Cloud SQL alerting

Summary

This issue proposes introducing a minimal set of core Cloud SQL alerts, aligned with our use of the gcp-deploy-boilerplate. The alerts outlined below are to be implemented in theory. As part of this work, we will also assess the practical feasibility of implementing each alert.

• Instance unavailable
  • Metric: cloudsql.googleapis.com/database/up
  • Conditions:
    • Critical: metric equals 0 for 60 seconds
• CPU utilisation
  • Metric: cloudsql.googleapis.com/database/memory/utilization
  • Conditions:
    • Critical: p90 > 95% for 15 minutes
• Memory utilisation
  • Metric: cloudsql.googleapis.com/database/memory/utilization
  • Conditions:
    • Critical p90 > 95% for 15 minutes
• Disk space used
  • Metric: cloudsql.googleapis.com/database/disk/bytes_used
  • Conditions:
    • Critical > 95%