13 Monitoring and Cost Controls
Add uptime checks, alerting, cost monitoring for Cloud Run.
Note: This step content was written for the original dual-run architecture and is being updated for the current setup workflow. The guide overview reflects the current approach.
This step is being finalized. Content is accurate but will be updated and expanded before full release.
Deliverable
Server tagging endpoint monitored for availability, errors, and cost.
Validation
Use this final step to make the deployment operationally safe. Confirm there is a real uptime check for the tagging endpoint, a real alerting policy for failures, and a monitoring view that the team can use after go-live. This step is about making sure the system can be trusted and supported, not just configured.
Check the exact resources that will notify humans when the endpoint fails or degrades. Confirm the notification channels, alert thresholds, and dashboard coverage are all reasonable for the expected traffic and support model. If cost controls or traffic expectations matter for the environment, document them here as part of the operational baseline.
Do not leave this step as a vague promise to set up monitoring later. The deliverable is a tagging endpoint that is both technically working and operationally monitored.
Step Values
| Field | Current value |
|---|---|
| Uptime check name | Pending |
| Alert policy names | Pending |
| Notification channels | Pending |
| Baseline monthly cost | Pending |
13.1 Create Uptime Check for Tagging Service
In Google Cloud Monitoring > Uptime Checks, create a check targeting your tagging service /healthy endpoint.
Expected: A GCP uptime check pings the tagging service /healthy endpoint.
Checks
- API GCP uptime check is configured for the tagging service endpoint.
13.2 Create Alert Policy for Failures
In Cloud Monitoring > Alerting, create a policy that triggers on uptime check failures or sustained Cloud Run error spikes. Add a notification channel (email, Slack, PagerDuty).
Expected: A GCP alert policy triggers on uptime check failures with a notification channel configured.
Checks
- API GCP alert policy is configured for uptime check failures.
13.3 Set Up Metrics Dashboard
Create a Cloud Run metrics dashboard showing request count, latency, error rate, and billable instance time for the tagging service.
Expected: A Cloud Run metrics dashboard shows request count, latency, error rate, and billable instance time.
Checks
- Manual Metrics dashboard is set up for Cloud Run monitoring.
Documentation
Fields recorded during this step.
| Field | Description |
|---|---|
| Uptime check name | Name of the GCP uptime check resource. |
| Alert policy names | Names of configured alert policies. |
| Notification channels | Notification channels attached to alert policies. |
| Baseline monthly cost | Estimated monthly cost for Cloud Run services. |
Use the app to validate this step automatically.
Request access