-
Notifications
You must be signed in to change notification settings - Fork 75
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add e2e tests #32
Add e2e tests #32
Conversation
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: BobyMCbobs The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think what I'm actually looking for in #24 is mentioned by this part:
We should then make sure we have k8s e2e jobs that use registry-sandbox, and have them in a testgrid dashboard for this project.
We should have actual kubernetes e2e tests (like kubetest + gce) running in a testgrid dashboard for this project with alerting (conformance tests maybe? or just the parallel subset?) but using registry-sandbox.k8s.io as the kubernetes image registry with the latest code from this repo deployed to registry-sandbox.k8s.io (#24 (comment) => kubernetes/k8s.io#3411), so we know even the cloud run deployment etc is configured correctly, and that it works with a real client we can expect.
I think this will drive far higher confidence that we're ready to promote a change from the sandbox to production.
In #23 I think we can write something reasonable with a shell script + crane to spot check running a local registry binary and pull a known small image (pause?) through it, and add this as a make integration rule, which will give us a quicker way to test changes while developing / in presubmit, though with lower confidence. We can track crane in the hack/tools module and build and run an archeio binary locally with the make targets.
@@ -2,6 +2,25 @@ module sigs.k8s.io/oci-proxy | |||
|
|||
go 1.17 | |||
|
|||
require k8s.io/klog/v2 v2.40.1 | |||
require ( | |||
github.com/google/go-containerregistry v0.8.0 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
let's move the test tool to it's own module, so we can clearly keep imports minimal in the main binary and not require downloading + installing test-only deps during building.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
(or, perhaps rework entirely 😅 #32 (review))
we sort of have those e2e jobs today, but instead they're not in a dashboard we're tracking and they're using the production domain ahead of us rolling that out more widely, we're missing "auto deploy the latest code to registry-sandbox" and "run some realistic CI jobs against registry-sandbox". I think for running against the deployed registry, we should skip straight to an actual cluster pulling from it, and only use containerregistry/crane as the client for local integration tests where it's not as reasonable to use a cluster / we're not covering the deployment configs anyhow. |
Adds some e2e tests to target registry.k8s.io:
pause:3.6
Related #24