Deploy Temporal Worker

This is part of the Charmed Temporal Tutorial. Please refer to this page for more information and the overview of the content.

The Temporal Worker is the entity which listens and polls on specific task queue, and executes code in response to the task.

Deploy

In a production environment, a Temporal worker is deployed in a separate environment from the Temporal server. As our server is currently locally deployed, we will opt for a direct connection from the worker rather than using the ingress.

To deploy Charmed Temporal Worker to a different model, you need to run the following commands, which will fetch the charm from Charmhub and deploy it to your model:

juju add-model worker-model
juju deploy temporal-worker-k8s

Wait until the application is ready - when it is ready, juju status will show:

Model         Controller           Cloud/Region        Version  SLA          Timestamp
worker-model  temporal-controller  microk8s/localhost  3.1.5    unsupported  13:21:49+03:00

App                  Version  Status   Scale  Charm                Channel  Rev  Address         Exposed  Message
temporal-worker-k8s           waiting      1  temporal-worker-k8s  stable     5  10.152.183.187  no       installing agent

Unit                    Workload  Agent  Address      Ports  Message
temporal-worker-k8s/0*  blocked   idle   10.1.232.75         Invalid config: wheel-file-name missing

Configure Worker

In order to test our Temporal worker, we will use a sample resource which contains one workflow and one activity. You can run the following commands to clone the repository, build the necessary wheel file and attach it to the worker charm:

git clone https://github.com/canonical/temporal-worker-k8s-operator.git
cd temporal-worker-k8s-operator/resource_sample

poetry build -f wheel # Take note of the generated file's name

We will then configure our worker to the Temporal server deployed in the previous steps. Create a file config.yaml with the following content:

temporal-worker-k8s:
  host: "10.1.232.64:7233" # Temporal Server unit IP address
  queue: "test-queue"
  namespace: "default"
  workflows-file-name: "python_samples-1.1.0-py3-none-any.whl"
  # To support all defined workflows and activities, use the 'all' keyword
  supported-workflows: "all"
  supported-activities: "all"

Note: To get the Temporal Server unit IP address, you need to switch to the previously created model using juju switch temporal-model before switching back to this model using juju switch temporal-worker.

Run the following command to configure your charm:

juju config temporal-worker-k8s --file=path/to/config.yaml

# Verify that the charm has been configured with the correct value
juju config temporal-worker-k8s host

And then run the following command to attach the workflows file to the worker:

juju attach-resource temporal-worker-k8s workflows-file=./dist/python_samples-1.1.0-py3-none-any.whl

Wait until the application is ready - when it is ready, juju status will show:

Model         Controller           Cloud/Region        Version  SLA          Timestamp
worker-model  temporal-controller  microk8s/localhost  3.1.5    unsupported  13:45:16+03:00

App                  Version  Status  Scale  Charm                Channel  Rev  Address         Exposed  Message
temporal-worker-k8s           active      1  temporal-worker-k8s  stable     5  10.152.183.187  no       worker listening to namespace 'default' on queue 'test-queue'

Unit                    Workload  Agent  Address      Ports  Message
temporal-worker-k8s/0*  active    idle   10.1.232.78         worker listening to namespace 'default' on queue 'test-queue'

To further verify that the worker is functioning correctly, observe the output of the following command to ensure the absence of errors:

kubectl -n worker-model logs temporal-worker-k8s-0 -c temporal-worker -f

At this point, we have a Temporal worker connected to our Temporal server on the default namespace listening for tasks on the test-queue task queue.

See next: Run Your First Workflow