Jenny
06/16/2023, 2:54 PMprefect cloud webhook create your-webhook-name \
--description "Receives webhooks from your system" \
--template '{ "event": "your.event.name", "resource": { "prefect.resource.id": "your.resource.id" } }'
For more information, see the release notes! 🎉Chris Reuter
06/19/2023, 8:44 PMChris Reuter
06/28/2023, 3:38 PMprefect init
!Will Raphaelson
06/30/2023, 3:19 AMWill Raphaelson
07/06/2023, 2:38 PMjustabill
prefect config set PREFECT_EXPERIMENTAL_ENABLE_WORKSPACE_DASHBOARD=True
in your terminal.
Improvements to git_clone
deployment pull step git
Previously, you had to apply the appropriate format for your service credentials in a Secret
block using the access_token
field in git_clone
. The git_clone
pull step now includes an additional credentials
field, allowing you to leverage your existing GitHubCredentials
, GitLabCredentials
, or BitBucketCredentials
blocks when cloning from a private repository. See the docs for examples.
Check out the release notes for more information.justabill
justabill
my-deployment
for a flow with the name `my-flow`:
prefect deploy --name my-flow/my-deployment
This is especially useful when you have several flows with deployments that have the same name.
You can also now use environment variables in deployment steps. You can access environment variables during the pull
action at runtime or during the build
and push
actions when running prefect deploy
. For example, you can now use the following syntax to set an image tag of a Dockerized build by loading an environment variable during the build
action:
yaml
build:
- prefect_docker.deployments.steps.build_docker_image:
requires: prefect-docker>0.1.0
image_name: my-image/orion
tag: '{{ $CUSTOM_TAG }}'
But wait, there's more! but wait theres more
When there are multiple deployments with the same name, the prefect deploy
command now prompts you to choose which one to deploy:
For example, if you have the following `prefect.yaml`:
yaml
deployments:
- name: "default"
entrypoint: "flows/hello.py:hello"
- name: "default"
entrypoint: "flows/hello.py:hello_parallel"
running prefect deploy -n default
will now prompt you to choose which flow to create a deployment for.
The Prefect Dashboard - your heads up display 📊 📈
The response to the experimental Prefect dashboard was so enthusiastic that we've made it generally available as the default landing page in the Prefect UI. The dashboard provides an overview of all Prefect activity, surfaces the urgent information, and provides the context to understand that information. With the dashboard, you can:
• Confirm that all flows run in the past 24 hours behaved as expected
• Identify a flow run that recently failed and jump directly to its page
• See a work pool that is unhealthy and the work that is impacted
See the release notes for more information.Will Raphaelson
07/13/2023, 9:05 PMjustabill
Chris Reuter
07/25/2023, 4:55 PMChris Reuter
08/01/2023, 6:51 PMFOR
and IF
statements to an event-driven workflow with variable, uncertain payloads.
Thanks to everyone here who helped contribute via user interviews and feedback, and special shoutout to club42!
Ready, set, flow!!justabill
Jenny
08/10/2023, 8:05 PMjustabill
.visualize()
method. Calling this method will attempt to locally produce an image of the flow's schematic diagram without running the flow's code.
See the release notes for more information.
Happy Engineering! marvinWill Raphaelson
08/22/2023, 3:47 PMjustabill
justabill
Flow.serve()
, the "it just works" way to deploy flows.
The new Flow.serve()
method allows you to automatically create a flow deployment so that you can schedule or trigger runs via the Prefect UI and CLI. Check it out:
python
from prefect import flow
@flow
def hello(name = "Marvin"):
print(f"Hello {name}!")
if __name__ == "__main__":
# Creates a deployment named 'hello/hourly-greeting'
# which will run the 'hello' flow once an hour
hello.serve(name="hourly-greeting", interval=3600)
The script will stay running so that it can listen for scheduled or triggered runs of this flow; once a run is found, it will be executed within a subprocess.
> python hello.py
╭─────────────────────────────────────────────────────────────────────────────────────╮
│ Your flow 'hello' is being served and polling for scheduled runs! │
│ │
│ To trigger a run for this flow, use the following command: │
│ │
│ $ prefect deployment run 'hello/hourly-greeting' │
│ │
╰─────────────────────────────────────────────────────────────────────────────────────╯
To start serving your flows, check out our newly updated quickstart and tutorial.justabill
justabill
justabill
justabill
justabill
justabill
justabill
Jeff Hale
09/20/2023, 3:41 PMChris Reuter
09/21/2023, 1:00 PMSarah Krasnik
09/21/2023, 4:58 PMJeff Hale
09/21/2023, 8:53 PMprefect deploy -n flow-a/*
🛠️ Various fixes, including Prefect variable resolution in the deployments section of prefect.yaml
📚 Documentation updates and 🤝 new contributions from the community
See the release notes for more information. launchWill Raphaelson
09/21/2023, 9:12 PM