GCP Cloud Scheduler Job Builder

About This Tool

Google Cloud Scheduler is a fully managed cron job service that triggers targets on a schedule defined by cron expressions or time intervals. Targets can be HTTP/S endpoints, Pub/Sub topics, or App Engine routes, each with different authentication and configuration options. Configuring Cloud Scheduler jobs requires specifying the schedule in unix-cron format with timezone, the target type and endpoint, authentication method (OAuth or OIDC tokens for HTTP targets), retry policies, and attempt deadlines. This builder helps you construct job configurations with correct cron expressions, target settings, and retry parameters.

When to Use This Tool

• •Building a Cloud Scheduler job that triggers a Cloud Function via HTTP every 15 minutes with OIDC authentication
• •Creating a scheduled Pub/Sub publisher that sends messages to a topic on a cron schedule for batch processing pipelines
• •Configuring retry policies with exponential backoff for scheduled HTTP targets that may experience transient failures
• •Generating Cloud Scheduler job definitions for Terraform that trigger Cloud Run services with proper IAM-based authentication

Frequently Asked Questions

What cron expression format does Cloud Scheduler use?

Cloud Scheduler uses the standard unix-cron format with five fields: minute (0-59), hour (0-23), day of month (1-31), month (1-12), and day of week (0-6). It also supports the non-standard sixth field for year. Special characters include * (any), , (list), - (range), and / (step). Schedules are evaluated in the timezone you specify (e.g., America/New_York). For example, '0 9 * * 1-5' triggers at 9:00 AM on weekdays in the configured timezone.

How does Cloud Scheduler authenticate to HTTP targets?

Cloud Scheduler supports two authentication methods for HTTP targets: OAuth tokens and OIDC tokens. OAuth tokens use a service account to generate an OAuth 2.0 access token sent in the Authorization header — used for calling Google APIs or Cloud Functions. OIDC tokens generate an OpenID Connect ID token for the specified audience — used for Cloud Run, Cloud Functions (2nd gen), and custom services that validate OIDC tokens. The service account must have the necessary IAM permissions to invoke the target.

What happens when a scheduled job fails?

When a job fails (non-2xx HTTP response, Pub/Sub publish error, or App Engine error), Cloud Scheduler retries based on the configured retry policy. You can set the minimum and maximum backoff duration, maximum retry duration, maximum number of doublings for exponential backoff, and maximum retry attempts. If all retries are exhausted, the job execution is marked as failed. Cloud Scheduler logs all attempts to Cloud Logging, and you can set up monitoring alerts for failed jobs.