Cloud Functions (1st gen) has been renamed to Cloud Run functions (1st gen). Switch to the Cloud Run functions documentation.

Workflows Tutorial (1st gen)

This tutorial shows you how to use Workflows to link a series of services together. By connecting two public HTTP services (using Cloud Run functions), an external REST API, and a private Cloud Run service, you can create a flexible, serverless application.

Objectives

In this tutorial, you use the Google Cloud CLI to create a single workflow, connecting one service at a time:

Deploy two Cloud Run functions services: the first function generates a random number, and then passes that number to the second function which multiplies it.
Using Workflows, connect the two HTTP functions together. Execute the workflow and return a result that is then passed to an external API.
Using Workflows, connect an external HTTP API that returns the log for a given number. Execute the workflow and return a result that is then passed to a Cloud Run service.
Deploy a Cloud Run service that allows authenticated access only. The service returns the math.floor for a given number.
Using Workflows, connect the Cloud Run service, execute the entire workflow, and return a final result.

The following diagram shows both an overview of the process as well as a visualization of the final workflow:

Costs

In this document, you use the following billable components of Google Cloud:

To generate a cost estimate based on your projected usage, use the pricing calculator. New Google Cloud users might be eligible for a free trial.

Before you begin

Security constraints defined by your organization might prevent you from completing the following steps. For troubleshooting information, see Develop applications in a constrained Google Cloud environment.

Sign in to your Google Cloud account. If you're new to Google Cloud, create an account to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.

Install the Google Cloud CLI.

To initialize the gcloud CLI, run the following command:

gcloud init

Create or select a Google Cloud project.

Create a Google Cloud project:
```
gcloud projects create PROJECT_ID
```
Replace PROJECT_ID with a name for the Google Cloud project you are creating.
Select the Google Cloud project that you created:
```
gcloud config set project PROJECT_ID
```
Replace PROJECT_ID with your Google Cloud project name.

Make sure that billing is enabled for your Google Cloud project.

Enable the Artifact Registry, Cloud Build, Cloud Run functions, Cloud Run, Cloud Storage, and Workflows APIs:

gcloud services enable artifactregistry.googleapis.com cloudbuild.googleapis.com cloudfunctions.googleapis.com run.googleapis.com storage.googleapis.com workflows.googleapis.com

Install the Google Cloud CLI.

To initialize the gcloud CLI, run the following command:

gcloud init

Create or select a Google Cloud project.

Create a Google Cloud project:
```
gcloud projects create PROJECT_ID
```
Replace PROJECT_ID with a name for the Google Cloud project you are creating.
Select the Google Cloud project that you created:
```
gcloud config set project PROJECT_ID
```
Replace PROJECT_ID with your Google Cloud project name.

Make sure that billing is enabled for your Google Cloud project.

Enable the Artifact Registry, Cloud Build, Cloud Run functions, Cloud Run, Cloud Storage, and Workflows APIs:

gcloud services enable artifactregistry.googleapis.com cloudbuild.googleapis.com cloudfunctions.googleapis.com run.googleapis.com storage.googleapis.com workflows.googleapis.com

Update the Google Cloud CLI components:
```
gcloud components update
```

If you are running commands inside Cloud Shell, you are already authenticated with the gcloud CLI; otherwise, sign in using your account:
```
gcloud auth login
```

Create a service account for Workflows to use:

export SERVICE_ACCOUNT=workflows-sa
gcloud iam service-accounts create ${SERVICE_ACCOUNT}

To allow the service account to call authenticated Cloud Run services, grant the run.invoker role to the Workflows service account:

gcloud projects add-iam-policy-binding PROJECT_ID \
    --member "serviceAccount:${SERVICE_ACCOUNT}@PROJECT_ID.iam.gserviceaccount.com" \
    --role "roles/run.invoker"

Replace PROJECT_ID with your Google Cloud project ID.

Set the default location used in this tutorial:

gcloud config set project PROJECT_ID
export REGION=REGION
gcloud config set functions/region ${REGION}
gcloud config set run/region ${REGION}
gcloud config set workflows/location ${REGION}

Replace REGION with the supported Workflows location of your choice.

Deploy the first Cloud Run functions service

After receiving an HTTP request, this HTTP function generates a random number between 1 and 100, and then returns the number in JSON format.

Create a directory called randomgen and change to it:
```
mkdir ~/randomgen
cd ~/randomgen
```

Create a text file with the filename main.py that contains the following Python code:

import functions_framework
import random
from flask import jsonify


@functions_framework.http
def randomgen(request):
    randomNum = random.randint(1, 100)
    output = {"random": randomNum}
    return jsonify(output)

To support a dependency on Flask for HTTP processing, create a text file for the pip package manager. Give it the filename requirements.txt and add the following:
```
flask>=1.0.2
functions-framework==3.0.0
```

Deploy the function with an HTTP trigger, and allow unauthenticated access:

gcloud functions deploy randomgen \
    --runtime python37 \
    --trigger-http \
    --allow-unauthenticated

The function might take a few minutes to deploy. Alternatively, you can use the Cloud Run functions interface in the Google Cloud console to deploy the function.

