yangwwei commented on pull request #369: URL: https://github.com/apache/incubator-yunikorn-k8shim/pull/369#issuecomment-1075940644
hi @anuraagnalluri There are a couple of things: 1. The unit test failure was unrelated to your patch, looks like it was because [here](https://github.com/apache/incubator-yunikorn-k8shim/blob/4fa4544c07e8820031c7acc4dc05fd068c0482c5/pkg/cache/placeholder_manager.go#L154) we need to use a read lock when to get the cleanupTime. We need a separate JIRA to get this fixed. For this one, as long as we can have a good UT run, we are good. 2. The failed e2e tests are with `--plugin` option, which means they are testing the scheduler-plugin mode. The panic happens at ` /home/runner/work/incubator-yunikorn-k8shim/incubator-yunikorn-k8shim/test/e2e/recovery_and_restart/recovery_and_restart_test.go:120 +0x156c` which was because the allocations list retrieved from the scheduler was empty (we can also see that from the debug output). Have you looked at the history if it was always failing under the plugin mode? if that's the case, I suggest rerunning this simple scenario in plugin mode and see if it introduces a different behavior after scheduler restart. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
