[
https://issues.apache.org/jira/browse/HADOOP-15627?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16553587#comment-16553587
]
Steve Loughran commented on HADOOP-15627:
-----------------------------------------
Cause here is that if the test FS has s3guard enabled always, then
the{{Assume.assumeTrue(getFileSystem().hasMetadataStore());}} check at the
start of the test holds, but the FS creation code is only consistent if you set
the -Ds3guard on the test run. otherwise, metadata setup in maybeEnabledS3Guard
is skipped, and your test filesystems are all copies of the FS as
created/configured by default. Which does have s3guard enabled, and with
whatever writeback/auth options that comes with.
Fix: change how the new FS instances are configured and created
> Failure in ITestS3GuardWriteBack.testListStatusWriteBack
> --------------------------------------------------------
>
> Key: HADOOP-15627
> URL: https://issues.apache.org/jira/browse/HADOOP-15627
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3, test
> Affects Versions: 3.2.0
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Priority: Major
>
> Repeatable failure in {{ITestS3GuardWriteBack.testListStatusWriteBack}}
> Possible causes could include
> * test not setting up the three fs instances
> * (disabled) caching not isolating properly
> * something more serious
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
