Brightlake Insurance's operations team changes load balancer settings, firewall rules, and instance templates by hand in the Google Cloud console. Two recent outages were caused by undocumented manual changes that differed between staging and production. The CTO asks for the change that best aligns with the operational excellence pillar of the Google Cloud Well-Architected Framework.
What should you recommend first?
- ARequire a second engineer to watch each manual console changePeer observation reduces some mistakes but keeps changes manual, unrepeatable, and hard to audit.
- BGrant the operations team fewer IAM permissions in productionRestricting access may reduce change volume but does not make changes consistent or reviewable.
- CDefine the infrastructure in Terraform stored in version control, review changes through pull requests, and apply them through a CI/CD pipeline to every environmentInfrastructure as code with reviewed, automated deployment makes changes repeatable, auditable, and identical across environments, which is a core operational excellence recommendation.
- DTake weekly screenshots of production settings to compare with stagingScreenshots document drift after it happens but do not prevent it or make changes reproducible.
Why: The operational excellence pillar emphasizes operational readiness, incident management, resource management, automation of changes, and continuous improvement. Managing infrastructure as code and deploying through pipelines removes the configuration drift that comes from manual changes and gives every change a review and an audit trail. The deciding detail is that the outages came from undocumented manual differences between environments. The distractors add process around manual work instead of replacing it.