This sample shows how to send structured logs to Stackdriver Logging.

Read more about Cloud Run logging in the Logging How-to Guide.

For more details on how to work with this sample read the Google Cloud Run Node.js Samples README.

export SERVICE_NAME=logging-manual
export CONTAINER_IMAGE=gcr.io/${GOOGLE_CLOUD_PROJECT}/logging-manual
npm run e2e-test

The url.sh script derives the automatically provisioned URL of a deployed Cloud Run service.

export SERVICE_NAME=logging-manual
export REGION=us-central1
test/url.sh

The deploy.sh script deploys a Cloud Run service.

export SERVICE_NAME=logging-manual
export CONTAINER_IMAGE=gcr.io/${GOOGLE_CLOUD_PROJECT}/logging-manual
export REGION=us-central1
test/deploy.sh

The runner.sh script:

• Deploys the service to Cloud Run based on the deploy.sh script.
• Sets the BASE_URL and ID_TOKEN environment variables.
• Runs any arguments passed to the runner.sh script.
• Tears down the Cloud Run service on completion.

test/runner.sh sleep 20

• BASE_URL: The Cloud Run service URL for end-to-end tests.
• ID_TOKEN: JWT token used to authenticate with Cloud Run's IAM-based authentication.
• REGION: [us-central1] Optional override region for the location of the Cloud Run service.
• SERVICE_NAME: The name of the deployed service, used in some API calls and test assertions.
• GOOGLE_CLOUD_PROJECT: If used, overrides GCP metadata server to determine the current GCP project. Required by the logging client library.

• express: Web server framework.
• got: Used to pull the project ID of the running service from the compute metadata server and make system test HTTP requests. This is required in production for log correlation without manually setting the $GOOGLE_CLOUD_PROJECT environment variable.