:4200/deployments. deployment did appear • create

kubernetes

queue manually. The name seems to be hardcoded according to the error logs in GCP agent. This step fixes the error. •

prefect deployment run Backtest/dev

• the flow run was created. But it is stuck in pending. • The appropriate KubernetesJob was created. But it is stuck in error state with the error All connection attempts failed. and ConnectionRefusedError: [Errno 111] Connect call failed (‘0.0.0.0’, 4200). Please help! Thank you in advance :)

Is it possible to start multiple agents from the command line by only informing the tag? For example, If I execute:

prefect agent start -t test

I can see that a new “Work Queue” is added to the UI (

Agent queue test

) But if I start a new agent (on another tab) with the same command, the UI keeps the same. Is this the expected behavior? Would it be possible to add a

--name

parameter on the start command to allow starting multiples agents to consume the same tag?