[ https://issues.apache.org/jira/browse/HDFS-13731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16590928#comment-16590928 ]
Xiao Chen commented on HDFS-13731: ---------------------------------- Thanks a lot [~zvenczel] for the work here. Good find! For the {{ConcurrentModificationException}}, I agree it should be better protected. I think we should protect this in {{ReencryptionUpdater#processCheckpoints}} with a {{synchronized(handler)}} too. Currently handler has FSDirectory readlock, and updater has FSDirectory writelock, which doesn't protect the removal of the list. Could you share a little more about what happened in the 2 other failures? Great if we can have a failure log. {{saveNamespace}} is synchronized call, why would {{waitFor}} help us? > ReencryptionUpdater fails with ConcurrentModificationException during > processCheckpoints > ---------------------------------------------------------------------------------------- > > Key: HDFS-13731 > URL: https://issues.apache.org/jira/browse/HDFS-13731 > Project: Hadoop HDFS > Issue Type: Bug > Components: encryption, test > Affects Versions: 3.0.0 > Reporter: Xiao Chen > Assignee: Zsolt Venczel > Priority: Major > Attachments: HDFS-13731-failure.log, HDFS-13731.01.patch > > > HDFS-12837 fixed some flakiness of Reencryption related tests. But as > [~zvenczel]'s comment, there are a few timeouts still. We should investigate > that. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org