🙋♀️ The University of California Curation Center (UC3) helps researchers and the UC libraries manage, preserve, and provide access to their important digital assets.
This page outlines shared development practices across UC3 repositories. These guidelines establish a consistent baseline for how we build, review, deploy, and operate our services.
For broader context, see our UC3 Product Development Approach.
- All work must be tracked in GitHub issues
- Issues should include:
- Clear description
- Acceptance criteria
- Relevant context and dependencies
- Work is prioritized through the Product Manager
- All changes must be submitted via pull request
- PRs require at least one review before merging
- PRs should include:
- Summary of changes
- Testing instructions
- Related issue(s)
- Reviews should validate:
- Functionality
- Code quality
- Security considerations
- Feedback should be clear, constructive, and actionable
- Automated tests are expected where feasible
- Changes should be validated before merging
- Accessibility and performance should be considered as part of development
- All product teams adhere to UC3 Security practices
- Keep releases small and incremental
- Avoid deployments late in the day, on Fridays, or before holidays
- Monitor systems after deployment
- Communicate releases to relevant stakeholders
Use the standard release template: https://github.com/CDLUC3/.github/blob/main/ISSUE_TEMPLATE/release.md
- Infrastructure should be managed as code wherever possible
- Services should include:
- Logging
- Monitoring
- Alerting
- Use shared AWS patterns and tooling when available
When migrating EC2-based services to ECS:
- Containerize applications and reduce host dependencies
- Design for stateless services where possible
- Externalize configuration using environment variables
- Store secrets in AWS SSM or Secrets Manager
- Use managed services for persistence (S3, RDS, EFS)
- Avoid reliance on local disk
- Build and deploy container images through pipelines
- Use versioned image tags
- Ensure centralized logging (CloudWatch, OpenSearch)
- Monitor service and container health
- Use rolling or blue/green deployments where possible
- Ensure rollback via previous task definitions
- Monitor utilization and right-size services
- Evaluate ECS vs EC2 tradeoffs
- At least two developers should be familiar with each service
- Document architecture and operational processes
- Raise blockers and risks early
- Share knowledge across teams