-
Notifications
You must be signed in to change notification settings - Fork 594
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
Consolidate E2E environment variables with integration varaibles #4727
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #4727 +/- ##
=======================================
+ Coverage 77.6% 77.8% +0.1%
=======================================
Files 164 165 +1
Lines 18453 18477 +24
=======================================
+ Hits 14331 14386 +55
+ Misses 3299 3273 -26
+ Partials 823 818 -5 ☔ View full report in Codecov by Sentry. |
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.
One nit and conflicts to resolve.
E2E (targeted) tests with KIND-based clusters were started at https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6566314300 |
More conflicts. #4766 removed some generic E2E utilities that weren't specific to <3.4 gates. While we chose a required version newer than any of the extant gates now, we probably will have similar gates in the future, so keeping the generic functions around with |
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.
The nolint:unused
comment did not take its effect. Is there any difference between the local make lint
and lint job on the CI ?
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.
In order to prevent the linter fail you'd need to add
, "-U1000"
to
checks = ["all", "-ST1000", "-ST1005"] |
We run both staticcheck
and golangci-lint
as part of lint
and they don't take into account the same code comment markers.
istio tests failed in e2e suite: https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6566314300/job/17836886646 |
6ea5cc0
to
97c8025
Compare
E2E (targeted) tests with KIND-based clusters were started at https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6592767672 |
https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6592767672/job/17914168070 Istio run now happy. Do we have a way to run GKE tests off a branch? It hopefully just needs the same thing with a static controller image (I'd neglected to add the new input splitter to both it and Istio), but https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6592767672 just skips it. It shouldn't have additional complications given that it's just the main E2E job with GKE flipped on, but if we can run it, may as well. Edit: https://github.com/Kong/kubernetes-ingress-controller/actions/runs/6592883775 the nightly job has a branch dispatch option, so that should get it and anything else that differs from the label variant. |
* chore(e2e) use testenv.ClusterVersion() * chore(e2e) use testenv image/tag overrides * chore(e2e) use testenv pull credentials * chore(e2e) use testenv cluster info * chore(e2e) use testenv image load flag * chore(e2e) move Github env into testenv * chore(e2e) remove e2e-specific envvars * chore(e2e) update workflow envvars
What this PR does / why we need it:
Source E2E test environment setup variables from the same utility package used for integration. Align E2E variables with their integration counterparts.
E2E now uses separate
TEST_(KONG|CONTROLLER)_(IMAGE|TAG)
variables where it previously used a singleTEST_KONG(_CONTROLLER)_IMAGE_OVERRIDE
variable with both the repo and tag.CI inputs still use a single value that the CI scripts split.
Which issue this PR fixes:
Fix #4593
Special notes for your reviewer:
GKE-related variables are defined by the related KTF module, and have not been changed.
License data was also using a KTF-defined variable from the Kong addon, although it isn't actually using the Kong addon. Integration does use the same var AFAIK, however, since it does use the Kong addon. I left this unchanged.
https://github.com/rainest/kubernetes-ingress-controller/actions/runs/6305128682/job/17117913357 and https://github.com/rainest/kubernetes-ingress-controller/actions/runs/6305130237/job/17119050768 have debug-enabled CI runs on a fork. GKE and Enterprise license failures are expected for the fork.
This is mostly to confirm the split script is populating the expected variables, since those variable names actually changed in E2E:
Other variables just got redefined at a new location.
PR Readiness Checklist:
Complete these before marking the PR as
ready to review
:theCHANGELOG.md
release notes have been updated to reflect any significant (and particularly user-facing) changes introduced by this PR