[ https://issues.apache.org/jira/browse/HDFS-11045?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Daniel Templeton updated HDFS-11045: ------------------------------------ Attachment: HDFS-11045.009.patch This patch addresses all your comments, [~xiaochen], except for the 0.9/1.1 concern. The flakiness in the tests is always that the run time is too high, which means that the ratio will never be more than about 1.01 of expectation. I used 1.1 just to be extra safe. 0.9 is aspirational. If the tests are working correctly, we shouldn't see anything more than 2% off expectation. When the FS lock is held too long (as detailed above), we get running times that are wildly too high. To make the assert tolerant of those cases, I'd have to use 0.7 or less. I've upped the retry count in hopes that in those cases we can retry until we get a clean run. If that works, then 0.9 will be fine. I'm eagerly waiting to see what [~jzhuge] sees in his testing. Based on his results I'll either lower the 0.9 or not. > TestDirectoryScanner#testThrottling fails: Throttle is too permissive > --------------------------------------------------------------------- > > Key: HDFS-11045 > URL: https://issues.apache.org/jira/browse/HDFS-11045 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode > Affects Versions: 3.0.0-alpha2 > Reporter: John Zhuge > Assignee: Daniel Templeton > Priority: Minor > Attachments: HDFS-11045.001.patch, HDFS-11045.002.patch, > HDFS-11045.003.patch, HDFS-11045.004.patch, HDFS-11045.005.patch, > HDFS-11045.006.patch, HDFS-11045.007.patch, HDFS-11045.008.patch, > HDFS-11045.009.patch > > > TestDirectoryScanner.testThrottling:709 Throttle is too permissive > https://builds.apache.org/job/PreCommit-HDFS-Build/17259/artifact/patchprocess/patch-unit-hadoop-hdfs-project_hadoop-hdfs.txt -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org