> On Feb. 5, 2019, 11:53 p.m., Jason Dere wrote: > > ql/src/java/org/apache/hadoop/hive/ql/exec/Utilities.java > > Line 1829 (original), 1838 (patched) > > <https://reviews.apache.org/r/69903/diff/1/?file=2123940#file2123940line1838> > > > > No "if" - this dedup strategy does not work with speculative execution > > enabled.
Based on my understanding these are the two scenarios, 1. speculative execution succeeds, it has attempt ID 1. The original attempt ID is 0. The logic picks speculative one, regardless of original one's outcome. This works fine. 2. speculative execution fails, throws exception. Let me know I am getting it wrong. - Deepak ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/69903/#review212581 ----------------------------------------------------------- On Feb. 5, 2019, 10:10 p.m., Deepak Jaiswal wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/69903/ > ----------------------------------------------------------- > > (Updated Feb. 5, 2019, 10:10 p.m.) > > > Review request for hive and Jason Dere. > > > Bugs: HIVE-21214 > https://issues.apache.org/jira/browse/HIVE-21214 > > > Repository: hive-git > > > Description > ------- > > MoveTask : Use attemptId instead of file size for deduplication of files > compareTempOrDuplicateFiles() > > > Diffs > ----- > > ql/src/java/org/apache/hadoop/hive/ql/exec/Utilities.java 8937b43811 > > > Diff: https://reviews.apache.org/r/69903/diff/1/ > > > Testing > ------- > > > Thanks, > > Deepak Jaiswal > >
