Configuring Arcade Deploy

The arcade deploy command deploys your worker to the cloud.

Requirements

Python 3.10 or higher
Verify your Python version by running python --version or python3 --version in your terminal.
Arcade Account: Sign up for an Arcade account if you haven’t already.
Arcade CLI: Install the Arcade CLI with pip install arcade-ai.

Deployment Config

The worker.toml file is used to configure your worker. Running arcade deploy will use the worker.toml file in the current directory or you can specify a different file with the -d flag.

Running the CLI command arcade new will automatically create an example worker.toml file for the created toolkit.


# worker.toml
 
# Arcade Deploy supports configuring multiple workers in a single file.
# Configurations below each [[worker]] block will be deployed as a separate worker.
[[worker]]
 
# (Required) The worker configuration
[worker.config]
 
# (Required) The unique identifier for the worker.
id = "my-example-worker"
# (Optional) Whether the worker is enabled.
enabled = true
# (Optional) The timeout for the worker for requests to the worker.
timeout = 30
# (Optional) The number of times to retry a connection to the worker before giving up.
retries = 3
# (Required) The secret for the worker.
# The default "dev" secret or empty string is not allowed.
# Environment variables can also be used by setting the secret with the format "${env:YOUR_VARIABLE_NAME}"
secret = <your secret>
 
# (Optional) Localy packages to deploy with the worker.
# You can specify a list of directories that contain a toolkit.
# The directory must contain a pyproject.toml file or a setup.py file.
[worker.local_source]
packages = [ "./my_toolkit1", "./my_toolkit2"]
 
# (Optional) Pypi packages to install for the worker.
[worker.pypi_source]
packages = [ "arcade-math", "my-pypi-package"]

Secrets

You can use a secret from the environment to configure the worker with the ${env: MY_SECRET} syntax.


[[worker]]
 
[worker.config]
secret = "${env: MY_ENVIRONMENT_SECRET}"

Using with a local engine

To register the worker with your local engine, you can use the host and port flags.


arcade deploy -h localhost -p 9099 --no-tls

Logs

To see the logs for the worker, you can use the arcade logs command.


arcade logs my-example-worker