Once the function is deployed, you can confirm the httpsTrigger.url property:
```
gcloud functions describe randomgen
```
You can try out the function with the following curl command:
```
curl $(gcloud functions describe randomgen --format='value(httpsTrigger.url)')
```
A number is randomly generated and returned.

Deploy the second Cloud Run functions service

After receiving an HTTP request, this HTTP function extracts the input from the JSON body, multiplies it by 2, and returns the result in JSON format.

Create a directory called multiply and change to it:
```
mkdir ~/multiply
cd ~/multiply
```

Create a text file with the filename main.py that contains the following Python code:

import functions_framework
from flask import jsonify


@functions_framework.http
def multiply(request):
    request_json = request.get_json()
    output = {"multiplied": 2 * request_json['input']}
    return jsonify(output)

To support a dependency on Flask for HTTP processing, create a text file for the pip package manager. Give it the filename requirements.txt and add the following:
```
flask>=1.0.2
functions-framework==3.0.0
```

Deploy the function with an HTTP trigger, and allow unauthenticated access:

gcloud functions deploy multiply \
    --runtime python37 \
    --trigger-http \
    --allow-unauthenticated

The function might take a few minutes to deploy.Alternatively, you can use the Cloud Run functions interface in the Google Cloud console to deploy the function.

Once the function is deployed, you can confirm the httpsTrigger.url property:
```
gcloud functions describe multiply
```

You can try out the function with the following curl command:

curl $(gcloud functions describe multiply --format='value(httpsTrigger.url)') \
    -X POST \
    -H "content-type: application/json" \
    -d '{"input": 5}'

The number 10 should be returned.

Connect the two Cloud Run functions services in a workflow

A workflow is made up of a series of steps described using the Workflows syntax, which can be written in either YAML or JSON format. This is the workflow's definition. For a detailed explanation, see the Syntax reference page.

Navigate back to your home directory:
```
cd ~
```

Create a text file with the filename workflow.yaml that contains the following content:

- randomgen_function:
    call: http.get
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/randomgen
    result: randomgen_result
- multiply_function:
    call: http.post
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/multiply
        body:
            input: ${randomgen_result.body.random}
    result: multiply_result
- return_result:
    return: ${multiply_result}

This links the two HTTP functions together and returns a final result.

After creating the workflow, you can deploy it, which makes it ready for execution.
```
gcloud workflows deploy WORKFLOW_NAME --source=workflow.yaml
```
Replace WORKFLOW_NAME with a name for your workflow.
Execute the workflow:
```
gcloud workflows run WORKFLOW_NAME
```
An execution is a single run of the logic contained in a workflow's definition. All workflow executions are independent, and the rapid scaling of Workflows allows for a high number of concurrent executions.

After the workflow is executed, the output should resemble the following:
```
result: '{"body":{"multiplied":120},"code":200,"headers":{"Alt-Svc":"h3-29=\":443\";
...
startTime: '2021-05-05T14:17:39.135251700Z'
state: SUCCEEDED
...
```

Connect a public REST service in the workflow

Update your existing workflow and connect a public REST API (math.js) that can evaluate mathematical expressions. For example, curl https://api.mathjs.org/v4/?'expr=log(56)'.

Note that since you have deployed your workflow, you can also edit it through the Workflows page in the Google Cloud console.

Edit the source file for your workflow and replace it with the following content:

- randomgen_function:
    call: http.get
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/randomgen
    result: randomgen_result
- multiply_function:
    call: http.post
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/multiply
        body:
            input: ${randomgen_result.body.random}
    result: multiply_result
- log_function:
    call: http.get
    args:
        url: https://api.mathjs.org/v4/
        query:
            expr: ${"log(" + string(multiply_result.body.multiplied) + ")"}
    result: log_result
- return_result:
    return: ${log_result}

This links the external REST service to the Cloud Run functions services, and returns a final result.

Deploy the modified workflow:

gcloud workflows deploy WORKFLOW_NAME --source=workflow.yaml

Deploy a Cloud Run service

Deploy a Cloud Run service that, after receiving an HTTP request, extracts input from the JSON body, calculates its math.floor, and returns the result.

