generated from AlabamaWaterInstitute/awi-open-source-project-template
-
Notifications
You must be signed in to change notification settings - Fork 4
Open
Labels
CI/CDRelated to Continuous Integration and Delivery/DeploymentRelated to Continuous Integration and Delivery/DeploymentP0Priority level. P0: Critical, P1: High, P2: Medium, P3: LowPriority level. P0: Critical, P1: High, P2: Medium, P3: Low
Description
To keep the forcingprocessor workflows clear and modular, @harshavemula-ua and I have discussed the following structure for the github workflows.
- Unit tests aimed at ensuring the fundamental tooling of forcingprocessor is functional. These are pytests and located in workflows named
forcingprocessor_*.yaml. These tests exist, no development required. - Docker build tests aimed at ensuring the forcingprocessor tooling is functional in the docker container environment builds on both x86 and arm. These are docker builds and then pytests in workflows named
docker_build_x86.yaml,docker_build_arm.yaml,docker_build_arm_push,yaml, etc. These workflows do not yet exist. - Integration tests with tools like DataStreamCLI that both build the docker containers (on each platform) and then issue DataStreamCLI commands to ensure coupling. For DataStreamCLI integration, these workflows were mentioned in docker test and build for x86 and arm #21 and are under development in CI/CD workflows for building and testing ForcingProcessor Docker containers on x86 and ARM architectures #25 . We can name these workflows
integration_datastreamcli_x86.yamlfor example.
This way, we can easily understand at which layer a bug exists and have confidence forcingprocessor works as advertised even if coupled tooling does not.
Metadata
Metadata
Assignees
Labels
CI/CDRelated to Continuous Integration and Delivery/DeploymentRelated to Continuous Integration and Delivery/DeploymentP0Priority level. P0: Critical, P1: High, P2: Medium, P3: LowPriority level. P0: Critical, P1: High, P2: Medium, P3: Low
Type
Projects
Status
In Progress
Status
Todo