Define and conduct a V&V Plan with Github Actions CI/CD pipeline for eoAPI (based on RD BB) [#2301]

[j08lue]

• Objective: Involves setting up automated pipelines that ensure every code change made in eoAPI is automatically tested and reliably deployed into the production environment. The focus will be on enhancing code quality, reducing manual errors, speeding up the release process, and ensuring the high availability and stability of eoAPI. Implementing CI/CD practices aims to streamline development workflows, facilitate quicker iterations, and enable a more agile response to the needs of users and stakeholders.
• Lead contributor: DevSeed
• Priority: 1
• Upper Bound Estimate: 7
• Lower Bound Estimate: 5
• Requirements:
  • None

Implementation

Ideally the V&V plan / tests should be maintained in a single place but feature in the EOEPCA building block docs in https://github.com/EOEPCA/data-access.

However, docs integration is secondary. First step is to create the content: what tests (will) exist in eoapi-k8s and how do they run? Could be in eoapi-k8s README or so first.

Where to document is not settled yet:

We do not have a template for the V&V plan. I agree it is a good idea to include it within the main BB doc - but perhaps we need an additional Testing heading under Administration for this.

Our tests will also need to run in the system end-to-end test suite:

Note also, the expectation to contribute to the automated system end-to-end test suite - which has been started at https://github.com/EOEPCA/system-test - currently populated with some example tests for demonstration. Associated to this there is a github action linked to the deployment - https://github.com/EOEPCA/eoepca-plus/blob/deploy-develop/.github/workflows/run-smoke-tests.yaml - which executes the system test (smoke tests only).

Quarterly breakdown

• [ ] https://github.com/EOEPCA/data-access/issues/60

[ranchodeluxe]

Quick status update:

Accomplishments:

1. We've successfully deployed everything into the Argo data-access app. Live here
2. Our existing smoke tests are succeeding (other pre-existing tests are failing)

Next Steps:

1. We still have to choose a subset of tests to use for the titiler-pgstac smoke tests (Issue)
2. We really want to use the Crunchydata Postgresql Operator but there's a couple issues preventing this that need to be investigated (Issue)
3. break out the eoapi-support chart that gives us monitoring and the requestRate auto-scaling we want (Issue)