You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
vdemeester opened this issue
Nov 3, 2023
· 0 comments
Labels
area/releaseIndicates an issue on release (process, tasks).area/testingIssues or PRs related to testingkind/featureCategorizes issue or PR as related to a new feature.
The operator should be tested on a variety of Kubernetes version, from the lowest we support to the latest available.
We should nightly as well as on pull-requests (running them in parallel). Today, we only have one check which is pull-tekton-operator-integration-tests. Instead we should have several of them, one per k8s version we "support". We do not necessarily need to run the full test-suites on all version (if they are taking ages), but a selected number that should cover the basis (and most components if not all).
We can go the chains route and start by using GitHub workflows (and kind)
We can then (or initially) go the tekton+matrix routes and hook the dogfooding tekton cluster to this repository.
I would suggest we should go ahead with the GitHub workflow to work on the "tooling" to do this. Porting it to tekton then should be relatively straightforward.
area/releaseIndicates an issue on release (process, tasks).area/testingIssues or PRs related to testingkind/featureCategorizes issue or PR as related to a new feature.
The operator should be tested on a variety of Kubernetes version, from the lowest we support to the latest available.
We should nightly as well as on pull-requests (running them in parallel). Today, we only have one check which is
pull-tekton-operator-integration-tests
. Instead we should have several of them, one per k8s version we "support". We do not necessarily need to run the full test-suites on all version (if they are taking ages), but a selected number that should cover the basis (and most components if not all).chains
route and start by using GitHub workflows (andkind
)I would suggest we should go ahead with the GitHub workflow to work on the "tooling" to do this. Porting it to tekton then should be relatively straightforward.
cc @tektoncd/operator-maintainers
Note: e.g.
tektoncd/chains
is doing that on pull-request (like here)The text was updated successfully, but these errors were encountered: