Skip to content

Conversation

@danail-branekov
Copy link
Member

@danail-branekov danail-branekov commented Jan 30, 2026

Is there a related GitHub Issue?

No

What is this change about?

Allow up to a minute for the k8s manager to start

Our tests start the k8s manager in a separate thread as
k8sManager.Start() blocks while the manager is running. Once we start
the manager, we check whether it is up by invoking its webhook server
started check.

We have been experiencing flakes caused by the webhook server failing to
start within a second, for example
https://ci.korifi.cf-app.com/teams/main/pipelines/main/jobs/run-tests-main/builds/1896

The interesting thing is that the failure is always in the
controllers/services/osbapi suite. What distinguishes this suite from
the rest of the suites is that it does NOT configure a default
eventually timeout, hence the default (one second) is used. One second
seems to be insufficient, therefore hardcode the webhook server check
timeout to a minute.

Tag your pair, your PM, and/or team

@georgethebeatle @uzabanov

Explicitly invoke the manager cancel function in `AfterEach`. While
doing that in `DeferCleanup` is technically correct, the code looks
weird and could raise some eyebrows
Our tests start the k8s manager in a separate thread as
`k8sManager.Start()` blocks while the manager is running. Once we start
the manager, we check whether it is up by invoking its webhook server
started check.

We have been experiencing flakes caused by the webhook server failing to
start within a second, for example
https://ci.korifi.cf-app.com/teams/main/pipelines/main/jobs/run-tests-main/builds/1896

The interesting thing is that the failure is always in the
`controllers/services/osbapi` suite. What distinguishes this suite from
the rest of the suites is that it does NOT configure a default
eventually timeout, hence the default (one second) is used. One second
seems to be insufficient, therefore hardcode the webhook server check
timeout to a minute.
@danail-branekov danail-branekov merged commit a0a47d7 into main Jan 30, 2026
11 checks passed
@danail-branekov danail-branekov deleted the fix-flakes branch January 30, 2026 14:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants