[
https://issues.apache.org/jira/browse/FLINK-18262?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17161934#comment-17161934
]
Robert Metzger commented on FLINK-18262:
----------------------------------------
I agree with Chesnay that it is unlikely that the "tar" command itself has an
issue. Maybe we are trying to compress more than we are expecting (maybe there
are (unexpected) large files in the directory, or we are following some
symlinks somewhere?)
Does it make sense to run the tar command with verbose output? (if that's
possible), so that we can see what is it compressing, and where it starts
hanging?
We could also run the entire bash script with "set -x" to make sure it is
really the "tar" call, and not the "mkdir" or "docker cp" call.
> PyFlink end-to-end test stalls
> ------------------------------
>
> Key: FLINK-18262
> URL: https://issues.apache.org/jira/browse/FLINK-18262
> Project: Flink
> Issue Type: Bug
> Components: API / Python, Tests
> Affects Versions: 1.12.0
> Reporter: Robert Metzger
> Priority: Critical
> Labels: pull-request-available, test-stability
>
> https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=3299&view=logs&j=c88eea3b-64a0-564d-0031-9fdcd7b8abee&t=1e2bbe5b-4657-50be-1f07-d84bfce5b1f5
> {code}
> 2020-06-11T17:25:37.2919869Z We now have 2 NodeManagers up.
> 2020-06-11T18:22:11.3398295Z ##[error]The operation was canceled.
> 2020-06-11T18:22:11.3411819Z ##[section]Finishing: Run e2e tests
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
