[ 
https://issues.apache.org/jira/browse/YUNIKORN-321?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17168641#comment-17168641
 ] 

Adam Antal commented on YUNIKORN-321:
-------------------------------------

There wasn't much documentation around the e2e tests, but I could find that 
they require a Kubernetes cluster to run the tests against it. It uses the kube 
config file which is pointed to my local docker-desktop cluster. I think the 
documentation definitely needs to be enhanced.

Taking into account the above I am not sure if it's the exact error seen in 
YUNIKORN-177, but we do have failing tests on master, so we should keep 
investigating.

> TestPlacementRules e2e test fails on master
> -------------------------------------------
>
>                 Key: YUNIKORN-321
>                 URL: https://issues.apache.org/jira/browse/YUNIKORN-321
>             Project: Apache YuniKorn
>          Issue Type: Bug
>          Components: shim - kubernetes
>    Affects Versions: 0.9
>            Reporter: Adam Antal
>            Priority: Blocker
>
> Logs from GoLang:
> {noformat}
> === RUN   TestPlacementRules
> Running Suite: CI Test Suite
> ============================
> Random Seed: 1595944701
> Will run 3 of 3 specs
> ............STEP: create development namespace
> STEP: Deploy the sleep pod to the development namespace
> Failure [11.064 seconds]
> [BeforeSuite] BeforeSuite 
> /Users/adamantal/git/yunikorn/yunikorn-k8shim/test/e2e/ci/ci_test.go:43
>   Unexpected error:
>       <*errors.errorString | 0xc0002bce20>: {
>           s: "timed out waiting for the condition",
>       }
>       timed out waiting for the condition
>   occurred
>   /Users/adamantal/git/yunikorn/yunikorn-k8shim/test/e2e/ci/ci_test.go:67
> ------------------------------
> Ran 3 of 0 Specs in 11.076 seconds
> FAIL! -- 0 Passed | 3 Failed | 0 Pending | 0 Skipped
> --- FAIL: TestPlacementRules (11.08s)
> {noformat}
> The failure happens at the following line:
> {code:go}
> err = kClient.WaitForPodBySelectorRunning(dev,
>   fmt.Sprintf("app=%s", sleepRespPod.ObjectMeta.Labels["app"]),
>   10)
> {code}
> so we time out waiting for a pod to run. We should fix this issue, as it 
> blocks other jiras like YUNIKORN-177.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@yunikorn.apache.org
For additional commands, e-mail: issues-h...@yunikorn.apache.org

Reply via email to