[YUNIKORN-2043] SimplePreemptor e2e test doesn’t really wait for pod’s Killing event

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: 1.4.0
• Component/s: test - e2e
• Labels: pull-request-available
• Assignee: Yu-Lin Chen
• Reporter: Yu-Lin Chen
• Created: 14/Oct/23 07:42
• Updated: 20/Nov/23 16:42
• Resolved: 17/Oct/23 09:25

Description

In the e2e simple_preemptor_test, the current timeout for ‘waiting Killing event’ is in nanosecond.

1. 120 ns: test/e2e/simple_preemptor/simple_preemptor_test.go#L169
2. 1200 ns: test/e2e/simple_preemptor/simple_preemptor_test.go#L215

The interval is Millisecond*100(code).
So there is no actual 'waiting' happening here.

This issue will trigger an error in the CI even though the pod has been killed.
(Note: Not all SimplePreemptor fails are caused by this issue, there are still other unkown issues to be investigated.)

CI error: https://github.com/apache/yunikorn-k8shim/actions/runs/6495711809/job/17641528620#step:5:2046

• [FAILED] [15.593 seconds]
SimplePreemptor [It] Verify_basic_simple_preemption. Use case: Only one pod is running and same pod has been selected as victim
/home/runner/work/yunikorn-k8shim/yunikorn-k8shim/test/e2e/simple_preemptor/simple_preemptor_test.go:148

  [FAILED] Unexpected error:
      <context.deadlineExceededError>: 
      context deadline exceeded
      {}
  occurred
  In [It] at: /home/runner/work/yunikorn-k8shim/yunikorn-k8shim/test/e2e/simple_preemptor/simple_preemptor_test.go:170 @ 10/12/23 

Issue Links

• links to: GitHub Pull Request #689