mseth10 edited a comment on issue #18618: URL: https://github.com/apache/incubator-mxnet/issues/18618#issuecomment-649870530
Thanks for creating this issue @leezu . This is what I have observed: This timeout error is flaky and has broken CD "CPU" pipeline many times in the past. Earlier I linked a similar error in operator tests to the commit https://github.com/apache/incubator-mxnet/commit/10b6b4887ef494ea985c9ea75a393a985476d08e which used "NaiveEngine" setting to run operator tests in CD. I reverted that change in https://github.com/apache/incubator-mxnet/commit/a361f33497c8e87a4eab48a666fcb4a586a607b1 , which fixed the timeout issue for operator tests which was much more frequent. But there are still four non-operator tests that are flaky and fail with timeout error: - test_contrib_DataLoaderIter - test_while_loop_for_foreach - test_while_loop_rnn - test_cond Please refer to this pipeline in which all four tests fail: http://jenkins.mxnet-ci.amazon-ml.com/blue/rest/organizations/jenkins/pipelines/restricted-mxnet-cd/pipelines/mxnet-cd-release-job/runs/1216/nodes/262/steps/288/log/?start=0 ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
