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
Test running without k8s api proxying. #4054
Conversation
Currently failing CI with:
but I can't for the life of me see why that var is unbound on line 398. Neither can shellcheck :/ Let me know if you've any ideas. |
Nice, tests now running, but looks like a valid failure when creating a private app repo. I'll reproduce that one locally and fix tomorrow. |
@absoludity #4051 is merged into master. Feel free to go ahead! |
Excellent, thanks Rafa. I'll rebase and take it out of draft once I find the culprit of the failure :) |
.circleci/config.yml
Outdated
@@ -686,6 +686,7 @@ jobs: | |||
DEFAULT_DEX_IP: "172.18.0.2" | |||
TEST_UPGRADE: "1" | |||
USE_MULTICLUSTER_OIDC_ENV: "true" | |||
TEST_OPERATORS: "" |
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.
Thanks for introducing the flag for CI.
Shouldn't we add another step where we test the operators?
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.
Oh - I'll be setting the value here to "1" before landing so that we continue to test with operators in CI. After landing I'll then look at whether to also run a job without operator support.
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.
OK, so I've verified that this PR passes CI when the operators and the k8s api proxying are disabled, and have now set the TEST_OPERATORS
to true so that it's now testing operators (and proxying to the k8s api) again to land, until we're ready to switch CI.
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson <minelson@vmware.com>
1a01af9
to
1459147
Compare
Signed-off-by: Michael Nelson <minelson@vmware.com>
Signed-off-by: Michael Nelson minelson@vmware.com
Description of the change
Follows on from both the work on #3896 as well as @castelblanque 's work in the recent #4051 . Since the latter isn't yet landed, I've targeted a branch that I created which is just an updated master with Rafa's changes. I'll retarget to master once Rafa's PR lands. (Edit: done)
This PR updates the frontend config so that it no longer proxies to the k8s API server, unless operators are enabled.
It also updates ci so that the operator switch (and hence whether Kubeapps proxies to the k8s api server) can be configured via the env. I set it to disable operators and the k8s api proxying to verify that it passed, and am now updating it to re-enable the operators in CI for now.
Finally, it also updates the dashboard so that we no longer request api groups and api endpoints at all, when operators are not enabled.
Benefits
No more proxying of k8s api server! (unless operator support is enabled).
Possible drawbacks
Some yet-to-be discovered place where we still depend on the k8s API being accessible, that is not used in e2e, causing a failure (but should be easy to fix, if so)
Applicable issues
Additional information
I've tested this IRL and haven't hit any issues. Let's see what CI says.