Create a directory called floor and change to it:
```
mkdir ~/floor
cd ~/floor
```

Create a text file with the filename app.py that contains the following Python code:

import json
import logging
import os
import math

from flask import Flask, request

app = Flask(__name__)


@app.route('/', methods=['POST'])
def handle_post():
    content = json.loads(request.data)
    input = float(content['input'])
    return f"{math.floor(input)}", 200


if __name__ != '__main__':
    # Redirect Flask logs to Gunicorn logs
    gunicorn_logger = logging.getLogger('gunicorn.error')
    app.logger.handlers = gunicorn_logger.handlers
    app.logger.setLevel(gunicorn_logger.level)
    app.logger.info('Service started...')
else:
    app.run(debug=True, host='0.0.0.0', port=int(os.environ.get('PORT', 8080)))

In the same directory, create a Dockerfile with the following content:

# Use an official lightweight Python image.
# https://hub.docker.com/_/python
FROM python:3.7-slim

# Install production dependencies.
RUN pip install Flask gunicorn

# Copy local code to the container image.
WORKDIR /app
COPY . .

# Run the web service on container startup. Here we use the gunicorn
# webserver, with one worker process and 8 threads.
# For environments with multiple CPU cores, increase the number of workers
# to be equal to the cores available.
CMD exec gunicorn --bind 0.0.0.0:8080 --workers 1 --threads 8 app:app

Create an Artifact Registry standard repository where you can store your Docker container image:
```
gcloud artifacts repositories create REPOSITORY \
    --repository-format=docker \
    --location=${REGION}
```
Replace REPOSITORY with a unique name for the repository.

Build the container image:

export SERVICE_NAME=floor
gcloud builds submit --tag ${REGION}-docker.pkg.dev/PROJECT_ID/REPOSITORY/${SERVICE_NAME}

Deploy the container image to Cloud Run, ensuring that it only accepts authenticated calls:

gcloud run deploy ${SERVICE_NAME} \
    --image ${REGION}-docker.pkg.dev/PROJECT_ID/REPOSITORY/${SERVICE_NAME}:latest \
    --platform managed \
    --no-allow-unauthenticated

When you see the service URL, the deployment is complete. You will need to specify that URL when updating the workflow definition.

Connect the Cloud Run service in the workflow

Update your existing workflow and specify the URL for the Cloud Run service.

Edit the source file for your workflow and replace it with the following content:

- randomgen_function:
    call: http.get
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/randomgen
    result: randomgen_result
- multiply_function:
    call: http.post
    args:
        url: https://REGION-PROJECT_ID.cloudfunctions.net/multiply
        body:
            input: ${randomgen_result.body.random}
    result: multiply_result
- log_function:
    call: http.get
    args:
        url: https://api.mathjs.org/v4/
        query:
            expr: ${"log(" + string(multiply_result.body.multiplied) + ")"}
    result: log_result
- floor_function:
    call: http.post
    args:
        url: CLOUD_RUN_SERVICE_URL
        auth:
            type: OIDC
        body:
            input: ${log_result.body}
    result: floor_result
- create_output_map:
    assign:
      - outputMap:
          randomResult: ${randomgen_result}
          multiplyResult: ${multiply_result}
          logResult: ${log_result}
          floorResult: ${floor_result}
- return_output:
    return: ${outputMap}

Replace CLOUD_RUN_SERVICE_URL with your Cloud Run service URL.

This connects the Cloud Run service in the workflow. Note that the auth key ensures that an authentication token is being passed in the call to the Cloud Run service. For more information, see Make authenticated requests from a workflow.

Deploy the modified workflow:

gcloud workflows deploy WORKFLOW_NAME \
    --source=workflow.yaml

Execute the final workflow:

gcloud workflows run WORKFLOW_NAME

The output should resemble the following:

result: '{"Floor":{"body":"4","code":200
  ...
  "Log":{"body":"4.02535169073515","code":200
  ...
  "Multiply":{"body":{"multiplied":56},"code":200
  ...
  "Random":{"body":{"random":28},"code":200
  ...
startTime: '2023-11-13T21:22:56.782669001Z'
state: SUCCEEDED

Congratulations! You have deployed and executed a workflow that connects a series of services together.

To create more complex workflows using expressions, conditional jumps, Base64 encoding or decoding, subworkflows, and more, refer to the Workflows syntax reference and the Standard library overview.

Clean up

If you created a new project for this tutorial, delete the project. If you used an existing project and want to keep it without the changes added in this tutorial, delete resources created for the tutorial.

Delete the project

The easiest way to eliminate billing is to delete the project that you created for the tutorial.

To delete the project:

In the Google Cloud console, go to the Manage resources page.
Go to Manage resources
In the project list, select the project that you want to delete, and then click Delete.
In the dialog, type the project ID, and then click Shut down to delete the project.

Delete tutorial resources

Delete the Cloud Run service you deployed in this tutorial.
Delete the workflow you created in this tutorial.
Delete the container image from Artifact Registry.
Delete the Artifact Registry repository.

Remove the Google Cloud CLI default configurations you added during the tutorial setup:

gcloud config unset functions/region
gcloud config unset run/region
gcloud config unset workflows/location
gcloud config unset project